PodSearch

The Talk Show

366: ‘Measure Seven Times, Cut Once’, With Glenn Fleishman

 

00:00:00   Glenn, happy holidays. Happy holidays to you.

00:00:02   Was Philadelphia an ice storm extravaganza like Seattle? No, you know what?

00:00:07   We, it was cold ice cold. I mean, really. I mean, and again,

00:00:11   I don't even want to complain cause I know the temperature. Yeah. Uh,

00:00:15   basically I don't want to complain. It got really cold,

00:00:18   but we avoided all of the precipitation or, you know,

00:00:22   other than like a little bit of rain,

00:00:23   but I didn't even have to salt our sidewalks.

00:00:25   So given how much terrible precipitation, snow, ice,

00:00:29   whatever has hit everybody over the last week, I don't want to complain.

00:00:34   Yeah. We got, we got coded on the Friday before Christmas,

00:00:37   Seattle turned into a,

00:00:39   just a solid sheet of unbroken ice across the entire city.

00:00:42   Like we'd been put under some kind of acrylic layer.

00:00:46   And if you didn't have to go out, I was talking to,

00:00:49   I had a medical appointment yesterday and I was talking to all the medical

00:00:51   staff and I was like, do you have to go out? And they're like, I was,

00:00:54   I was supposed to come in and I was getting the car ready.

00:00:56   And then they're like, don't come in. And the doctor I had is like, well,

00:00:58   I wasn't going to go in, but a patient needed me and he's from the Northeast.

00:01:02   So we drove very carefully and we met at the hospital. Oh, I'm sorry.

00:01:05   I saw a thing and again,

00:01:07   Philadelphia does not get by far and away the worst winter weather,

00:01:12   but it's, you know, we, we definitely, you know, get icy roads.

00:01:15   And I saw that with the weather nationwide that like a surging

00:01:20   Google trend was how to drive on icy roads.

00:01:24   Don't. Yes, exactly.

00:01:27   I lived in Maine for two winters, two winters in one summer,

00:01:31   which is the wrong way to do it.

00:01:33   And the trick is in climates that get a lot of bad weather, typically, you know,

00:01:37   you know, you get the snow and it gets packed down and you drive on snow and you

00:01:40   have snow tires, you have chains in your car if you need them.

00:01:42   And it's the ice that gets you and people are like, well,

00:01:45   if I can drive on snow, I can drive on ice. No, you can't drive on ice.

00:01:48   That's don't know. No, it's like tip number one. Don't tip number two.

00:01:53   Seriously. Don't tip number three, seriously leave your car where it is.

00:01:57   You know, if you're worried about the battery,

00:02:00   start the car up and just leave it in park for five minutes to warm it up and,

00:02:04   and then just turn it off and go back inside. And number four, if you really,

00:02:08   really have to go somewhere that the only secret to driving on ice is

00:02:13   to go like an inch a minute. Oh, there is no,

00:02:18   you don't the traction. Oh, there's,

00:02:20   we have a Subaru and I noticed the first time this year it was,

00:02:23   is a gift from my in-laws when they stopped driving and I was like,

00:02:26   what is that X button X mode button? I don't know what this does.

00:02:29   And I look it up and it's like, Oh,

00:02:31   it engages every kind of mode to improve traction and low gear and all at one

00:02:36   separate tire function. And I'm like, Oh, I didn't even know we,

00:02:40   this it's like X mode is really like ice and bad hill mode. I'm like, okay,

00:02:46   well next time we'll press the button. I guess, I don't know if you want to,

00:02:49   if you want to treat and no one was hurt every time we have ice in Seattle,

00:02:53   you just type in like queen and ice or something like that into,

00:02:57   and that's our tallest Hill in Seattle. And there's one very steep Hill.

00:03:01   That's the counterbalance Hill where they used to run a cable car and they had a

00:03:04   counterbalance that's still present underneath,

00:03:07   even though there's no cable cars and people, you know, we're like, well,

00:03:09   it's just, it's not that bad. And then you go,

00:03:11   it's like if you were curling with cars and this was happening all the time,

00:03:16   all over queen Anne and people were like, Oh, Nope, they're either,

00:03:18   they'd be shooting it out the window. Oh no, they, Oh,

00:03:22   now it's going down and the person in the car would just slide sideways and

00:03:25   bumper car. And Oh anyway, we got through it.

00:03:29   Well, I hope everybody out there is all right.

00:03:31   One thing that I feel certain is guaranteed given the quantity of people is

00:03:36   that there's no doubt in my mind that there are numerous people who will be

00:03:41   listening to this episode,

00:03:42   hearing your and my voices who were hit by the Southwest airlines.

00:03:47   I don't even know what to call it. Collapse fiasco.

00:03:51   I'm fascinated by it because I'm always interested in the way systems break

00:03:56   down. Yes.

00:04:00   And I'm always interested in companies that are customer focused,

00:04:05   you know, and, and I am, I am a fan. I, you know,

00:04:10   I wrote about it briefly yesterday or the day before. I,

00:04:13   I like Southwest airlines a lot. I don't fly them.

00:04:16   I used to fly them as my primary airline and in the early days of doing daring

00:04:21   fireball professionally,

00:04:23   it was really the difference between me being able to afford to go to all of

00:04:29   these various Apple events that I went to and not,

00:04:33   it was like around the mid two thousands, 2006,

00:04:37   2007 that that era early days of DF,

00:04:40   me doing it professionally Southwest had super low flight.

00:04:44   I mean just unbelievably lower cost flights than any other airline from Philly

00:04:49   to the Bay area, to Austin for South by Southwest.

00:04:53   Oh yeah. And a couple other places.

00:04:55   One of the things that they had done was that they had,

00:04:58   it reminds me of Apple.

00:05:00   Remember when Apple locked in the flash memory prices?

00:05:05   It was like in the, in the iPod era. Yeah. Yeah. Yeah.

00:05:09   And it was like everywhere they had prices that were ridiculous compared to

00:05:13   everyone else cause they'd bought up all the capacity, right?

00:05:15   Or a huge amount of capacity.

00:05:17   It was like a maybe the move that secured Tim Cook as the future leader of the

00:05:22   company.

00:05:23   Oh yeah. Right. And then they kind of did it again with the iPad too when it was

00:05:26   released.

00:05:27   That was I think why that price differential was so huge as they had secured a

00:05:31   massive amount of expense or not expensive at parts that when you buy them on

00:05:35   the open market or we're much higher. Yeah. But like 15 ish years ago,

00:05:40   Southwest did something similar with fuel prices where they had locked in like a

00:05:45   longterm contract on jet fuel and,

00:05:49   and the prices skyrocketed.

00:05:50   But Southwest uniquely had locked in these low prices and for a couple of years

00:05:55   it was just like night and day, many,

00:05:59   many hundreds of dollars to fly coast to coast from Philly to the Bay area

00:06:03   versus, you know,

00:06:04   I remember getting like $129 coast to coast flights, you know,

00:06:07   on Southwest and,

00:06:09   and flying Southwest is not like flying what you think of as a budget airline.

00:06:14   Right. It's, you know, it's different.

00:06:20   They have the weird rule that throws everybody off. I think they still do it.

00:06:24   It's been a while since I flew Southwest where you don't get a reserved seat and

00:06:29   you sort of bored like, I don't know,

00:06:31   like getting on a bus or actually Amtrak has started doing reserved seats,

00:06:36   which I actually really like, but more or less, you know,

00:06:39   it's like you get like a boarding group number and first on first choice of

00:06:44   seats and you just sit wherever it, there's an open seat.

00:06:49   And they figured out that by doing it that way,

00:06:53   they actually fill the plane faster than if there are reserved seats.

00:06:57   And that was the number one reason why they did it that way.

00:07:00   And it throws everybody who's used to other airlines off,

00:07:03   but once you get used to it, it's, it works out.

00:07:05   There's often the single biggest hiccup is like, Oh, you know,

00:07:10   and then the flight attendants will come on the PA system and say,

00:07:13   there's like a mother and a young child. And there's only like two middle seats,

00:07:17   10 rows apart that are open. Would any,

00:07:21   would anybody be willing to move their seats?

00:07:24   So this mother and young child can board together. Whereas, you know, and again,

00:07:28   well, I'm a rotten person, Glenn. I never raised my hand,

00:07:32   but I gotcha. God'll get you John. Well, but you know, it's, but you're, see,

00:07:36   this is, this is that I don't think Southwest,

00:07:39   some other airlines are much worse about this, right?

00:07:40   In which they sort of pit our personal and are like our generous and selfless

00:07:45   interests against each other and Southwest from what I've,

00:07:48   I've never flown Southwest.

00:07:49   They don't really service the areas that I go to from Seattle or they're not the

00:07:53   best deal.

00:07:53   I can always get Alaska is often our kind of airline for lower cost and certain

00:07:58   routes and Alaska has nonstop to a lot of places really affordably.

00:08:01   So we're really lucky in our hub,

00:08:02   but Southwest doesn't seem to put everybody into like a prisoner's dilemma or a

00:08:07   Hobson's choice or any of the other scenarios they, so when you get that,

00:08:11   you know, Hey, could somebody move? Because we have, you know, that situation,

00:08:14   that's like a minority of the time, right? It's not,

00:08:17   you're not constantly feeling like you can't get your bags. Like, you know,

00:08:20   I just flew to,

00:08:22   my kid did a very long European trip bumming around Europe recently and I joined

00:08:26   him for the last two weeks. I was in central Europe for a couple of weeks in

00:08:28   November, December, and just flying these long haul, you know,

00:08:31   the like two segments, they used miles both ways and they're, you know,

00:08:35   you got on the plane, like it feels like there's like freight car storage above

00:08:39   you. They're like, we might not have room for all the bags. I'm like, what the deal,

00:08:42   you, what are you doing?

00:08:44   It's like you could put several people up there and you're like, well,

00:08:47   if you don't get on first, if you're not in the boarding,

00:08:48   this ticket class doesn't include a carry on bag this bigger and you're not

00:08:52   here. And I'm like, Oh, just, you know, you're trying to make every dollar.

00:08:56   And all you're doing is making everybody, you know,

00:08:58   the airlines are determined and maybe Southwest before this excluded to make

00:09:03   everyone feel about them like the cable companies and telephone companies.

00:09:06   And I'm like, come on, you could be, you could be better,

00:09:09   but they're so eager to get it to scrape every dollar out. I don't know.

00:09:12   Well, Southwest,

00:09:13   so you've probably read all these great accounts from people who have airline

00:09:16   system experience and it's like South.

00:09:19   I thought the best description I read was Southwest. This is,

00:09:22   I think the a seat you were listening to that seat 31 B it was that Southwest is

00:09:26   a, is a, you know, it's national airline,

00:09:28   but it likes to pretend it's kind of a regional point to point airline when it

00:09:33   suits them because they want to argue that they don't have the same,

00:09:36   they shouldn't be subject to the same rules as the mainline. Yeah. No airlines,

00:09:40   like a United or something. And it's like, no, they really are.

00:09:43   And this is when it breaks down because their system is designed on a knife's

00:09:47   edge and the knife cut through it.

00:09:48   Yeah, it's, it's really, I mean, and again,

00:09:51   they are a customer focused airline, you know,

00:09:54   and maybe by the standards of overall customer

00:09:59   focused companies period,

00:10:01   even the best airline in the United States is actually not that great, but, but,

00:10:07   but given the state of U S air travel, Southwest,

00:10:11   in my opinion is an excellent company, but it, it, it's like a,

00:10:15   I don't know,

00:10:16   some kind of worst case scenario where everything that could go wrong for

00:10:20   Southwest went wrong and they had a crew

00:10:25   at certain airports stranded who couldn't get to where they needed to go

00:10:30   and get into an airport hotel. Right.

00:10:32   Cause Southwest doesn't have other arrangements with hotels.

00:10:34   They couldn't get them rested in time to redeploy them. Right. And you know,

00:10:38   for, for good reason, you know, there are,

00:10:40   are rules about how much rest, you know, airline staff, you know,

00:10:45   pilots and flight attendants need, you know, before they can work.

00:10:48   And they had planes at other airports, you know, but, but no crew.

00:10:53   So they had planes at certain airports, crew at other airports,

00:10:56   and no way to catch up, which is just so strange.

00:11:01   It doesn't think like something that could happen, but it did. And to me,

00:11:06   the most telling fact was that yesterday, Wednesday,

00:11:11   99% of all canceled flights in America were Southwest flights.

00:11:16   Yup. And they're still at 50% when we record this.

00:11:20   And they're about, they claim on December 30th,

00:11:22   there'll be back to essentially full operation, you know,

00:11:26   unless some of the post-it notes fall off the wall where they've been tracking

00:11:29   flights. Right. And it's, you know, and it's one of those things where I,

00:11:33   I always remember this phrase, I forget who I learned it from,

00:11:35   but better necessarily implies different, right?

00:11:40   That to be better, you have to be different. And it's,

00:11:45   it's such a good phrase to keep in mind because human nature is to

00:11:49   eschew differences, right?

00:11:51   We tend as humans to

00:11:55   avoid things that are weird or different. And, you know,

00:11:59   we could go on and on about how that sort of was the,

00:12:02   the problem Apple faced with the Mac for decades, right? Was that, ah,

00:12:07   the Mac is too different and it, it was different and that's what made it better.

00:12:12   But people are like, ah, I don't want something that I'm not used to.

00:12:15   And Southwest is sort of like that for an airline where it's like, ah,

00:12:19   it's weird. You don't get a reserved seat.

00:12:22   They don't participate in any of these cross airline things where,

00:12:26   where like worst case scenario,

00:12:28   you're flying United and something bad happens and they don't have

00:12:33   airplanes.

00:12:34   It's like United can somehow work something out with American and maybe stick

00:12:38   you on an American flight. If there's like an empty seat,

00:12:42   and you don't even have to pay. And because they have some kind of,

00:12:45   I don't know what they call it, like a partner.

00:12:47   I learned the term it's called an interline agreement. I just,

00:12:50   now you learn something new when things belt down.

00:12:52   There you go. Interline agreement. Southwest has no interline agreements.

00:12:57   And you know,

00:13:00   in some ways that served them well over the decades because it's made them

00:13:04   different. But then something like this happens and they're like,

00:13:07   and they're telling these poor customers and I'm again,

00:13:11   again, I'm sure that some of them are out there listening to us talk about it.

00:13:14   And, and my,

00:13:15   to me,

00:13:16   I am not phobic in the least bit about airplane crashes because I

00:13:21   truly, I, for whatever reason,

00:13:24   the way my brain works,

00:13:26   I've completely internalized the statistical

00:13:31   truth that the safest way to get from point a to point B is in an

00:13:35   airplane. Right? Yeah. Yeah. So I'm not worried about that. To me,

00:13:39   my phobia whenever I travel air travel is exactly what happened

00:13:44   to people trying to travel on Southwest this week,

00:13:47   which is you get to the airport, you go through security,

00:13:51   and now you're on the other end of security and flights start getting canceled.

00:13:55   And there's a six hour line, literally six,

00:13:59   six hour line to talk to somebody at the desk. And what,

00:14:04   what do you do?

00:14:04   You're waiting. Yeah. You're stuck. You know,

00:14:06   I'll tell you my worst airplane thing.

00:14:08   And this happened to me once is although not as bad as some of the incidents you

00:14:11   heard a few years ago is when they put you on the plane,

00:14:14   they take you out and then they're like,

00:14:15   we're delayed and you're on the plane for hours. Then they're like,

00:14:18   you know, you can't get up and use the bathroom. They're like, all right,

00:14:20   we're going to let you use the bathroom. They're like, okay,

00:14:21   all the bathrooms are now closed and you're like, and, and we can't, oh,

00:14:25   and there's no gate for us to go back to. Right. Like some kind of,

00:14:27   and then that happened.

00:14:28   There were a few planes that happened to a few years ago and that that's when

00:14:31   the department of transportation pushed out a bunch of new, like it was,

00:14:34   it took that impetus to put out new rules and penalties and, and things. I,

00:14:38   I also expect what Southwest has been doing, I think along the way, like, you know,

00:14:42   despite this meltdown, you and I know, and I think all the listeners know,

00:14:46   unlike some organizations and companies,

00:14:48   nothing they're doing here is malicious or in the interest of making more money.

00:14:52   In fact, they would be delighted to not be in this situation.

00:14:54   There's nothing good about this for them and they didn't do it intentionally.

00:14:58   Although they put out some nasty note to their workers a few days ago that may

00:15:02   have had an impact where they're like, people are taking too much,

00:15:04   too many fake days off or something. And that, you know, soured the pot,

00:15:08   I think for what's coming on. But, but that said,

00:15:11   every statement they've made is like, they literally put on their website,

00:15:15   this is unacceptable or this is, you know, they, in their public statements,

00:15:19   they've said, what we're doing is wrong and bad. And I'm like, all right,

00:15:21   well that is kind of disarming. And they've told everyone who's asked,

00:15:24   we're going to pay you back.

00:15:26   And so it doesn't help when you had to spend 48 hours in an airport and miss

00:15:29   Christmas. The money doesn't help, but you're not going to be monitored.

00:15:32   If you had a book, an airline, like some people are like,

00:15:34   I had to spend a thousand dollars to rebook on United.

00:15:36   We rented a car and we spent $700 and we drove and we had to put it air,

00:15:40   whatever.

00:15:40   And Southwest so far has been saying we're going to pay you back for all the

00:15:44   expense you incurred. I just gonna have to send a lot of receipts.

00:15:47   So it's a hassle, but they're not like, well, this isn't our problem.

00:15:50   The historic thing and ice, blah, blah. They're not trying to play that game.

00:15:53   And if they do this well, I mean, some people are complaining already.

00:15:57   They're having trouble with refunds cause the airport plane airlines still

00:15:59   melted down. But if they do this well,

00:16:01   they come out of it and they put it cause they're going to have to push out like

00:16:05   all their flyers. They're gonna have to push out vouchers to everybody.

00:16:07   They're going to have to pay them back. They're going to have to have fair sales.

00:16:11   It's going to take them years to reestablish and they're all gonna have to

00:16:14   promise like we're going to rebuild our system, you know,

00:16:16   and rebuild relationships with all of their employees who have been all abused

00:16:20   in different ways.

00:16:21   And all the gate agents who've been screamed at by people have been stuck in

00:16:25   airports for 48 hours. There's no excuse for doing it.

00:16:27   Cause not the gate agents problem,

00:16:28   but people are going to do it cause they lost it there. They haven't slept.

00:16:32   So anyway, there's years of rebuilding, but so far I'm like,

00:16:35   I can't think of another company that has been as objectively like, yeah,

00:16:39   this is our fault at Southwest when something during the meltdown part,

00:16:43   not during the face saving part later.

00:16:45   Yeah. Yeah. I think they're doing an okay job with the crisis management,

00:16:49   you know, and, and you know,

00:16:51   thankfully nobody has actually gotten hurt because of this. So,

00:16:54   so maybe it's not the best analogy,

00:16:56   but of course the gold standard is the Tylenol recall back in the

00:17:01   eighties where there was what, what was the, as I recall the, the basic,

00:17:05   it really was true. What it was like some, some lunatic never caught,

00:17:10   right? Never, never. I forgot that part. I was looking it up recently. Yep.

00:17:13   They like injected cyanide into some tablets or replaced some tablets with

00:17:18   cyanide, but it was like a handful at a handful of stores, right?

00:17:22   It was very limited. They absolutely knew the serial numbers or whatever,

00:17:27   you know, whatever the numbers were.

00:17:29   And it was a certain tablet from a certain factory and then they,

00:17:32   they could have just recalled just, just the laced packages.

00:17:37   And instead their CEO at the time was like, Nope,

00:17:40   let's every single thing with the name Tylenol in it,

00:17:44   whether it's cough syrup or, or tablets or whatever, if it says Tylenol,

00:17:49   everything comes off the shelves, everything gets destroyed.

00:17:51   And we'll start over from scratch and guarantee that this will never happen

00:17:56   again. And that's when they rolled out the tamper proof caps too,

00:17:59   or the flute seal. That was part of that operation. I think as they retooled,

00:18:03   they took months, I think there was no Tylenol.

00:18:06   You can only get generic or whatever on the market. And then they, right.

00:18:09   But there was, there was, there was speculation when it first happened. And,

00:18:14   you know, and that, you know, it really wasn't, it wasn't like an urban legend.

00:18:17   There really were a limited small number of Tylenol packages that had cyanide

00:18:22   from a lunatic.

00:18:23   And there was a lot of speculation that this would be the end of the brand,

00:18:26   that they would have to shut down. Nobody's ever going to buy it again.

00:18:29   And instead because they reacted so strongly and so

00:18:34   everything to put everybody's mind at ease by, by literally taking everything,

00:18:39   whether they knew that it couldn't possibly have been involved with the same guy

00:18:43   or not everything off the shelves, like within six months,

00:18:46   their brand trust numbers were actually higher than before the incident,

00:18:51   which is crazy. I don't know that that's going to happen with Southwest.

00:18:54   Honestly, I, I, this is going to take to me,

00:18:57   this is going to take some time because if you do the math and I did the math and

00:19:02   I was like, just the back of the envelope math, I was like, wait,

00:19:04   that's too many people that can't be right. But Southwest,

00:19:08   part of their gimmick or,

00:19:10   or their operational excellence is that they only fly one type of airplane

00:19:14   Boeing 737.

00:19:15   Yeah. Yeah. Yeah. It makes it much simpler for, right.

00:19:18   That's one of their things. If you don't like that plane, don't.

00:19:20   Yeah. Yeah. If you don't like a Boeing. Yeah.

00:19:23   And they have slightly different seating arrangements where there's like always

00:19:27   the it's, they, they don't have first class or business class,

00:19:31   but what they have are very generously spaced exit rows,

00:19:35   which is, you know, if you fly Southwest, that's like,

00:19:37   that's like the in demand seats,

00:19:40   including one on every plane that doesn't have a front,

00:19:44   a seat in front of it. So you can really stretch your legs out.

00:19:46   There's like one seat with no, no seat in front of it in the exit row.

00:19:50   Sometimes it's on the left, sometimes it's on the right,

00:19:53   but it has all of their planes are Boeing 737.

00:19:56   So like the engine parts are all interchangeable and

00:20:01   the, the number of seats is identical.

00:20:04   So any Southwest flight can fly any, any,

00:20:08   any Southwest plane can stop, can fly any Southwest flight.

00:20:12   And it's part of their general operational excellence where they,

00:20:17   you know,

00:20:17   tend to lead the industry and on-time departures and arrivals and stuff like

00:20:21   that. I forget where I was going with it.

00:20:23   It's the homogenization of the right, right.

00:20:26   There's where I was going with it is that I,

00:20:31   I think that people are going to be a little bit slower to forgive

00:20:36   the people who were affected. Cause if you do the back,

00:20:39   the back of the envelope math,

00:20:40   which is where I was going is there's 143 seats on a Southwest flight. Oh,

00:20:44   I see. Okay. And they tend to fly.

00:20:47   They also tend to lead the industry in how full their planes are,

00:20:51   which is how they're so efficient. So 2000 and they, this week,

00:20:56   they've averaged more than 2000 canceled flights a day. Cancel not. Oh,

00:21:00   I see. Yeah. Yeah. So you were talking,

00:21:02   do the math 2000 times 140 is like 330,000.

00:21:07   So 330,000 passengers a day with canceled flights for like,

00:21:12   I really thought about the numbers.

00:21:13   So a million plus people this week or a million passengers easily. It's easily,

00:21:17   it's easy to, yeah.

00:21:19   Easy to figure out that there's a million people and there just isn't that much

00:21:23   surplus.

00:21:24   Even if you can squeeze people onto United and American and Alaska and whatever

00:21:29   other airlines, there's just no way to,

00:21:30   there's no way to put a million people on onto otherwise booked planes. So I,

00:21:35   I think it's going to be a while before Southwest recovers, but you know,

00:21:39   I wouldn't bet against them because I do think that fundamentally it's a company

00:21:43   that is customer focused. So we shall see.

00:21:46   It's it's fun to fly different airlines because you get to see,

00:21:50   get to see how they perform. And I flew Lufthansa. This is again, under my,

00:21:54   you know, United partners. So I used to use miles and I got the, you know,

00:21:57   the coach Lufthansa experience is honestly better than most, you know,

00:22:02   it's, it's not quite business class, but it's, it ain't too shabby for their

00:22:06   long haul flights. So it was kind of funny. I was like, Oh yeah, that's right.

00:22:10   You can run a really profitable airline that, that is,

00:22:13   it's really sharp in the air too. So I don't know. It's, I think, you know,

00:22:17   actually there's,

00:22:18   I think there's the biggest change that I think in airlines is that the new

00:22:22   seats are so thin compared to those old bulky ones.

00:22:25   And I think they're generally more comfortable too.

00:22:27   And they didn't give us all the room back in, you know, in knee space.

00:22:31   But I think I feel like the newest planes,

00:22:34   even in the worst coast sections are more comfortable than some of the better

00:22:38   planes in the, you know, 20, 30 years ago. I don't know. I don't,

00:22:41   maybe it's just me. I really,

00:22:42   my behind falls asleep and my back goes wonky on longer flights.

00:22:47   And I managed to get through, you know,

00:22:48   like 14 hours in one direction and 12 and the other pretty with two segments.

00:22:52   So, yeah, I, I know I agree with that. I think when you,

00:22:55   when you first board a plane and you, you sort of look around and,

00:22:59   and you can quickly discern, was this plane recently renovated?

00:23:03   Or is it, are these seats like 30 years old?

00:23:06   A new plane definitely is more comfortable. No, there's something,

00:23:09   there's something to the, the seat technology that they've, you know, they've,

00:23:13   they, you know, they, they, they squeeze every inch out of the leg space,

00:23:18   but by making the seat a little thinner,

00:23:21   they kind of gain back an inch or two and an inch.

00:23:24   Cause you gotta be a certain depth, right?

00:23:26   You're always going to have to be back in the seat a little bit.

00:23:28   So I can like cross my legs and coach and I flew my first 737 Max's on a

00:23:33   trip to Ohio in October. And I was like, huh, it's like 737 Max. Like, well,

00:23:38   they've been flying enough of them long enough. I'm not going to worry about it.

00:23:40   It was still like, all right, first time out on one of these is the planes are fine.

00:23:44   It didn't, I didn't have any, any worries,

00:23:46   but it was just kind of funny to suddenly be like, Oh yeah,

00:23:48   that was that thing, wasn't it? Before the pandemic. Yeah.

00:23:52   That was the thing everybody was talking about. Oh, geez. All right. Anyway,

00:23:55   let me take a break here.

00:23:56   Thank our first sponsors are good friends at Linode go to L I N O D E.com/the

00:24:01   talk show linode.com/the talk show and see why Linode has been voted the top

00:24:08   infrastructure as a service provider by both G2 and trust radius.

00:24:14   They've got award winning tech support offered 24 hours a day,

00:24:18   seven days a week, 365 days a year. New Year's day is coming up.

00:24:23   You need tech support for your Linode account.

00:24:25   Somebody is going to be there to talk to you. That is true.

00:24:28   It is absolutely great tech support and it's just so easy to use and set up.

00:24:33   This is where I host during fireball. I've been there for years. I would do it.

00:24:38   Even if they didn't sponsor my podcast, I don't know where,

00:24:41   I don't even know what my second choice would be.

00:24:42   That is how much I love hosting my stuff at Linode,

00:24:45   even though my needs are relatively very simple,

00:24:48   but you can deploy your entire application stack with our one click app

00:24:52   marketplace or build it all from scratch.

00:24:56   Manage everything from source yourself with supported centralized tools like

00:25:00   Terraform. So you want to get nerdy, you want to be at the command line,

00:25:04   you want to build it all yourself. You can do it. You want to just click,

00:25:07   click, click and do prebuilt stuff that they've guaranteed to work.

00:25:11   You can do it that way too. You can mix and match. They've got it all.

00:25:14   And they offer the best price to performance value for all compute instances,

00:25:18   including GPU's as well as block storage, Kubernetes,

00:25:22   bare metal, all of it.

00:25:24   They've got a new database hosted service,

00:25:28   my SQL, Mongo DB, Postgres SQL,

00:25:31   fully managed databases for all of the SQLs that you could want Mongo DB

00:25:36   more than that. And it is easily deployed high to high performance database

00:25:41   clusters with backup disaster recovery, data replication,

00:25:46   everything you can need. It's all built in.

00:25:48   That's one of their new features for 2022 go to Linode,

00:25:52   L I N O D E linode.com/the talk show.

00:25:56   By going to that URL with the talk show as the slug at the end of the domain,

00:26:01   you can create a free account using your Google or GitHub account,

00:26:06   or just use your email address. That's what I do.

00:26:09   And you will get a hundred dollars in credit just for going there.

00:26:13   Linode.com/the talk show a little bit of news.

00:26:18   It's surprising to me a little bit how news Apple newsy December has been.

00:26:23   Usually it's pretty quiet, but two pieces that sort of seem related to me,

00:26:28   related to Apple Silicon, for lack of a better catch all term.

00:26:33   Mark Gurman had an excellent report in the,

00:26:37   at Bloomberg on the Mac Pro, which is the obvious,

00:26:42   we all know this, this isn't speculation or rumor.

00:26:45   When Apple announced the move to Apple Silicon at WWDC

00:26:50   2020,

00:26:51   Tim Cook said that this transition will be complete by the end of 2022.

00:26:56   And it turns out it's not because of one, one Mac,

00:27:02   the Mac Pro, which is the only one that is still only on Intel.

00:27:05   It's part of the backstory is this spring

00:27:10   when Apple unveiled the Mac studio, which is a new form factor.

00:27:15   I think one of the big stories of 2022,

00:27:17   cause it's a form factor that Mac Pro Mac users

00:27:22   have been dreaming about for, I think literally decades,

00:27:27   you know, certainly since the G4 Cube,

00:27:29   which was the last sort of attempt at a sort of

00:27:34   small form factor, bigger than a mini,

00:27:38   but pro spec features Mac when they unveiled it,

00:27:43   John Ternus in their pre filmed event said this, you know, you know,

00:27:47   we've got one more Mac to go, the Mac Pro, he even mentioned it,

00:27:51   that, you know, the Mac Pro is still coming this, you know,

00:27:55   more or less, I think to emphasize that

00:27:58   however well spec the Mac studio is, and it is very well specced.

00:28:05   It is not the Mac Pro on Apple Silicon.

00:28:08   And I think that if Ternus hadn't said that in their event,

00:28:12   a reasonable conclusion for everybody outside Apple would be, Oh,

00:28:17   maybe the Mac Pro brand is dead. And in the Apple Silicon era,

00:28:21   the Mac studio is the Mac Pro.

00:28:23   This is the beefiest Mac desktop you're going to be able to buy.

00:28:28   So deal with it. Let's, you know, let's treat it that way. But they didn't,

00:28:33   they came out and said the Mac Pro is still coming. And the Gurman report,

00:28:37   I don't know how best to summarize it,

00:28:39   but basically is that they ran into that.

00:28:44   They had planned to have an M one based something, something, you know,

00:28:48   like something like a double M one ultra,

00:28:52   like two ultras attached together.

00:28:55   And that would be the foundation of the Mac Pro and it didn't happen.

00:28:59   And now who knows?

00:29:01   Although it makes, I mean, it made sense. I would have been shocked.

00:29:05   It would have, well,

00:29:06   it would have been interesting cause I would have been shocked if they'd release

00:29:09   an M one based Mac Pro because we all knew an M, you know,

00:29:12   we all knew there's another number coming right.

00:29:15   And we know that another number is going to represent an evolution of, you know,

00:29:18   what they learned in the first one. And it's well as you know,

00:29:22   as you've discussed many times,

00:29:23   I think you were one of the best people writing about technology who talks about

00:29:28   lead time because everybody who reports in the space is always like, well,

00:29:32   Apple made a sudden change. And you're like, look,

00:29:33   let's back that out to eight months ago when they had a design it, test it and,

00:29:38   or whatever, or sometimes 14 months or two years.

00:29:40   So the M two was well underway when the M one shipped,

00:29:44   like it was deeply figuring out, you know,

00:29:46   from the early stages of M one production. So, you know,

00:29:49   you're in that standpoint,

00:29:50   you're a product manager or Tim cook or whatever level the company,

00:29:52   these kinds of decisions are being talked about to percolate up.

00:29:56   And I would imagine you'd be hard pressed to say, well,

00:29:58   let's just do an M one version cause we could do it. Be like, we said two years,

00:30:01   we got some leeway. The M two is, you know,

00:30:04   X percent more efficient on that and this, we figured out this other problem.

00:30:07   Let's, we need the double M two ultra or whatever we're going to do.

00:30:10   That's going to be the pro chip and everybody like, Oh, okay, we'll get,

00:30:14   you know, the studio will ship that's on the horizon.

00:30:17   That's going to fill a certain amount of demand and we're going to kind of prime

00:30:20   the pump and we're going to, and unlike Apple's usual moves,

00:30:23   we're going to tell people at the event. Right. And so it,

00:30:26   I think it all makes logical sense except that they could not apparently produce

00:30:29   the M two they wanted to. Right. That seems like, I mean,

00:30:32   it seems like fundamentally there was an issue with, with production,

00:30:36   like actually getting it into manufacturing, into the fab.

00:30:40   The M here's I'm quoting from Gurman's report.

00:30:43   The M two ultra chip is destined to have some serious specifications for

00:30:46   professional users, including up to 24 CPU cores,

00:30:49   76 graphic cores and the ability to top out the machine with at least 192

00:30:54   gigabytes of memory. All right. I'm going to stop there and just say, that's,

00:30:59   that's a lot of Ram, but,

00:31:03   but with the current Mac pro, the,

00:31:06   the last Intel based Mac pro it's a terabyte of memory,

00:31:11   I believe. Yeah, it was a huge, right. Cause they were very careful about that.

00:31:14   Right. Right. Speaking of that after the lawsuit. So if you,

00:31:18   but they claim and the M series is supposed to be so much more memory efficient

00:31:23   though, I mean, I think that's part of the argument, right?

00:31:24   Well, I don't know it, but it is a weird downgrade, right? It's, it's, you know,

00:31:30   literally like one fifth of the Ram at the maximum end.

00:31:34   And I realize there are very few people who have memory needs like

00:31:39   that, but if you really do need that much memory, you need the memory, right?

00:31:43   There are some people in scientific computing and who are doing certain things

00:31:47   who are memory constrained at a terabyte, right? You know,

00:31:51   at the extreme end, you're always going to find outliers like that.

00:31:55   And if you're already happy to have a terabyte,

00:31:59   if you wish you could have more than a terabyte maxing out at 192 gigabytes is

00:32:04   a serious downgrade,

00:32:05   no matter how efficient apple silicon is overall versus the Intel platform.

00:32:10   All right. Continuing from Gurman's report, which is the,

00:32:14   the M two ultra would be the equivalent of the high end chip.

00:32:18   That's currently in the the Mac studio, right?

00:32:21   What he's calling in again,

00:32:23   he's not saying that this would be the marketing name,

00:32:26   but maybe it would be the M two extreme would be two M two ultra is put

00:32:30   together in the way that the M one ultra is two M one maxes put

00:32:35   together. Right? So M one ultra is two M one maxes.

00:32:40   Gurman is saying that their original plan was to have an M something extreme,

00:32:45   which would be two M whatever ultras,

00:32:48   the M two extreme would have doubled that to 48 CPU cores,

00:32:53   152 graphics cores. But here's the bad news. The company is likely likely,

00:32:58   that's a very key likely scrapped that higher end configuration,

00:33:03   which may disappoint apple's most demanding users, blah, blah, blah.

00:33:06   There's a thing hidden in here too is all of the M excuse me,

00:33:11   all the M one chips to date have only on like a sock like system on

00:33:16   a chip memory. Right? Right. So this,

00:33:18   to get 192 gigabytes even you would have to,

00:33:21   you they can put that on the same die. That's going to be external memory.

00:33:24   So conceivably some higher end ones, if those shipped, those would have,

00:33:29   you know, so it's, I'm right there, right? Is that that would all have to,

00:33:32   there'd have to be some amount of external memory or not. Maybe with 192,

00:33:36   maybe not. But it feels like you start to get, are they,

00:33:39   they're still in the four nanometer process size. So how big would a die?

00:33:43   I'm already thinking how big the die is with 24 CPU is to 76 graphic cores.

00:33:48   You know, you start to get, I don't know. I mean there are limits, right?

00:33:51   There's limits and yield issues and manufacturer issues.

00:33:54   And I wonder if that's not like I'm an expert in this topic.

00:33:57   I've studied it in part, but I'm not a chip maker. But I know there are,

00:34:00   that's where they start to get into problem areas if they can achieve the right

00:34:04   yield.

00:34:05   And we saw a little bit of that with the M one and M two where some of their low

00:34:08   end chips that don't achieve the, the,

00:34:10   the testing get sold as the seven CPU or, or other GPU and so forth.

00:34:15   So they're, they already approached that with some, you know,

00:34:18   some end of what they're doing.

00:34:19   Yeah. To me, it's one of the most exciting,

00:34:22   even though I'm definitely not going to buy one, I have no,

00:34:24   no reason to get a Mac pro. I don't even need the Mac studio.

00:34:28   I'm very happy with my maxed out,

00:34:30   no pun intended and one Mac book pro as my one and only main Mac.

00:34:37   But as an observer, I'm super,

00:34:40   super excited to see what the Mac pro story is.

00:34:44   It's gotta be coming in 2023. And I don't know, maybe I,

00:34:49   this, this is not informed speculation at all.

00:34:52   No little birdies have told me this,

00:34:54   but I've just had it in my head that for the given the Mac studio,

00:34:58   once we saw the Mac studio and how beefed up it was,

00:35:01   but the Mac studio is still based on the idea

00:35:06   that whatever it is you order, it's all soldered onto the board.

00:35:11   So if you know, however much Ram you want to get,

00:35:14   you have to order it when you order your Mac studio.

00:35:17   So my thinking is that, well, what if, is there a way,

00:35:22   is that feasible that they could do a Mac pro that has a

00:35:27   bigger chassis so that it's expandable and friendly to an

00:35:32   enthusiast, you know, like,

00:35:33   like Mac pros over the years typically have been where you can open it up.

00:35:37   And this would be the one and only Mac that in addition to having

00:35:43   X amount of Ram soldered onto the systems on a

00:35:48   chip and to get that unified memory performance

00:35:53   advantage where the same Ram is used by the graphics system as is used by the

00:35:58   CPU system.

00:35:59   But what if there was also another area where you could just put

00:36:04   regular Ram chips in?

00:36:06   So if you,

00:36:07   what you really want to do is just max out the Ram and you

00:36:11   don't really need massive amounts of parallel CPU or GPU

00:36:16   processing,

00:36:17   it's really just Ram that you just want to put all this in Ram that they could

00:36:22   have the, the main Ram,

00:36:24   which is the Ram that's on the unified memory architecture that is soldered onto

00:36:28   the boards.

00:36:29   But then there's another level of Ram that's expandable and it would only be

00:36:33   used in cases where you're doing a computational

00:36:38   task that literally takes hundreds and hundreds of gigabytes of Ram,

00:36:42   if it's available.

00:36:44   And that secondary Ram wouldn't perform quite as well as the

00:36:49   unified memory,

00:36:50   but it's better than not having the Ram at all, right?

00:36:55   It's sort of like treating the unified memory that's on

00:37:00   the systems on a chip as sort of like a cash, you know,

00:37:04   like in CPU terms and the, the rest of the Ram,

00:37:08   if you add it is just there. I don't know.

00:37:11   I don't know if that's feasible or not, but it sounds to me like,

00:37:15   what is the point of still having the Mac pro if it's not expandable? If,

00:37:19   if this one product is not expandable in that way,

00:37:23   even if no other Mac hardware has user expandable memory.

00:37:28   Well, and yeah, and I mean, even from the storage standpoint, you know,

00:37:32   Apple became allergic to slots. They like, they like jacks, not slot.

00:37:36   And which is an int, which I'm sorry, it sounds obscene way of saying,

00:37:40   but the, which is weird because in a,

00:37:43   in a Thunderbolt world and a PCI-E world and a got it a word like

00:37:47   M2 connectors for SSD world or so running

00:37:52   over NVME trying to get all my acronyms in there in that world.

00:37:57   The difference between a port and a slot aren't the same as they used to be,

00:38:02   but because it used to be slots were faster and ports were slower, more or less,

00:38:06   right? Not uniformly, but it's kind of the thing you want to direct memory access.

00:38:09   You had to go to a slot inside something that you did a wide, you know,

00:38:13   multi pin card that went in for maximum parallel pin access at once for these

00:38:18   wide data paths. So that's kind of one area that is sort of funny.

00:38:22   And the other is that, you know, Apple's obsession with the sock,

00:38:25   I understand it because they brought, you know,

00:38:27   it reduces your cost of manufacture enormously, reduces the failure rate.

00:38:31   Like there's incredible it's savings.

00:38:34   And it's also as a user standpoint thing like this issue keeps coming up.

00:38:37   This is a little sidebar,

00:38:38   but I think it's relevant to the whole is I get email regularly at the Mac 911

00:38:42   email at Mac world about people with questions about like booting off external

00:38:46   drives with external volumes with an M1 or M2. And I have to remind people,

00:38:50   I'm like, remember if your internal drive dies on an M1 or M2,

00:38:55   your machine will not work.

00:38:56   There's no way because of policy management issues and it's part of the

00:38:59   security features. This is still, I think,

00:39:01   relatively less known because SSDs are generally very reliable.

00:39:06   Apple, you know,

00:39:07   typically buys the most reliable ones and they do burn in and all the rest.

00:39:11   So it's very, very unlikely now between the sock and the SSD,

00:39:16   the odds of getting a Mac that dies,

00:39:17   like I would love to know the hard failure rate in like three months and a year

00:39:21   on max now that aren't like a power supply issue. I've got to believe it's very,

00:39:25   very low. And it's also Apple doesn't have to source out memory, right?

00:39:29   It's got all these advantages. They're doing it in the dye and it's all great,

00:39:33   but it does seem to be, you know, that they kind of contracted into that.

00:39:37   And I feel like there's this great opportunity though,

00:39:40   cause there's still a lot of work being done in slot based products.

00:39:43   And there was a big movement at one point, which I think has died about the eGPU,

00:39:47   you know, that was a port based thing. But there's, I don't know,

00:39:49   I may maybe we're evolved past that.

00:39:52   Maybe you don't need external or you know,

00:39:54   slot based graphics cards or slot based memory anymore.

00:39:58   But I think your argument is good.

00:39:59   I don't see how Apple would put a terabyte on a die.

00:40:01   I was misspeaking a little bit here though.

00:40:03   192 gigabyte Ram on an Apple M whatever chip is not that ridiculous cause they're

00:40:08   putting 128 gigs on some on the M one ultra.

00:40:14   You can configure it to 128 gigs so they can do that today.

00:40:18   So 192 is not, you know, bizarre, but I think you're right. I mean,

00:40:22   there is a point which you can't get up to a terabyte or two terabyte for memory

00:40:26   bound issues. But, and I also think it's a correct too,

00:40:29   is there's going to be memory intent or sorry, yeah,

00:40:31   memory intensive applications and drive intensive applications.

00:40:35   And some of those will be CPU bound and some won't be the same way.

00:40:39   The CPU won't be waiting for memory or it's the delay is,

00:40:43   is not as a big a deal as being able to have all the adjustable memory available.

00:40:47   So it feels like there's a space for that.

00:40:50   How many people do you think need a Mac pro in the world now with them in the

00:40:55   Mac studio, you know, in with that out in the market,

00:40:59   I felt like that really shaved the edge off a lot of potential Mac pro buyers

00:41:04   into fairly high end applications.

00:41:06   But these are also going to be incredibly lucrative, you know, high, high dollar,

00:41:11   high margin machines. So Apple has a motivation to do it.

00:41:14   I, you know, that's a good question. And I do think, you know,

00:41:16   listening to various other podcasts where people talk about it and you know, I,

00:41:20   I, it comes up all the time is why does Apple even bother? Right?

00:41:25   Maybe, maybe there won't be a Mac pro or maybe the Mac pro will just be the Mac

00:41:29   studio in a bigger chassis and you get it just for putting additional

00:41:34   internal SSDs inside. You know,

00:41:36   I wouldn't be shocked if that's what happens, but I don't think so.

00:41:41   And I think the reason why isn't that it makes business sense that there are so

00:41:45   many people who need something above and beyond the Mac studio,

00:41:49   that it really makes great business sense in and of itself.

00:41:53   But I think it's similar. And I, I know I've made this analogy before,

00:41:58   but it's sort of like why car car makers get into formula one,

00:42:03   right? Why do Mercedes and Porsche and Honda and companies like that fund

00:42:09   formula one racing teams and build this technology because it,

00:42:14   there's a trickle down effect where pressing being at the

00:42:19   forefront of what is

00:42:21   theoretically possible today in raw compute power is

00:42:28   not a bad advantage to have for

00:42:32   what will trickle down to literally to iPhones three, four years from now.

00:42:38   Right. And, and I think that's been true for car makers, right? And it's not,

00:42:43   and the advantage,

00:42:44   the difference is with car makers getting into like making F1

00:42:49   race cars, it's just for the sport of it ultimately, right?

00:42:53   You're just making it.

00:42:54   Whereas making the equivalent of an F1 race car

00:42:59   desktop computer is actually practical for the

00:43:04   handful of top tier compute users. You know,

00:43:08   whether you're doing special effects for major motion pictures at

00:43:13   8k resolution or scientific simulations for, you know,

00:43:17   to go back to the Southwest thing,

00:43:19   you're an airplane engineer Boeing,

00:43:21   and you're running complex aerospace engineering simulations on a new

00:43:26   wing design for an airplane. And,

00:43:28   and you need as much CPU as much parallelism,

00:43:33   as much Ram as the fastest IO possible, you can't get enough.

00:43:38   You know, you're, you're nowhere near, you're,

00:43:40   you're still waiting for the computer for whatever it is that you do.

00:43:44   I think that there is an advantage to Apple being in that game

00:43:49   and being there that will trickle down a handful of years down

00:43:54   the road to the mass market products that truly fund

00:43:59   the company. I really do believe that that's true.

00:44:02   And I think they believe it.

00:44:03   And I also think Apple itself has many,

00:44:07   many people who need that sort of computer.

00:44:10   Yeah. I think you bring up a superb,

00:44:14   like other end of the telescope thing too, is the M1,

00:44:17   the initial M1 chip is the most powerful thing Apple made the wrong way up,

00:44:21   right? Like they released the lowest end possible chip in this series,

00:44:26   but it was such an outperforming chip.

00:44:28   And usually Apple is trickling down technology. They've,

00:44:32   they've developed for more sophisticated purposes and they have versions that

00:44:35   are meant for more consumer level ones. Right? Right. So the M1 is,

00:44:38   is was a buildup operation and they're still building up.

00:44:42   So I think that's right. As you get a cycle that through and we're seeing,

00:44:45   you know, the original M1s are falling off the plate. Eventually they'll move.

00:44:49   I would imagine they're going to move entirely to M2s unless there's a cost

00:44:53   reason to retain it. But the M1s are such good.

00:44:55   The machines that are M1 based are, have such good performance.

00:44:59   They're not like they're outdated. They're just a little slower.

00:45:01   And that's the amazing thing. So maybe you have to wait,

00:45:03   they come up with the M3 and then you lose the M1 generation.

00:45:06   But I don't know if that's totally new for Apple,

00:45:09   but I don't feel like they would build a new machine around the slowest possible

00:45:13   chip in that new architecture. And in this case,

00:45:15   the slowest possible chip was so fast,

00:45:18   had so many savings that it made sense for them to shift over there and then

00:45:22   gradually ratchet all their models up. So, but yeah, I mean if they only sell,

00:45:26   I think that argument is also valid to the Formula 1,

00:45:29   1 because Apple has to be aspirational. It has to have products that people want

00:45:33   to buy and sometimes they're going to go into purchasing and say,

00:45:35   I need a $40,000 machine. And they're going to say,

00:45:38   you can have a $10,000 one. We all right,

00:45:40   we'll get $10,000 and what now and we'll put more money in the budget next year.

00:45:44   But the fact that the $40,000 machine exists may give them more leverage to buy

00:45:48   a cheaper, cheaper one. And there's,

00:45:50   we also see all the time that when you have more computational power available

00:45:54   on a kind of in a desktop environment,

00:45:56   you start doing more sophisticated things that couldn't be done because they

00:45:59   required supercomputers.

00:46:00   So part of the AI revolution has been that neural networking could be done on

00:46:06   with originally with specialized graphic cards and then less and less requiring

00:46:10   that as computers, just the basic computational, you know,

00:46:14   central chips or the socks have improved. You can do incredible AI stuff.

00:46:18   And Apple devotes parts of its chip to this specific activity without having to

00:46:22   have machines that are so tricked out that they're unaffordable.

00:46:25   So as you get,

00:46:27   if you have a Mac pro that's an M two extreme based or whatever it is,

00:46:31   or an M three or whatever the next thing is,

00:46:32   that's going to provoke more research that will allow, you know,

00:46:36   open up new fields of research or so it'll open up new areas that researchers

00:46:39   will buy these machines specifically for because they have the capability on the

00:46:43   desktop to perform that kind of computation. Yeah.

00:46:47   The other story along the Apple Silicon line,

00:46:50   and this was published on December 23rd by Wayne Ma

00:46:55   at the information, which is paywalled excellent publication,

00:46:58   very expensive though. It's like 400 bucks a year. And every time I think, Hey,

00:47:03   it's a lot of money. Maybe I'll unsubscribe.

00:47:04   Then they drop a story like this that is right in my wheelhouse. And I'm like,

00:47:09   all right, I'll keep my subscription. But this is the messy soap opera back,

00:47:13   but, but,

00:47:14   but messy soap opera that affects a multi-trillion dollar company or a trillion

00:47:17   dollar company. It's so tricky. Quoting from Wayne Ma's story inside Apple's war

00:47:22   for chip talent. And a lot of this comes down to, and this is a fact, I mean,

00:47:26   it's not speculation, but,

00:47:28   and I don't think it's surprising given the success of the Apple Silicon

00:47:34   entire division that an awful lot of talented people have left Apple,

00:47:39   either to start their own startups doing chip design,

00:47:43   or they've been poached by Qualcomm or Intel or other companies, you know,

00:47:47   but that, you know,

00:47:48   it's the way it goes when one company has tremendous success in a certain area,

00:47:52   the talented engineers in that division become the

00:47:57   target of intense bidding from competitors,

00:48:01   right? That's, that's the way the market works,

00:48:04   but Apple has had some level of brain drain from their chip division. Now,

00:48:08   is it so much that the division is in trouble because

00:48:13   they've lost so much talent or is it that they have so much talent that they had

00:48:17   talent to spare and they can still continue to lead the industry? Who knows?

00:48:21   We'll, you know, we'll find out soon enough,

00:48:23   but I guess stories like this are sort of, you know,

00:48:26   certainly the information story to me is written with

00:48:31   the overall takeaway point that maybe Apple's

00:48:35   Silicon advantage is regressing to the mean and, you know,

00:48:40   which is it's what happens, whether it's sports, right? You know,

00:48:44   the Chicago Bulls are no longer the six time champions they were when Michael

00:48:49   Jordan and Scottie Pippen were playing 25 years ago. You know, it's,

00:48:52   things ebb and flow.

00:48:55   And even when you have a decade long run of tremendous

00:48:59   success, it tends not to last forever. Nothing,

00:49:04   you know, nobody, especially in Silicon, you know, even Intel,

00:49:08   the mighty Intel no longer is,

00:49:10   it was once unthinkable to think that Intel would be so far behind the state of

00:49:15   the art in chip performance, but here we are today and they are,

00:49:19   but maybe they'll come back.

00:49:20   We used to all be on MySpace also.

00:49:22   And it's sad to think that the saddest thing that the same pattern happens,

00:49:27   right? I mean, Intel had so many stubbles,

00:49:29   they like lost the mobile chip market like four times, three times,

00:49:33   at least X scale and a bunch of other stuff and a bet on the wrong wireless

00:49:36   technology.

00:49:37   And sold their stake in what do you call it? The arm, right?

00:49:41   Yeah, exactly. I tell you, here's the passing of a moment.

00:49:43   I have had a website registered Y max, Y max news,

00:49:48   dot blog.com or Y max net news.com or something.

00:49:51   I've actually forgotten the name for 20 years because I had my wifi networking

00:49:56   news site and then I launched some ancillary blogs with less news,

00:49:59   like cellular and public safety and so forth.

00:50:01   And those never got that much traction, but the Y max one did for a bit when,

00:50:05   and I was like,

00:50:06   I got a renewal notice and I'm like, why am I still paying for Y max news?

00:50:10   In 2022, I got to give it up. So I gave it up this year.

00:50:14   All right, let me read this story. The,

00:50:16   basically the gist of Wayne Ma's story is that the a 16

00:50:21   chip that is only available today in the eighth app,

00:50:24   iPhone 14 pro models, right? Infamously this year,

00:50:28   the non pro iPhone fourteens are still using

00:50:34   the a 15 chips from last year's iPhones.

00:50:38   Although the iPhone 14 non pro have the

00:50:43   better version of the a 15 that was in the eight,

00:50:46   the iPhone 13 pro has trickled down to the

00:50:51   regular non pro iPhone 14.

00:50:53   And that was a little baffling. I remember at the time everyone was trying to

00:50:55   assess that out. All right. But for the a 16, here's quoting from Wayne Ma,

00:51:00   Apple planned a generational leap for the graphics processor in the latest

00:51:04   version of the tie in smartphones, the iPhone 14 pro,

00:51:07   but engineers were too ambitious with adding new features and early prototypes

00:51:12   drew more power than what the company had expected based on software

00:51:16   simulations that could hurt battery life, make the device too hot,

00:51:20   according to two people, blah, blah, blah,

00:51:22   obviously because Apple discovered the mistake late in development,

00:51:27   it had to base the graphics processor in its iPhone 14 pro line,

00:51:31   largely on the design of the chip that went into last year's iPhone model.

00:51:36   According to four people familiar with the matter,

00:51:38   those people described the snafu,

00:51:41   which hasn't been previously reported as unprecedented in the group's history.

00:51:46   The iPhone 14 pro models,

00:51:48   which went on sale in September showed only small gains, blah, blah, blah.

00:51:51   One of the features, let me see if I can find it here in his story.

00:51:54   One of the features that was supposedly set and had

00:51:59   to be cut from apparently according to his report was hardware based

00:52:04   ray tracing. One of the, one of the,

00:52:07   one of the features that Apple had to cut from the iPhone 14 pros graphics

00:52:11   processor was ray tracing a computer,

00:52:13   computer lighting a technique that brings an extra level of realism to games.

00:52:17   A lack of big improvements in chips can make upgrading, blah, blah, blah.

00:52:20   I have heard this from friends. I have,

00:52:22   I stay informed in the games world through

00:52:27   my son who is a legit gaming enthusiast and some friends who

00:52:31   are have both feet in that industry. And I,

00:52:35   before this story in the information came out, that was something that I,

00:52:40   friends and my son had both put onto my radar,

00:52:44   which is that the state of the art and graphics,

00:52:46   one of the next things to look for 2023 2024 is built in ray

00:52:51   tracing support. And I've seen some demos, you know, like,

00:52:56   Oh,

00:52:56   here's this certain game with hardware based ray tracing on versus off.

00:53:00   And it's like, you know, the way all these modern 3d games look is so good.

00:53:05   It's like, so the, the before, as it were,

00:53:09   it still looks really good,

00:53:11   but then you look at the after with hardware based ray tracing and it is like

00:53:16   a real step towards photo realism. And

00:53:20   I mean, they've done non real time ray chasing has been a feature of computer

00:53:24   animation for 30 years, maybe longer than that without question.

00:53:28   But like, what's the speed, you know, you're running at, I don't know.

00:53:30   So one time it's probably a hundred or 200 times real time. Right. So,

00:53:34   Oh, even slower than that.

00:53:36   Cause I remember being at Drexel in the nineties when I was studying computer

00:53:39   science and there was, I remember every year the graduate students, you know,

00:53:43   their thesis projects, you know, would, would be there.

00:53:47   And you know, a lot of, a lot of stuff that you could do for a thesis in computer

00:53:52   science is it's just purely mathematical. There's nothing to look at,

00:53:57   but there in the mid nineties, ray tracing was huge. And there were definitely,

00:54:01   every year there were a couple of, you know,

00:54:03   graduate students whose thesis is, were in the ray tracing realm.

00:54:08   And therefore they had really cool demos or printed output.

00:54:12   Like look at this thing that, you know, the software they wrote wrote,

00:54:16   but you could watch it run on like what at the time were these,

00:54:20   we had a bunch of sun workstations in the computer science department that were

00:54:24   more or less, you had to be a grad student to get access to,

00:54:26   and you could watch them work. And it was like pixel by pixel by P you know,

00:54:32   like it was,

00:54:34   it made like the image writer dot matrix printer seemed like it,

00:54:38   it output stuff fast, right? Ray tracing a single frame,

00:54:42   super duper slow toy story. Of course, the first breakthrough feature film.

00:54:47   Yeah. Yeah.

00:54:48   It's like when you and you look at the original toy story and how

00:54:53   low fidelity it is compared to the modern Pixar films and other, you know,

00:54:58   3d animated movies of that kind,

00:55:00   but you talk to the Pixar or hear the Pixar people talk about how

00:55:05   long it took to render a frame.

00:55:10   It's been a thing in computer science graphics for a while.

00:55:13   It's the new frontier. It's,

00:55:16   it makes total sense to me that that would have been,

00:55:19   it was slated to be an iPhone 14 pro feature this year and got cut.

00:55:24   Cause it's definitely not, they're very, very interesting to me that, you know,

00:55:29   just the basic story that the,

00:55:31   the a 16 graphics were supposed to be a lot better than they were,

00:55:35   or then they actually turned out to be now they're not bad. I actually like,

00:55:39   you know, is geek bench metal,

00:55:42   the best benchmark. I don't know,

00:55:45   but it's certainly the easiest way to compare.

00:55:49   It looks like the a 16 is about seven or 8% faster at

00:55:55   graphics than the a 15 from a year ago. So, you know, nothing to sneeze at.

00:55:59   It's, you know, not a bad year over year improvement, but not groundbreaking.

00:56:04   And the other thing, here's the thing,

00:56:06   and you hinted at this a couple of minutes ago, neither,

00:56:10   my main frustration with both Wayne Ma story and the information and Gurman story

00:56:15   in Bloomberg about the Mac pro is that there's no dates of when

00:56:20   these internal crises were determined.

00:56:23   When did Apple determine that, Oh,

00:56:27   this design for the ACE a 14 or a 16 chip

00:56:31   is going to take up way too much heat and energy.

00:56:36   When was this a year ago?

00:56:38   A year ago, two years ago.

00:56:40   By your math and thinking, I got to believe this was,

00:56:43   was it like two years before the iPhone 14 pro shipped? I, I would have,

00:56:48   I mean, I don't know enough about, I don't know enough about the cycles,

00:56:51   but I would have a hard time believing that they went into production and the

00:56:55   chip. No, I mean, yeah, it's cause they need the,

00:56:59   the amount of time for each of those stages is too long.

00:57:01   So maybe it was 18 months ahead of the iPhone 14 pro shipping,

00:57:06   but that sounds,

00:57:07   that sounds really tight in terms of like scaling up the fab to get to the

00:57:13   yields you need and then integrating. I don't see, I just don't,

00:57:16   I don't believe it. So it could have been two years, right? I mean,

00:57:18   let's say outside is 18 months, but it could have been two years,

00:57:21   maybe more, but, but maybe more.

00:57:23   But I think the lay person reading these stories would think that these were

00:57:26   issues that came up like a year ago, January of 2022 or something.

00:57:31   And it's really frustrating to me and I'm not disputing their reporting.

00:57:35   I believe, you know,

00:57:37   but both Gurman and Wayne Ma and the information in Bloomberg are certainly

00:57:40   reputable. I, all my complaints about Bloomberg's, the big hack story,

00:57:44   which is why the reason I complain about it is it's,

00:57:47   it's so exceptional for never been repeated since then that they've done

00:57:51   something that boneheaded, right? It's, you know, if it were,

00:57:55   if that sort of fiasco were par for the course for Bloomberg,

00:58:00   nobody would listen to Bloomberg reports.

00:58:02   It's the fact that they're overall so credible,

00:58:05   but with the sources they have to me, they should be able to have dates.

00:58:09   And I would just love to know when these decisions were made, because to me,

00:58:13   it's one of the most fascinating things about Apple is how long their hardware

00:58:18   lead times are and, and how secretive it is for whatever

00:58:23   reason. One of the holiest of Holy secrets that they have,

00:58:27   they do not talk about these timelines of how,

00:58:31   how far in advance and they, they don't want, they don't want to disabuse,

00:58:36   the notion that they can make last minute decisions, even though I know they can't,

00:58:40   everybody who's involved knows they can't. Right. I remember,

00:58:44   I forget his name, but back when it was one of the first,

00:58:49   truly, wow, I'm, I'm, this is unbelievable.

00:58:52   I can't believe that I'm one of the media people here for the

00:58:57   antenna gate.

00:58:58   I went to the West when you went to the anechoic room and that whole thing,

00:59:01   right? Yeah. Yeah. Yeah. And their lead, you know, it was, I don't know,

00:59:06   a hundred people from the media were invited on the, you know,

00:59:09   a day in advance to go to the press conference, you know,

00:59:12   and Steve jobs was there fresh off, you know,

00:59:15   curtailed vacation in Hawaii and they played the Jonathan Mann story.

00:59:20   If you don't like it, take it back or something like that. Remember that?

00:59:23   Yeah. It was great. It was one of his great, his viral moments.

00:59:27   It's the press conference itself was by far and away the most interesting press

00:59:31   event I've ever attended at Apple. But then I was part of,

00:59:35   maybe there were like a dozen of us who got like this backstage tour of their

00:59:40   anechoic chambers and to see some of these places where they've tested their

00:59:44   antenna designs. And it was about a dozen of us in the media.

00:59:49   The tour was led by their lead cellular antenna

00:59:54   engineer. I don't remember his name off the top of my head,

00:59:57   but, but he was doing the talking,

00:59:58   but with us were Phil Schiller and Katie cotton

01:00:03   as, as our escort, but nobody else from Apple.

01:00:06   It was the only three people from Apple, this engineer,

01:00:09   Phil and Katie. And at one point, you know,

01:00:14   and they were encouraging questions and we'd seen these things.

01:00:17   And after one of the chambers, you know,

01:00:19   there were smaller ones and bigger ones that super freaky. If you've, you know,

01:00:23   again,

01:00:24   a Ruben Caballero or Caballero senior that's about his name. I remember his name.

01:00:29   You'd hear, I mean,

01:00:29   cause those people's names are rarely uttered in public also with their titles.

01:00:33   So yes.

01:00:34   Yep. Ruben Caballero. That was him. And he was charming and delightful. And,

01:00:38   and, but you could tell he wasn't practiced at this because,

01:00:42   because up until that point,

01:00:44   his work had been utterly secret and they're like, okay, Ruben, we need,

01:00:48   we're going to invite a dozen people to the media. Can you explain what you do?

01:00:54   And you know, if you've never been in one of these anechoic chambers,

01:00:57   it is super freaky. It, it, it, you just can't believe it.

01:01:00   And some of them were very big, you know, where, where, you know,

01:01:04   five or six of us could go in at the same time, you know,

01:01:06   and have plenty of room,

01:01:07   but you would open your mouth and talk to one another and you could hardly hear

01:01:11   each other. Oh, it's, it's so freaky.

01:01:14   You're just in this room with these foam walls in this weird sort of pyramid

01:01:18   shape and you're,

01:01:20   you're two or three feet away from anybody else and you'd talk at a normal

01:01:24   volume and the sound from your voice just disappears.

01:01:28   It was super freaky, very interesting, all sorts of interesting information.

01:01:32   But one of the people in the press asked something about like, well,

01:01:35   when did you start working on this external antenna design?

01:01:40   Cause that was the whole thing with the iPhone four.

01:01:42   And Ruben Caballero, he started to answer,

01:01:47   and I think he said something to the effect of like two years.

01:01:49   And as soon as he started this, it was something like that.

01:01:52   I have it in my notebook somewhere,

01:01:54   but Phil Schiller jumped in immediately and was like,

01:01:58   we're not going to talk about timelines. And it wasn't he, you know,

01:02:01   and Phil was Phil and it was very smooth, but I noticed it.

01:02:05   And he was just like, we're not going to talk about timelines. And,

01:02:08   and Ruben immediately, like I could see it on his face. He was like, Oh yeah,

01:02:13   I wasn't supposed to talk about timelines. And it wasn't the word, you know,

01:02:16   he didn't spill state secrets all, you know, the,

01:02:20   the gist of what he gave away was that they'd started working on the iPhone

01:02:24   four's external antenna design around when the iPhone three came out, right?

01:02:32   Three. And then a year later was three GS and two years later was the iPhone

01:02:37   four.

01:02:37   So it was like one year after the original iPhone was when they were like looking

01:02:44   at down the road, where do we want to be two years ago?

01:02:47   What if we do this flat back flat front flat sides and we'll move the antenna

01:02:52   outside the frame. And that way we'll solve this whole,

01:02:57   because if you think back to the first three iPhones,

01:03:00   the first iPhone had this gorgeous aluminum back,

01:03:03   but then to fit the antennas,

01:03:05   the bottom inch or so on the back was black plastic.

01:03:08   And it was sort of a weird transition, still a beautiful device,

01:03:12   but they had this problem where all the antennas had to be behind plastic,

01:03:17   not the aluminum. And then to solve that, you know, to, to work around that,

01:03:21   the three, the three G and the three GS had all plastic backs,

01:03:26   which were more antenna friendly, but not as nice a device.

01:03:31   And so, you know, two years,

01:03:34   but Phil Schiller did not want people to know that re it was very,

01:03:38   very clear again, wasn't uncomfortable. It wasn't like,

01:03:41   I'm sure Ruben did not get reprimanded afterwards or was in trouble.

01:03:46   Did not want that to come out.

01:03:48   I'm so curious about that with these Silicon issues. I am so interested.

01:03:53   Yeah, it's, I think, yeah, I mean, that's the looking back thing.

01:03:58   And I think, I mean, the article,

01:04:00   the information article is notable too because it's trying to tie together.

01:04:04   Wayne Ma is trying to tie together. And again, I think the timeline would help,

01:04:07   but I, I, maybe he just can't because it would reveal sources too. Right.

01:04:11   You know, right. At the time, but he's, he's trying to tie together, you know,

01:04:14   the,

01:04:14   the leadership of this division within Apple is being painted with the same kind

01:04:18   of it's the same kind of discussion usually here reserved for female company

01:04:23   leaders and politicians who are difficult to deal with. Like Amy Klobuchar,

01:04:27   she was getting covered.

01:04:28   She made her staff wash a comb and yelled at him because he didn't bring a fork

01:04:32   on the plane is the famous one. Right. And I'm like,

01:04:34   so this fellow Johnny's Ruggie is the Apple exec who's documented here as the

01:04:38   head of the that's the semi division.

01:04:40   So make an actor division and it's like, well,

01:04:42   he's really brusque and this and he won't hire people. And I'm like, yeah, it's,

01:04:46   I mean, I would hate to be, I,

01:04:48   I'm just trying to imagine how you hire people in one of the most rarefied

01:04:51   industries for the best company making Silicon in the world.

01:04:56   So how much of this is,

01:04:58   this is a really difficult space and the guy's a little prickly and he's hiring,

01:05:03   you know,

01:05:03   it's just a very difficult place to hire into and to keep people because they're

01:05:07   extremely valuable. There's a lot of money flowing around.

01:05:10   And they're talking about on this Exodus of Apple employees and it's like, well,

01:05:13   maybe he's not the best boss. I don't know. And maybe it could be,

01:05:16   that might be one factor.

01:05:18   Maybe people aren't given enough opportunities for advancement or whatever,

01:05:21   but it's also,

01:05:22   you got to think about the salaries that are being offered to people to jump

01:05:25   ship to these startups because there's so many, I mean,

01:05:28   there's so many billions and billions of dollars,

01:05:30   trillions ultimately involved in these spaces. So, and you know,

01:05:34   in acquisitions too is the next company, some of these spinoffs that,

01:05:37   where everyone's mad that they Apple people went off to join these or start

01:05:41   these other companies or join them, they could be the next Apple acquisition for,

01:05:45   you know, half a billion dollars or a billion dollars or more.

01:05:47   And we don't know that either. So it's kind of a, again,

01:05:50   I think the reporting seems solid, but I'm kind of like, well, maybe is it,

01:05:54   you know,

01:05:55   one former junior Apple employee was told he needed to get Srouji's permission

01:05:58   before he could participate in a hackathon.

01:06:00   So he abandoned the request given the effort approval would take. And I'm like,

01:06:03   well, that okay. But that doesn't seem that unreasonable.

01:06:06   You're working in one of the most secretive aspects.

01:06:08   So maybe you'd reveal a technique honestly,

01:06:10   that you learned that they don't want to know. So I don't know.

01:06:13   There's a little bit of that in there where I'm like, this is,

01:06:15   this is kind of inside baseball management stuff and,

01:06:19   and, and whatever. But if it's harming that division, it's an issue.

01:06:23   I suppose shareholders might be interested or, or what have you,

01:06:26   and they need to figure out how to retain and encourage talent.

01:06:29   I think the other part of the article though,

01:06:31   that I think I'm more so than the drama part is the,

01:06:34   I mentioned smile inches in passing. It's this thing about, you know,

01:06:37   it's kind of like the end of time coming, right?

01:06:39   Wait Moore's law has been remarkably accurate.

01:06:42   And it's the fact that it was so vaguely formulated. It's not perfect line,

01:06:46   but it's pretty good. And we're hitting the end, you know,

01:06:48   they're going down to three nanometer processes and then two nanometers.

01:06:52   And there's this new nano sheet technology to keep current from leaking across

01:06:57   transistors and so forth. But it's sort of, there's just a lot of noise and that's,

01:07:00   you know, 20, 25, I guess is when we're going to see maybe two nanometer process

01:07:04   chips. So we're getting down to the smaller and smaller size with more and more

01:07:07   limits that are just, you know,

01:07:09   you started getting down to quantum effects and all of this stuff.

01:07:12   There's a point at which they won't be able to make smaller chips with the

01:07:15   current like approach that has essentially been unbroken with modifications and

01:07:20   compensations for the history of the Silicon industry. So what happens then?

01:07:26   And that's not that many years away. So there is an issue.

01:07:29   You were talking about reversion of the mean, but then there's also like,

01:07:32   what is the next thing? It's all in the lab, I'm sure.

01:07:35   And research universities and R and D deep within sealed vaults inside Intel and

01:07:40   arm and others. But what happens after 2025? What's the next big breakthrough?

01:07:47   And I don't, I don't know what that is.

01:07:48   I'm not sure there is one that's going to have the same kind of gains that have

01:07:51   been happening consistently for this many years.

01:07:54   Yeah. I, it's interesting to see where that goes. The one,

01:07:58   the one area where I do see it coming,

01:08:00   I don't have any guesses as to how high end compute for what we consider personal

01:08:05   computers continues going forward and personal computers,

01:08:13   meaning Macs, iPads, iPhones,

01:08:16   things that have a user interface and run apps. And you,

01:08:20   you can kind of think of I'm using this as a computer, right?

01:08:24   When you use your iPhone, maybe most people don't think of it as a computer,

01:08:29   but you and I know, and the people who listen to the show know it's a computer.

01:08:32   How does that keep going forward in the face of these physical limitations at the

01:08:37   atomic literally getting close to the atomic level of, of, you know,

01:08:41   how the processes can continue improving? I don't know.

01:08:45   I don't know enough about it,

01:08:46   but the one area where I can see that is continuing to advance the,

01:08:51   continuing to advance at a remark at the same remarkable clip that computers

01:08:56   have kept getting better and more exciting my entire life, right?

01:09:03   Ever ever since I first realized that a computer was a thing and became

01:09:06   infatuated by the idea in the seventies as a little kid with video games is the

01:09:12   miniaturization and the computerization of everything.

01:09:20   AirPods would be my primary example where you literally could not make AirPods

01:09:25   10 years ago. There's the,

01:09:28   the Silicon did not exist because they run as they're too little computers.

01:09:33   There's a computer in your left ear and a computer in your right ear.

01:09:36   And the Silicon didn't exist 10 years ago.

01:09:41   It really wasn't possible. I mean, maybe in theory, 10 years ago,

01:09:45   but certainly 15 years ago,

01:09:47   there was no way to make anything like an AirPod because the chips didn't exist.

01:09:51   And that's the direction. And again,

01:09:55   I'm such a fan of my AirPod pros. I just knew it's one of the rare devices that

01:10:00   I continue not to take for granted. And I just keep thinking, my God,

01:10:04   these things are astonishingly amazing and I can hardly even feel them in my ear.

01:10:09   They're so lightweight. But obviously all this stuff,

01:10:14   everybody's thinking about AR glasses and you know, beyond VR goggles,

01:10:19   but to get to the point where you could just wear regular eyeglasses that look

01:10:23   like normal eyeglasses and you keep them on all day and they're not heavy,

01:10:27   they're just as comfortable as regular glasses,

01:10:30   but they actually are computers and you know,

01:10:33   projecting some kind of heads up display in front of you at all times and

01:10:37   listening to you. Obviously having AirPod style,

01:10:42   super lightweight chips with very long battery life is part of that.

01:10:47   And who knows how many untold,

01:10:51   I was going to say dozens, but hundreds of other type areas,

01:10:55   tiny little chips like that can come from.

01:10:57   And Apple is obviously very interested in that sort of direction for the future

01:11:02   for hardware, you know, and their Silicon team is key, right?

01:11:06   The Apple's Silicon prowess is so central to the AirPod success.

01:11:12   Story and will be central to their hopeful success in areas like AR and VR and

01:11:17   who knows, you know, what else? Or Apple watch is another great example, right?

01:11:24   Oh yeah. Goodness. Right. Yeah.

01:11:26   I'd love you to go back and claim chowder what people said when Apple bought PA

01:11:29   semi. I think you've done it before. It is, it is. People were like,

01:11:32   why would Apple get into Silicon? How can they compete with that? You know,

01:11:35   and you know, that's,

01:11:37   that was the beginning of the end for Intel's ability to lead the industry.

01:11:41   Although maybe it's,

01:11:42   there's some talk about maybe where they're at now is a better place,

01:11:45   but it just seemed impossible at a computer manufacturer,

01:11:48   especially one at Apple size at that time,

01:11:50   despite them already throwing off tens of billions of dollars in cash a year,

01:11:54   they were still relatively small.

01:11:56   And the fact that they've been able to custom engineer chips that have produced

01:12:01   those kinds of gains, but, and I think it's like this, it's to several points,

01:12:05   right?

01:12:06   Not only do they have,

01:12:07   were they able to make a better mobile chip and then,

01:12:10   and then integrate system on a chip design that makes for even better efficiency

01:12:15   and not only are they able to create the M one and produce a chip that is that,

01:12:20   I mean, it is hilarious.

01:12:21   Your site's one of the better places to find these comparisons,

01:12:24   but to see the people who are in the non Apple world,

01:12:26   try to justify that the M one or two is not as good as it claims by looking at

01:12:31   windows, laptops and Linux, whatever. It's just so funny. You're like, no,

01:12:34   I'm sorry, you're going to have to change your worldview. I was, I never,

01:12:37   when the M one came out, I really didn't believe it.

01:12:40   I actually had to own a machine before I truly accepted what a change it was.

01:12:44   And I'm like all in an Apple. I've been using Apple products since 1985. I'm not,

01:12:48   I'm not even dubious. And I was telling her, okay, this,

01:12:51   this is more of a game changer than I thought. But so it's not only that,

01:12:54   but it's also that they can produce these micro chips that, you know,

01:12:58   these tiny microchips, I should say, that allow them to do these, you know,

01:13:02   to do the air pods. And that's what the glasses, if Apple releases glasses,

01:13:06   you can imagine it's going to have the I've style thing of like,

01:13:09   we've packed in the battery into the stems and like the, you know,

01:13:13   all the custom circuitry. So you're going to wear a pair of maybe slightly chunky,

01:13:16   but very stylish, a black, you know, black glasses. And then the frame is going to

01:13:21   be a hundred percent full of electronics or something, right.

01:13:24   With induction charging in the stems or whatever.

01:13:27   Yeah. I'm,

01:13:29   I'm guessing that they're just going to stuff little tiny thin batteries all

01:13:33   throughout the glass.

01:13:34   Absolutely. Every, every corner, every edge. And then the,

01:13:38   the lenses I assume will be multi-lens or multi-layer LCD panels

01:13:43   interlaced with other kinds of probably more exotic stuff that get managed by

01:13:48   the microcomputers and the multiple computers probably running it. But you know,

01:13:52   then you're like, well, where do you put the processor? It's like, well,

01:13:55   they'll have designed some new chip form factor probably to stick the package in

01:13:58   the right spot.

01:13:59   With like a regular pair of eyeglasses.

01:14:01   It's like if you sit on your glasses or something and break the arm off,

01:14:05   you know, and it's like, oh,

01:14:07   you go back to your eyeglass place and you pay a nominal fee. It's like,

01:14:10   I'm sure with the Apple glasses, it might look like a normal pair of glasses,

01:14:13   but it'll be like,

01:14:14   it's like when you break the back of your iPhone and it's like how many

01:14:18   hundred dollars? And it's like, if you just break the side off your Apple

01:14:21   glasses, it'll, you know, it'll be like, oh, $250 just for the arm. It's like,

01:14:26   oh yeah, well there's all sorts of stuff in there. Apple's been prepping us with

01:14:29   the changes to Apple care though, where they're like, oh,

01:14:31   well now we're including an incident for this much. Oh, well accidents are

01:14:35   included. I'm like, yeah, yeah, you're gonna have to,

01:14:37   Apple glasses are going to have to have a very generous Apple care plan,

01:14:40   at least the early days to get people to go in on it. Yeah. Anyway,

01:14:44   before we leave the subject, that the thing,

01:14:45   one of the things I'm looking forward to 2023 is to me,

01:14:50   Apple is very clearly a company of annual patterns and we

01:14:55   still don't know what their pattern is for Apple Silicon max, right?

01:15:00   It's we've only gotten a second generation device.

01:15:04   The only ones that we have are the Mac book air and the

01:15:08   low end non pro 13 inch Mac book pro,

01:15:13   you know, the Mac book air in a, in a thicker chassis.

01:15:16   What is the pattern for the rest of Apple Silicon?

01:15:19   Because the one device we know is on an annual schedule is the iPhone, right?

01:15:23   Every year, every September, there's a new iPhone and a new generation of a,

01:15:27   a series chips, but everything else,

01:15:30   including the iPad is seemingly more like an 18 month schedule, right?

01:15:35   So iPads, all iPads, none of them get update,

01:15:39   updated annually, even though they've been on Apple Silicon forever.

01:15:43   And it just, it,

01:15:45   the thing that makes it seem unusual to me is that they're going

01:15:51   with this numbered M one M two integer pattern.

01:15:55   But I think that there will be max that skip generations,

01:16:01   right? That maybe, you know,

01:16:04   there won't be a Mac book pro that gets every single M

01:16:09   generation, you know, that at some point it's going to skip a generation.

01:16:13   Who knows? And obviously the Mac pro obviously skipped the M one

01:16:18   generation and

01:16:20   well, think about what you said earlier too, is that you don't need a Mac pro.

01:16:23   You've got a Mac book pro, right? This is your, and I,

01:16:26   I asked this question to people from time to time is, is what is the end of

01:16:30   upgrades, right? It's like my, I have an, uh, M one,

01:16:34   like first generation Mac book air with 16 gigs of Ram.

01:16:38   And I only have a half a gigabyte, sorry, half a terabyte rather.

01:16:42   I think a terabyte of storage on it. And I wish it just seemed like too much

01:16:47   money at the time to get a terabyte, but between iCloud and other things,

01:16:50   I make it work. So the only thing I'm short on is storage.

01:16:53   The only thing that would probably get me to upgrade,

01:16:55   I don't know when I will stop using my M one Mac book air because I don't,

01:17:00   I actually, for the first time in the history of owning laptops,

01:17:03   I don't need a faster one. Everything I need to do, I can do it the speed.

01:17:07   I need to do it more or less.

01:17:08   And I've done things like layout or work with giant InDesign files on it where

01:17:13   it's been essentially full speed, like real time, very little lag.

01:17:17   And that was even before Apple had fully transitioned to native,

01:17:20   native Apple Silicon code. So this is my question is like,

01:17:24   maybe Apple doesn't have to be on the same kind of cycle for many of the

01:17:28   machines because it's got, it's not terminal models, but the models are,

01:17:32   you know, what if they don't, if it's good enough for a couple of years,

01:17:36   and maybe people, I mean,

01:17:38   I don't think Apple wants people to own their computers necessarily for five to

01:17:41   seven years.

01:17:42   But you look at the history of how of ownership and kind of the support number

01:17:46   support years Apple's put it in Mac OS.

01:17:48   And it's like maybe we are going to see that lengthening that reduces their

01:17:52   revenue from max, except that they're expanding the Mac audience all the time.

01:17:56   And they still have, you know,

01:17:58   a hundred million people who need to upgrade to M machines who haven't yet.

01:18:02   So it feels like they could probably not worry as much about the speed of

01:18:07   revision of max and just keep focusing on why you should get the latest one.

01:18:11   If you have a machine that's, you know, three to seven years out of cycle.

01:18:15   Yeah, I think so too.

01:18:16   And I think they're very comfortable with the idea that PCs,

01:18:19   when the Mac is their PC as a, as a, uh,

01:18:24   you know,

01:18:25   literally 40 year old category that that a four or five year

01:18:29   upgrade cycle is fine, especially since they've,

01:18:32   they've held the line on prices, right? Where the,

01:18:36   the low end Mac book air is still nine 99,

01:18:40   even as the rest of the industry is sort of

01:18:44   filtered down to these three and $400 low end laptops,

01:18:49   Apple holding the line at a thousand dollars and selling most of these, you know,

01:18:53   the, the 14 and 16 inch Mac book pros start at $2,000

01:18:58   effectively. And really the, the configurations most people are buying,

01:19:02   I think are closer to $3,000,

01:19:03   which in today's PC world is very expensive, you know? And, and I think that

01:19:09   they're just fine selling.

01:19:10   They're very happy to sell you a $2,800 Mac book pro that you won't,

01:19:14   that you won't even think about upgrading for four years, you know,

01:19:17   with a 30 something percent margin. So they're like, that's fine.

01:19:20   We paid our money off you. And then, and this is the growth of this.

01:19:23   This is what I get. You got to bring this up, right?

01:19:25   We forget to do this sometimes that Apple is a services company now.

01:19:28   And of course they care less.

01:19:30   They have shifted enough people over to buying from the Mac app store. And,

01:19:34   you know, I'm ponying out that whatever the Apple one subscription is, I don't,

01:19:37   I don't know what it costs because it's less than it costs before for me to

01:19:40   buy all those things separately. And my family gets everything.

01:19:43   Some on family sharing. So they they're getting, I don't know,

01:19:46   $400 a year for me for Apple one. And I'm super happy about it.

01:19:51   I don't have any complaint about the price for the services provided. Yeah,

01:19:55   I agree. I do. I do agree. All right,

01:19:57   let me take a break here and thank our next sponsor.

01:19:58   It's our good friends at collide K O L I D E.

01:20:03   If you are listening to this podcast,

01:20:06   the odds are very good that at some point you or your company are going to

01:20:10   go through an audit like SOC two or ISO 27,

01:20:14   zero zero one. And when you do,

01:20:16   you'll have to answer some tough questions about end point security questions

01:20:20   like do all your company laptops have their discs encrypted?

01:20:25   Does everyone have the company's password manager installed?

01:20:28   Do you have a system in place to monitor and maintain compliance throughout your

01:20:33   cross platform fleet of PCs? If,

01:20:37   even if you're confident that you can answer all those questions. Yes.

01:20:40   The bigger question when you're faced with an audit is can you prove it?

01:20:45   If you're not quite sure how you'd go about proving compliance,

01:20:49   you need collide.

01:20:50   Collide is an end point security tool for Mac windows and now Linux devices that

01:20:55   does things that MDMs can't and gives you the visibility you need to achieve and

01:21:01   maintain and prove compliance. Best of all,

01:21:05   collide does not resort to surveilling employees or locking down devices.

01:21:09   Instead it works with end users to resolve issues and relies on their cooperation

01:21:14   and informed consent.

01:21:16   You can meet your security goals and pass your audit without compromising on

01:21:21   privacy.

01:21:22   Visit collide.com/the talk show to find out how.

01:21:27   Follow that link.

01:21:28   They'll hook you up with a goodie bag that includes a t-shirt just for

01:21:31   activating a free trial.

01:21:32   That's K O L I D E.com collide.com/the talk show.

01:21:37   What else do we got? We've got, uh, I want to talk,

01:21:43   speaking of password security. Oh man.

01:21:45   Last pass has had,

01:21:48   I would say a week that would make Southwest airlines seem like they've had a

01:21:53   good, a good week.

01:21:54   Yeah. Holy cow. You know, it's funny.

01:21:58   There's one of these things of like where you list off, well,

01:22:00   what password managers do people use? And it's like, well,

01:22:02   you'll be like, I always one password, you know,

01:22:04   there's a built in one in iCloud key chain.

01:22:07   You can use that and all of Apple products, but it's not, you know,

01:22:09   it's not good outside the ecosystem. Like, well there's, you know, last pass,

01:22:12   there's dash lane, there's, you start listing off others, right?

01:22:14   But last pass has always been kind of in that list and I don't know that it's

01:22:18   going to be in that list anymore. Like I don't think,

01:22:20   I don't think it should be. And I think there's a lot of reasons why it,

01:22:23   it's not because they were hacked. That's bad.

01:22:26   They made a number of operational mistakes that they've admitted some of clearly.

01:22:30   But the problem is I think, I think it is embarrassing to have master password

01:22:35   based security as your only line of defense in 2022 or maybe in 2015.

01:22:41   Right. And this is what's come out.

01:22:44   And I didn't realize this cause I haven't used last pass and I haven't looked at

01:22:47   their security for a while when I was,

01:22:49   since I was doing kind of more technical encryption looks and last pass,

01:22:54   you know, so every other, so I shouldn't say every other,

01:22:57   all the major password and secret management services out there you're likely to

01:23:02   use. And I will cite again, Apple's iCloud,

01:23:04   key chain operation and one password in particular because I know those more,

01:23:08   more in depth. I know there are others like this that are more specialized.

01:23:11   People call it bit warden,

01:23:13   which is an open source project that has not needs more apparently outside

01:23:17   auditing. They need to raise money to do certain things for more assurance.

01:23:20   But again, it's an option that's designed around this.

01:23:22   You need to have device based key material mix into anything so that if you're

01:23:27   archive is in a cloud storage environment and it's compromised that without

01:23:31   access to the devices and the ability to unlock a device.

01:23:35   So it's a physical proximity attack typically unless you're a sophisticated

01:23:39   nation state actor who could remotely install expensive one time use zero days

01:23:44   to crack your phone and extract device keys,

01:23:48   which isn't even possible in a secure enclave environment in most circumstances.

01:23:53   Right. Blah, blah, blah.

01:23:54   It's long list of things that you're still secure, right?

01:23:56   So if you're one password vault, if one password had a massive,

01:23:59   massive attack,

01:24:00   some horrible thing happened and all of their cloud stored vaults that they sync

01:24:04   for users and give access to through web apps were stolen.

01:24:07   There is zero risk that anyone's material would ever be decrypted because you

01:24:11   must have the secret keys that are device stored. Right. And that's that.

01:24:15   If all of your,

01:24:16   if Apple had an iCloud breach and all of its end to end security information,

01:24:20   including iCloud key chains were stolen and somebody had a billion years to work

01:24:24   on it,

01:24:25   they're not going to decrypt it because you need the keys stored only in secure

01:24:28   enclave and all of your Apple devices. It's absolutely, you know,

01:24:31   unless there was some exploit we don't know about and the design would be very

01:24:35   difficult to deal with with last pass. If somebody steals your last pass vault,

01:24:39   which bazillion were just lost and can guess through brute force,

01:24:43   your your master password, it's totally decrypted and that's it. Boom.

01:24:45   That's the story.

01:24:47   Yeah. That's a great summary.

01:24:48   And the problem with that is that the,

01:24:54   I think, and I read a lot about this this week,

01:24:57   but I think the problem is that with, with all of these vaults, I mean,

01:25:02   I guess millions, I don't, I'm not quite sure how many customers they have,

01:25:05   but you know, all of them, you know, the vaults are lost or,

01:25:09   or have been taken,

01:25:10   but they are secured by the user's own master password.

01:25:15   Right. So it is computationally unfeasible for anybody in possession of all of

01:25:24   those vaults to decrypt, certainly all of them,

01:25:27   but even to mass attack all of these stored vaults.

01:25:32   But if they wanted to target specific users,

01:25:36   it's computationally feasible to,

01:25:40   and especially, you know,

01:25:43   if some of these users don't have super long, super secure master passwords,

01:25:48   you know, it doesn't,

01:25:49   or if the master password was secured by the previous, this whole thing, you know,

01:25:53   some listeners will get this and some won't, and it's a deep cryptographic thing,

01:25:57   but it's like the number of rounds of hashing you do on a, on a password.

01:26:00   So you type in your password, you type in John Gruber rocks, right? And it gets,

01:26:05   that's not a good password. Don't use it. Folks has to be longer,

01:26:07   has to be longer and random.

01:26:08   But I do appreciate the sentiment. So if that was your password, I do.

01:26:12   I think I thank you for selecting it, but go change it.

01:26:15   So you type that in and then the hashing algorithm, the way hash works,

01:26:21   it's a one way cryptographic function.

01:26:23   So you put in one thing and the thing that comes up the other end is essentially

01:26:27   deterministically manipulated,

01:26:29   but you can't reverse it without an enormous asymmetrical amount of computational

01:26:33   power. So you do that once. That can be really strong.

01:26:35   That's how Shaw two and these other things use for certain kinds of pastures

01:26:39   storage work. But you could also, if you want to increase the difficulty,

01:26:42   you can use something, use an iterative routine.

01:26:44   So instead of just hashing it once you hash it, you know,

01:26:47   5,000 times, a hundred thousand times, 200,000 times, and each iterative,

01:26:51   I should say the shot to algorithm includes that as part of it.

01:26:54   The number of hashing operations that occur, but it's part of an algorithm,

01:26:57   right? So one password has this approach. Last pass,

01:27:00   others have their approaches and you can wind up if you haven't updated,

01:27:05   people have been finding as they look through their settings at last pass that

01:27:09   they are set to, you know, 5,000 iterations where last pass now,

01:27:13   I think the current setting is like a hundred thousand and 100 iterations and

01:27:19   5,000 is much easier to crack than the a hundred thousand.

01:27:23   And you can set it higher even. So there's,

01:27:24   there's the master password length and complexity. There's the,

01:27:28   so the amount of entropy in it, right? So how hard is it to crack?

01:27:31   There's the sort of social engineering guessability. And then there's the,

01:27:35   the, the, even if you have a really good password, your master password,

01:27:39   if it hasn't been iterated a sufficient number of times,

01:27:42   it can also be trivial. And John, you saw this, I didn't know this.

01:27:47   They don't encrypt the URLs in your password storage.

01:27:51   Yeah, that's that is going to leak. So if you're asking, so you say that,

01:27:56   what you said is valid. It's, they have some information,

01:27:59   an account information leaked.

01:28:01   So it's possible that a cracker would be able to match up a user with a vault

01:28:05   either through some leaked metadata about the account or failing that through

01:28:10   looking at URLs because some of those URLs are going to contain encoded

01:28:13   information about your user ID at an account because the URL is improperly

01:28:18   anonymized in the way that any given site,

01:28:21   you have a thousand passwords and some sites will encode your username into it

01:28:25   conceivably or maybe your email address. So in a different circumstance,

01:28:29   if everything in the vault were encrypted and no metadata were available in an

01:28:33   unencrypted form, then the vaults might still be fairly useless to other people,

01:28:38   even the ones that are weakly protected because it might be impossible for a

01:28:41   cracker to know which ones were weakly protected.

01:28:45   So they'd have to go through a lot of work even to figure out which ones they

01:28:48   could start to crack. But that's not the case.

01:28:51   Anyway, bad news for LastPass. I, I,

01:28:56   I don't know about you. I, you know, in terms of what I use,

01:29:00   I honestly have mostly overwhelmingly,

01:29:04   I would say rely on iCloud key chain and I,

01:29:09   I've trusted it for numerous years and it over one of the things I did last year

01:29:14   in 2022 was I moved most,

01:29:19   almost all of my two factor authentication codes. Like when you get those, Oh,

01:29:24   as a second factor,

01:29:26   go to Google authenticator or authy or any number of,

01:29:30   you know, there's an open protocol for generating these things.

01:29:33   And then once a minute you get a new six digit code and you,

01:29:39   you need to enter the current minutes,

01:29:42   six digit code for this account as your second factor to get in.

01:29:46   I moved most of those to iCloud key chain,

01:29:49   which started supporting those over a year ago. And

01:29:53   I couldn't be happier.

01:29:55   And I know that there's a lot of people out there who don't,

01:29:58   a one reason not to go all in with I call key chain is if you have cross

01:30:04   platform work needs.

01:30:05   If you use windows regularly or you want to use Android regularly or need to,

01:30:10   or just want to maintain your ability to,

01:30:18   to do that in the future. And you know,

01:30:20   you obviously can back out of iCloud key chain and switch from iCloud key chain

01:30:25   to one password or something else that's cross platform in the future.

01:30:29   But if you've already,

01:30:32   if you're already have all your two factor codes in authy or Google authenticator

01:30:37   and want to keep them there,

01:30:39   not because you think it's the best interface for them or the most convenient,

01:30:42   but because you like the idea that that's a cross platform way to do it.

01:30:46   I totally get that. But me personally,

01:30:50   I'm so all in on the Apple ecosystem,

01:30:53   including Safari as my main browser,

01:30:57   that the integration with iCloud key chain is just phenomenal.

01:31:02   And I know that there's a lot of people who hear Apple and cloud based syncing

01:31:07   and think it's always a disaster. And you know, I, I, I've been,

01:31:14   this is not my first rodeo. You know, I've,

01:31:17   I've seen Apple's struggles with cloud based syncing firsthand over a long

01:31:22   number of years.

01:31:23   I will say as somebody who's been at pretty much all in on iCloud key chain for

01:31:29   years, the syncing with iCloud key chain is as good as any syncing.

01:31:33   I've seen for any product anywhere. And I, you know, I,

01:31:38   I'm not personally vouching that they're never going to have a sync problem,

01:31:41   you know, that would be catastrophic for people.

01:31:44   And for my very most important accounts,

01:31:47   I've got saved codes printed out and stored in a safe location.

01:31:53   And you know, uh, uh,

01:31:57   and iCloud key chain catastrophe would be very, very annoying to me.

01:32:01   And there I might have a problem.

01:32:03   There might be some accounts that where I'm not really,

01:32:05   I haven't thought it through and I might be locked out and need,

01:32:09   I don't know what I would, you know,

01:32:10   but for the ones that I'm most concerned about, I know that I could recover,

01:32:15   even if I lost access to iCloud key chain with recovery codes, et cetera.

01:32:19   And I've never needed to.

01:32:21   And the integration with Safari and all the things that I've been doing,

01:32:26   the things I would want my password manager to do, I couldn't be happier.

01:32:31   And I, you know, I know a couple of the people who work on that team at Apple.

01:32:35   I think it is for lack of a better term, one of Apple's a teams,

01:32:39   I think they take it with the utmost care and it's,

01:32:45   you know, it's one of Apple's a teams working on that.

01:32:48   And it just keeps getting better. I don't know. What do you, what do you use?

01:32:51   I've been moving. I've used offie for, for my, what are they?

01:32:55   T O T P S time-based one time passwords, right?

01:32:59   And I've used that for years because that was the best Google authenticator was a

01:33:03   mess because it would, you know, you'd have a restore your phony,

01:33:05   it was all your code that was the worst, whatever. And so I,

01:33:08   at some point I found offie and a Twilio owns it now.

01:33:11   I don't know if they made it and it's a great, it's free for that purpose.

01:33:15   There's other stuff they sell,

01:33:16   but they don't market you and it sinks securely and it's available.

01:33:20   There's a Mac app that came out. It's not great, but it's fine.

01:33:23   So I used offie for years and years and then as soon as the, I guess,

01:33:27   Apple calls it, you know, verification codes or something.

01:33:29   When that rolled out with Monterey and iOS, iPad,

01:33:32   iOS 15 I was documenting it of course,

01:33:34   cause I read a book about iPad and iOS security and Mac security for the take

01:33:38   control series. And I thought, Oh, this is actually really, really well implemented.

01:33:43   It's best when you're using Safari on a Mac and it's best when you're doing kind

01:33:47   of anything in an app that's correctly integrated or in Safari on iOS and iPad

01:33:51   iOS devices. But you know, the cross platform thing,

01:33:54   you can just go on your iPhone, you just go to settings, passwords,

01:33:58   search on the thing and it shows you the verification code. So,

01:34:02   and I've had that issue. I mean,

01:34:03   occasionally there's some Safari integration issues where you can't edit the

01:34:07   list of websites to which a verification code applies.

01:34:11   So I've had some cases in which a site verifies on one site,

01:34:14   but then when you log in it has a different sub domain.

01:34:16   So occasionally I've had to go and be like, all right,

01:34:19   I've got to go to passwords in Mac OS and bring this up.

01:34:22   It doesn't give me the auto-fill touch ID option. I confess,

01:34:26   and I know I'm not the only person. I have a Mac mini and I bought a, uh,

01:34:30   a magic keyboard with touch ID almost well first to test cause I write about

01:34:36   these things. So I needed to own one and now I keep it attached all the time,

01:34:40   even though it's not my primary keyboard because I use touch ID on my Mac all

01:34:44   the time and it's such a delight compared to passwords or reducing security

01:34:49   levels that you know,

01:34:50   I'm using touch ID on my non laptop Mac for that purpose. I know Mac,

01:34:55   iMac users have had that option all along, but I specifically bought it.

01:34:59   But I think the feature works really well. I think it's,

01:35:01   I think the way it's presented, especially in like an in browser is good.

01:35:04   I have never, I've never had,

01:35:06   you said that and I'm thinking of all the iCloud sync problems I've had with

01:35:10   things like photos or calendar events,

01:35:12   things where I've had to go and do weird, you know,

01:35:14   resets and were have had no explanation why things were wrong for weeks.

01:35:18   I've never, that I can recall, have an iCloud key chain sync issue.

01:35:23   And I think it's got a different pathway.

01:35:25   So I'm sure the data is backed up at iCloud in this secured fashion.

01:35:29   But my belief is, I remember this is in the security document or not,

01:35:33   the iCloud security document.

01:35:34   I believe that it's kind of a push to your devices as opposed to like a sync

01:35:39   where it kind of accumulates it and then your devices kind of pole and whatever.

01:35:43   I think iCloud key chain items are essentially device to device synced with

01:35:48   iCloud as the conduit. And maybe describing that wrong,

01:35:51   but it does seem like it is more,

01:35:52   more rapidly done and more accurately done than any thing else I've ever had

01:35:56   trouble with on iCloud.

01:35:57   Yeah, I, I, that's, you know, everybody's experience is anecdotal, right?

01:36:02   But yeah, but some things to me sync better than others.

01:36:06   I also find that Apple notes has been the modern era of Apple notes ever since

01:36:11   they moved past I map as the backend sinking,

01:36:16   which was terrible to be honest. I mean,

01:36:18   I kind of understand why they did it because it was sort of a shortcut to get

01:36:22   sinking of some sort. But when,

01:36:25   when they switched to a native iCloud set of APIs for sinking your notes,

01:36:30   it's been rock solid for me. I mean, really, really good.

01:36:36   I kind of get the feeling, you know, in a very obvious way,

01:36:39   the way that you would think, Hey,

01:36:40   I would kind of hope that the iCloud key chain team is extra careful and is

01:36:45   there, you know, puts an even higher priority on not making mistakes,

01:36:52   even at the expense of moving slower year over year. Right. So,

01:36:59   yeah, you know, it's fewer new features move slower, but instead of,

01:37:05   you know, take the old carpenters adage of measure twice, cut once.

01:37:08   Maybe with something like iCloud cloud key chain,

01:37:11   you measure 10 times and cut once.

01:37:13   I learned something amazing about Russia a few decades ago from a Russian friend

01:37:17   at the time. I said, measure twice, cut once. He said, Oh, he said, in Russia,

01:37:21   we say measure seven times and cut once. And I said, Oh wow.

01:37:24   He said everything is sevens. And I was like, that taught me a lot.

01:37:27   The other thing is, you know, this is the neat feature, right?

01:37:29   Is that because it's integrated and this is probably why you do this.

01:37:32   It's why I'm doing it now as well as I trust, I trust what Apple's doing.

01:37:36   I believe their security documentation.

01:37:37   I haven't had negative experiences and iCloud key chain has worked solidly for

01:37:41   me all these years for regular passwords.

01:37:44   I've been using one password for many years also, and I still,

01:37:46   I use them both in different, slightly different ways, kind of integrated.

01:37:50   But it's the,

01:37:51   it shows you the QR code when you're doing the two factor enrollment and you

01:37:55   hold down, you know,

01:37:57   you press and hold on it on a touch device or you right click on it in Safari.

01:38:01   And it says, add as verification code. And that's your setup. I mean, come on.

01:38:05   That's, that's the kind of frictionless integration you want. And you're like,

01:38:09   well, why wouldn't I do that? Because now I'm there. There's issues.

01:38:12   People worry about out of band issues as well. Like, is this protected enough?

01:38:16   The classic one that comes up that I get contacted by people about regularly and

01:38:21   I've documented in my books is that when you do two factor, I mean,

01:38:25   it's funny that Apple isn't, hasn't adopted TOTPs for its own site.

01:38:28   It's using its own proprietary but very high security system that is kind of a

01:38:33   two step authentication in order to get a number. Right.

01:38:36   But when you do a login with a regular username and password,

01:38:40   and then it gives you the prompt and it's like, Hey, prove who you are here.

01:38:45   You need to enter this TOTP. Is this your location? And then what your device,

01:38:48   one of your devices, you enter the code, right? All that's good.

01:38:50   You want to trust this device. People have said to me, well,

01:38:53   why should I trust entering a code into the same browser?

01:38:57   Like if I'm on my machine or that the machine is on,

01:38:59   so I'm on a browser on my Mac and the Mac produces the code that Apple sent.

01:39:04   And it's like, because it's not in the same band path,

01:39:06   I'm not on my browser and the browser is presenting me my code.

01:39:10   I'm on the browser and a separate Apple system that is device again,

01:39:15   device locked like this is your Apple ID account logged into that Mac through a

01:39:19   verified connection that you've already done two factor authentication on for

01:39:23   iCloud and Apple ID on your system.

01:39:26   That pathway is being used exclusively over secure transport to present that

01:39:31   code to you. So it's trustworthy because it's not in browser.

01:39:35   And that's the same thing for the verification code.

01:39:38   You're in Safari and your verification codes in Safari,

01:39:41   but there's not a pathway where a website can pull that verification code out as

01:39:45   essentially separate sandbox entities that don't talk to each other.

01:39:49   Yeah. You're the one who explained this to me a few years ago and I'll always

01:39:53   appreciate it. And I'm still not sure I entirely understand it to be honest,

01:39:57   but I trust it at this point,

01:39:58   but I'll bet everybody listening to the show has had family members say to them,

01:40:03   how, how does this make sense when I'm logging in? I'm on my, I'm on this,

01:40:08   I'm on, I'm on my Mac book and I'm logging into an Apple website.

01:40:12   And to confirm I am who I am,

01:40:15   it's the same Mac book saying, is this a login that you want to approve?

01:40:20   But it's, it is, I guess the other, the flip side of the coin is that it is the,

01:40:25   the sort of counterintuitive way that,

01:40:29   that Apple's privacy initiatives are often in conflict with security

01:40:34   initiatives. And where I'm going with that is that the browser is so private

01:40:40   that the browser doesn't tell the website, this is a known,

01:40:46   this is the device I'm on, right? This unique identifier,

01:40:50   for this Mac book. No,

01:40:52   that you don't want your browser to give a unique identifier for your hardware

01:40:57   to any website, whether it's Apple's or Google's. But so like you said,

01:41:01   it's out of band of the communication of your browser to apple.com.

01:41:06   But when Apple says, okay,

01:41:09   we'll send a ping to all of your known devices that we trust and that you've

01:41:13   vouched for,

01:41:14   which includes the Mac book you're on to verify that this is right,

01:41:18   then you can just do it right on your Mac book. And once you understand that,

01:41:21   that it's sort of an entirely different communication channel between apple,

01:41:26   you know, it's your browser is not your device,

01:41:31   but that confirmation is your device saying, yes,

01:41:34   this trusted device is saying that this is okay.

01:41:37   It's because the browser to, to regular people, right?

01:41:41   The browser and your operating system appear to be monolithic.

01:41:44   They're all right. Same thing. Apple doesn't trust the browser.

01:41:47   But it trusts your device. So it's weird when it's like, well,

01:41:50   if you don't trust my browser, why are you trusting this? It's like, ah,

01:41:53   they're really, the browser's untrustworthy. So you could be phished.

01:41:56   And the thing to remember is it's not the Safari or any other browser,

01:42:01   whatever, but it's not Safari prompting you with that map that says,

01:42:06   oh, there's a login from somewhere. That's the system doing it,

01:42:10   which is out of it.

01:42:11   This is with advanced data protection, right? The newest thing,

01:42:15   which I think is super cool.

01:42:16   I assumed when I saw the first press release about it, oh,

01:42:19   you won't be able to access your data on iCloud.com anymore because it's all

01:42:23   going to be ended and encrypted. But no, Apple will let you do it. I don't know.

01:42:28   I've turned over a test account. I turned ADP on it so I could test it.

01:42:31   And you go to the website, iCloud.com and says, Hey, all your data,

01:42:35   except your email contacts and calendar entries are protected by ADP.

01:42:39   Do you want to access it? And if you say yes,

01:42:42   it does push to one of your devices that lets you enter a key that they have

01:42:46   then allows it to decrypt temporarily within the browser,

01:42:51   the data that sent from iCloud. So the data is still encrypted at iCloud.

01:42:56   It's sent to your browser,

01:42:57   cached in your browser and decrypting your browser with that temporary key.

01:43:01   And that is very clever. That's what one password does. For instance,

01:43:04   with when you use one password.com to access your end to end encrypted stuff,

01:43:09   it's never receiving the password. All the password access,

01:43:13   all the password interaction is local to your browser cache.

01:43:16   So it's never transmitted. And so I thought, you know, so that I knew,

01:43:19   one password was doing that,

01:43:20   but so Apple has found that trustworthy enough apparently that they're willing

01:43:24   to let you do that. So if you enable ADP, which I think is, you know,

01:43:28   there's plus pros and cons,

01:43:30   but it's could be very good as just an extra level of privacy and security.

01:43:33   Then you are not excluded from using iCloud.com for data for those newly

01:43:38   protected categories, which is cool.

01:43:41   But counterintuitive.

01:43:43   Yeah. And, and you could also, I don't, did you notice this little,

01:43:45   there's like a little switch too.

01:43:47   You can also now if you choose whether or not you have ADP and turned on,

01:43:51   you can separately choose in, in iOS six, let's do all the numbers, right?

01:43:55   It's Mac, MacOS 13.4, 13.1 Ventura,

01:43:59   Iowa's 16.2 iPad was 16.2.

01:44:02   So you have an option for ADP in the U S or you're at any of those versions

01:44:06   worldwide.

01:44:07   You can now disable iCloud.com access for your account.

01:44:11   And that's separate from ADP.

01:44:14   Yeah. It's a little subtle thing. I noticed it. I was like, wait, what is this?

01:44:17   And it's so you can have the normal encryption where some of your stuff is

01:44:20   encrypted at rest and some end to end like it is for almost everybody today.

01:44:24   And you're like,

01:44:25   I don't want to expose a phishing risk or any other risk to iCloud.com so I'm

01:44:30   just going to turn that off. And if you turn it off, you can turn it back on.

01:44:33   But if you can only do that from devices, you can't do it from a website.

01:44:36   So it's a, it's a different security choice, but it's an interesting one.

01:44:40   Yeah. It doesn't disable any sinking either.

01:44:42   Yeah.

01:44:43   And the one thing I can think of that that would do is it puts your device

01:44:48   security in front of everything, right? So all of your Apple devices have,

01:44:52   you know, face ID or a passcode or your Mac has a password and touch ID,

01:44:57   but there's no way to get into anything without getting into the device first.

01:45:03   Whereas iCloud,

01:45:05   you can use a Apple ID and a password and then a phone number.

01:45:10   You can have use your backup because Apple is still letting you do backup SMS

01:45:14   and automated voice calls. So if somebody were to, I mean, it's, you know,

01:45:17   it's a lot of steps, but if someone were to hijack a SIM, for instance,

01:45:21   at a cell phone company,

01:45:22   then they'd still,

01:45:23   they might be able to crack your iCloud through iCloud.com access.

01:45:27   But with this method, someone could,

01:45:30   I mean, they could log into certain services without a device because they can

01:45:35   still use an SMS method, but it wouldn't provide them the same amount of control.

01:45:38   Yeah. I think that, you know,

01:45:40   for most people I certainly have,

01:45:41   I'm going to leave iCloud.com access for me on and I think for most people

01:45:46   that's right, but you can, you know,

01:45:47   it's not too hard to squint your eyes and see how it's more secure to just turn

01:45:51   that off if you're genuinely paranoid or if you actually suspect that you're

01:45:56   more likely to be attacked.

01:45:59   Apple's now doing this tiered thing. You've got lockdown mode at the top.

01:46:02   You've got the new iMessage feature that'll roll out this year that warns you of

01:46:06   potential that there are people have intercepted or they have sniffers in the

01:46:09   middle of an iMessage conversation.

01:46:11   There's iMessage out of band confirmation for identity coming in 2023.

01:46:15   That's the top. Then you have ADP below that.

01:46:18   And then below that is iCloud.com access on or off.

01:46:22   And below that is fundamental at rest device based security with some elements

01:46:27   for iCloud syncing being end to end encrypted besides those data parts.

01:46:30   So it's a good, good layer of cake that's been built now.

01:46:33   It still makes me laugh thinking back to the original iPhone and it was just

01:46:39   slide to unlock. Yes. Oh, wait, no password. No passcode, right?

01:46:44   Well, you could set a password, I believe, right? But it wasn't,

01:46:47   it wasn't part of the golden. This is the way,

01:46:50   this is the way we intend you to use the device.

01:46:53   And I certainly didn't have a passcode on my iPhone for a couple of years.

01:46:56   No, cause slide to unlock was way too cool. It was. And then,

01:47:01   but once you slid to unlock, you just had all my email, had all your messages.

01:47:05   Sure. All right.

01:47:07   Let me take a break here and thank our third and final sponsor of the day.

01:47:10   It is our good friends at Squarespace. Oh,

01:47:13   Squarespace is the all in one platform for building your brand and growing your

01:47:17   business online. Stand out with a beautiful website,

01:47:21   engage with your audience and sell anything, your products, your content,

01:47:25   even your time through Squarespace, they've got member areas.

01:47:29   They make it easy for creators to monetize their content and expertise in a way

01:47:33   that fits their brand with member areas.

01:47:35   You can unlock a new revenue stream for your business and free up time and your

01:47:39   schedule by selling access to gated content like videos,

01:47:43   online courses, or even newsletters. They've got built-in analytics.

01:47:48   You can see where people are going on your website, where they're coming from,

01:47:52   which site, which sections of your site are popular,

01:47:55   which sections are not popular. Just a terrific interface for that.

01:47:58   Online stores, you can sell anything and they've got it all.

01:48:02   Everything from setting up the images and the catalog of the items you want to

01:48:06   sell to actual processing of the credit cards and,

01:48:09   and all of that e-commerce stuff all built into Squarespace.

01:48:13   They've got it all. They've got custom templates,

01:48:16   get started with the best in class templates,

01:48:19   and you can customize them to fit your needs, change the fonts,

01:48:22   change the colors, change the graphics, or just use the templates as,

01:48:26   as is they are hundreds to choose from. It's just terrific.

01:48:30   Where do you go to find out more? Go to squarespace.com/talkshow.

01:48:35   squarespace.com/talkshow. You go there, you get a 30 day free trial,

01:48:41   no limits, no watermarks on the site as you build it.

01:48:44   And then when your free trial's up and you're ready to launch,

01:48:47   just use that same offer code talk show,

01:48:50   and you save 10% off your first choice first purchase,

01:48:54   including prepaying for up to an entire year.

01:48:58   Just go to squarespace.com/talkshow. Last on my list was,

01:49:03   well, let me talk about this.

01:49:05   I want to talk a little bit about Twitter and Mastodon because I know people are

01:49:09   tired of talking about Twitter and Musk. I'm tired of it too, but it's so,

01:49:12   it's fascinating. Here, parlaying off our security discussion.

01:49:18   So I've been a Twitter blue subscriber for most of the year.

01:49:22   I think it came out earlier in 2022 is a $3 a month product that I pay happily

01:49:27   paid for to get the top articles feature.

01:49:30   I like the being able to have 30 seconds to revert your thing before it went

01:49:36   out.

01:49:37   I forgot that that was, I did not realize that was a Twitter blue feature,

01:49:41   even though, so it's like,

01:49:43   I still do most of my Twitter use from tweet bot,

01:49:47   so I don't in the third party doesn't have it,

01:49:49   but I use the first party app enough that I, you know,

01:49:53   I've seen that feature editing tweets is only a Twitter blue feature.

01:49:58   The top articles feature is it's like nuzzle used to be,

01:50:01   but basically it's a very simple concept. And, and to me,

01:50:06   Twitter is at its best when its features are very easily explained.

01:50:11   But the basic idea is all the people who you follow you,

01:50:15   the people you've chosen to follow it. If,

01:50:18   if a number of the people you follow have all tweeted links to the same article

01:50:23   that becomes one of the top articles of the day.

01:50:27   And you can just go to this page or tab in the app and just see

01:50:31   the articles across the web that are the most linked

01:50:37   to by the people you choose to follow on Twitter.

01:50:40   And then they have a second tab,

01:50:41   which is sort of the second order version of that,

01:50:46   where the people who you follow,

01:50:49   who they follow their top articles.

01:50:52   And they had Twitter had that relationship with paywall publication. Yeah.

01:50:56   So many things you could read through Twitter blue. Yeah.

01:50:59   You would have needed a subscription. So yeah,

01:51:01   it was my favorite thing since nuzzle also is the closest thing.

01:51:03   And nuzzle to me was a little bit better at

01:51:07   populating the list and nuzzle.

01:51:12   One reason for me as a news junkie is that with nuzzle, you could set it to like,

01:51:16   three hours instead of 24 hours.

01:51:18   Twitter's version is a little bit simpler and it's just,

01:51:21   you don't get that kind of granularity,

01:51:23   but more than made up for the fact that Twitter,

01:51:27   unlike nuzzle has these partnership deals with a lot of websites where you can

01:51:32   burst through paywalls and read,

01:51:34   read these articles without being a subscriber. Great feature.

01:51:38   I went through this thing. It seems like this is,

01:51:41   I don't know if it's Apple's fault or Twitter's fault, but I,

01:51:44   I paid for it through the iOS app.

01:51:46   So it was going through my iCloud or iTunes account, whatever you want to call

01:51:50   it to 99 a month. And I don't know,

01:51:54   earlier in December I started getting some kind of error.

01:51:59   It was inconsistent and it was something,

01:52:01   something that my payment method had expired and it didn't

01:52:05   make any sense to me because I knew that my iCloud,

01:52:09   I knew that my iTunes credit card was not expired.

01:52:13   It's the credit card I have on my iCloud account is, or iTunes account,

01:52:17   whatever you want to call it.

01:52:18   It's actually an Apple card and I know that the expiration date is,

01:52:22   I don't know, a year or two in the future,

01:52:24   but if any card is going to give me a proper warning when it

01:52:29   expires, it's an Apple card. Right. Uh,

01:52:32   but I think what really happened is that now that under Elon Musk,

01:52:37   the new Twitter blue, which includes verification check Mark,

01:52:42   which I, I have, but I didn't even ask for,

01:52:45   and I honestly don't care about, I honestly,

01:52:48   you could take away the blue check Mark on my at Gruber account and I really

01:52:52   don't care. That's not why I'm, I want to pay for it.

01:52:55   But I think what happened is now that the new version of Twitter blue is

01:52:59   $8 a month.

01:53:00   If you pay Twitter directly on their website and $11 a month through the app

01:53:05   to account for the 30% cut that Apple takes. And for a while,

01:53:10   it seemed like maybe they'd grandfathered in those of us who'd been paying $3 a

01:53:15   month. But I think they, I think they wanted, I think they messily very,

01:53:20   can you believe it under Elon Musk that something was slapdash and haphazard

01:53:25   and not entirely, but I think that basically they,

01:53:30   they wanted to cut all of us off who were paying $3 a month and make us pay

01:53:34   eight or $11 a month.

01:53:36   But the error message was about the payment credentials being expired,

01:53:41   which they probably just dumped all the tokens,

01:53:44   the receipt tokens from their database and let it just error out to annoy people

01:53:48   like us. I, I let mine lapse before that a little bit before that.

01:53:53   Well, I'm so,

01:53:54   but I'm so keen on the top articles feature that I wanted to resubscribe.

01:53:59   I was like, okay, I'll resubscribe. I don't know if you saw this.

01:54:02   I tweeted it the other day, but they won't let me, they won't let me sign up.

01:54:07   I want to give Twitter $11 a month cause I'd actually would rather pay for it

01:54:11   through iTunes and no better control, better control. I'll pay,

01:54:15   I'll pay three more dollars a month knowing that I can cancel at any time

01:54:20   through iTunes than to, especially at this point,

01:54:23   trust Twitter with my credit card and a recurring subscription.

01:54:26   I would like to pay $11 a month to them for this and they won't let me

01:54:31   because the phone number I have associated with my Twitter account is from an

01:54:36   unsupported carrier. And now what that means now,

01:54:39   the carrier in question is Google voice. Oh, Oh now,

01:54:44   and I tweeted about it and people are like, well,

01:54:46   that's like a security thing cause lots of scammers sign up for Google voices

01:54:49   free. You get a phone number for free and crying out.

01:54:53   But the reason I use a Google voice account,

01:54:56   I use this Google voice account for any,

01:55:00   pretty much anything that only supports SMS as a second factor,

01:55:05   which is actually not very secure and not, not, you know,

01:55:08   yeah. Cause you can get the codes via email, right? Well, and you can also log in.

01:55:12   And I, and I do think carriers, you know,

01:55:15   especially the big ones like Verizon and AT&T and T-Mobile have,

01:55:20   I think it certainly anecdotally,

01:55:23   it seems like SIM hijacking is less of a thing that they've, you know,

01:55:28   but I still don't trust I'm on Verizon. I've been on Verizon for a while.

01:55:31   I don't,

01:55:32   I trust Google with my main Google account security way more than I

01:55:37   trust Verizon and way more than I trust somebody. I think,

01:55:41   I honestly don't see how it's feasible that somebody could hijack my Google

01:55:46   account.

01:55:46   And whereas I don't necessarily trust that somebody who specifically wanted to

01:55:51   target me, John Gruber, the daring fireball guy who has, you know,

01:55:56   300,000 followers on Twitter, somebody could go into a Verizon store and say,

01:56:01   I'm John Gruber and I, you know, this is my phone number and you know,

01:56:04   it's not that hard to get my personal cell phone number. I lost my, you know,

01:56:08   I need a new SIM card.

01:56:09   I trust Google more than I trust Verizon with that even though I more or less

01:56:13   trust Verizon, but I,

01:56:14   All the carriers put ID, the possibility,

01:56:18   but not requirement for an ID code to protect that.

01:56:20   And so if you don't have an ID code set on your, your wireless account,

01:56:25   then that's possible. Although they're having stories of people,

01:56:27   also social engineering around the requirement for an ID.

01:56:31   But I think that's why the incidents has gone down. It's also cause you know,

01:56:34   Bitcoin is worth less. So people have less interest in hacking most people.

01:56:38   But I think the ID thing was enough of a deterrent that it became less low

01:56:42   hanging fruit. But I think also if you went in with a,

01:56:44   someone went in with a faked bill and a faked ID and said they were John Gruber

01:56:49   and had your phone number, they might still be able to talk somebody into it.

01:56:51   That's not, not implausible.

01:56:54   I just feel safer with my Google account. I do, you know,

01:56:57   I have lots of complaints about Google's overall customer experience,

01:56:59   but I think Google's account security is top notch. And if you have to,

01:57:03   if I have to use SMS as a second factor,

01:57:06   I always use my Google voice account if possible.

01:57:09   And I do have my Google voice account set up to forward all text

01:57:14   messages to my main email. So

01:57:18   when I do, when, when they do text me, quote unquote,

01:57:23   text me a, a six digit code,

01:57:26   all I have to do is look at my email and it's right there. And you know,

01:57:31   it's just as easy as checking it from an SMS message, super frustrating.

01:57:35   And it's so funny because it's one of the times where I've run into the,

01:57:39   and I hate to, I tried never to use the word fan boy,

01:57:43   but I can't think of a better term. The,

01:57:45   the Elon Musk fan boys who took the, well,

01:57:49   of course they don't allow Google voice. It's, you know,

01:57:52   it's keeps the scammers out. Well then how come I use my Google voice number,

01:57:57   with everything else with no problem and I'm not hacked.

01:58:01   And I, to me, it's just, it's just a sign of slips,

01:58:06   slipshottedness at, at Twitter. Very frustrating to me.

01:58:10   You know, John,

01:58:10   I didn't see your tweet cause I haven't been using Twitter since November 20th.

01:58:13   That's I ain't going,

01:58:14   I ain't going back unless there's some big changes over there. And you know me,

01:58:18   I have like Twitter poisoning. Like you could,

01:58:20   you could take a blood sample from me and I'm above 0.08 Twitter percentage at

01:58:24   any given Twitter C or something.

01:58:26   We're active alcohol content and I just, I just, you know,

01:58:29   I just said I can't be on this service. I mean,

01:58:32   I left Facebook when I felt like they were enabling genocide and I left a couple

01:58:36   of years ago. I was like, that's it. And I miss people.

01:58:39   I miss Twitter as a social mechanism and,

01:58:42   but I'm benefiting from the fact that there's a land of refugees that's polite

01:58:46   and kind and every everything is chirping birds and rainbows for this minute.

01:58:50   And the mastodon and the Fediverse have been pretty good.

01:58:53   And I'll give you one stat, which is, I had almost 29,000 followers on Twitter,

01:58:57   which I have no idea why because my Twitter handle has never really been

01:59:01   associated with any media property. So it is all organic. Like somehow,

01:59:05   I feel like I've amassed it by putting peas in a pile on a table until they get

01:59:10   rid of that pile. It gets really big one at a time. But I was like, this is great.

01:59:13   You know, so when I do a project or something, I can announce it.

01:59:15   And I get a little uptake from Twitter. It's been great for Kickstarter. It's,

01:59:18   you know,

01:59:19   it's always like I'm looking for a place for good conversation to learn about

01:59:22   things, to hear voices I don't hear otherwise,

01:59:24   to hear some major political and intellectual figures discuss stuff where they're

01:59:29   talking more freely and, you know, and for, you know, to,

01:59:32   to talk to colleagues and friends and so forth. Right. So, and, but,

01:59:35   but also projects have always been an important part of it.

01:59:37   So now over at at Mastodon after, I don't know,

01:59:40   six months of being a pretty good user over there and scraping my Twitter feeds

01:59:44   to follow people on Mastodon. I have over 7,000 followers on Mastodon,

01:59:49   or about 25%. And for me, that's a lot because they're,

01:59:52   A, they're all real people. It's not bots and it's not random.

01:59:55   People are making a much more intentional effort to follow people. And so the,

01:59:58   so there's great conversation there.

02:00:00   And I really enjoying this burgeoning community and worried at what point it

02:00:04   becomes untenable or moderation becomes an issue or balkanization or other

02:00:09   things.

02:00:09   But the real test will be the next time I launch a project and does anything

02:00:13   happen? Do I launch something there? And on Twitter,

02:00:15   it would have been like a third of the,

02:00:16   of the purchases or interest or whatever come from Twitter.

02:00:21   And with Mastodon it's like, you know, one person out of a thousand or something.

02:00:24   Does a project fail because I'm not on Twitter anymore? So that, you know,

02:00:28   that's like my professional concern,

02:00:31   but personally I'm really enjoying being in a mellower place right now.

02:00:34   For the record, I just looked it up, Glenn, your Twitter account,

02:00:39   which is I guess you, we could call it hibernating, right?

02:00:42   It's exactly which, and again, the Darth, I don't pass judgment.

02:00:47   I really don't on anybody who deleted their Twitter account, you know?

02:00:52   No, I, but if you asked my advice personally, I would say,

02:00:57   just,

02:00:58   just stop using it and let your old tweets stay there for,

02:01:02   Oh, that's what I'm doing. Yeah, I totally agree. I don't want to,

02:01:06   Elon could be a temporary phenomenon and I don't want to give up on Twitter

02:01:11   because it's been valued professionally, personally, intellectually, artistically.

02:01:15   It's been an incredible thing over its lifetime for me.

02:01:18   And so I don't want to give up on Twitter as a thing.

02:01:22   So people are talking about deleting it and what's been good is over a Mastodon,

02:01:25   if somebody says, Oh, I'm deleting my Twitter. People are like, don't do it.

02:01:27   Just put a thing, tell people where to find you and leave it in Amber.

02:01:32   Make it read only for yourself. But if you delete it,

02:01:34   then your handle frees up and your history is lost. And you know, you're not,

02:01:38   you're not helping Elon by keeping an active account that you're not using if

02:01:42   that's your concern,

02:01:43   but you're helping yourself by making sure you're not giving up your identity

02:01:47   and your, and what you've said and all the rest of it.

02:01:49   Yeah. And as much as most of us has spent most of the last 16 years,

02:01:53   more or less dicking or just, just screwing around on Twitter.

02:01:57   Some of our tweets are worth referencing, you know, and it's good to keep up.

02:02:02   Again, no, if, if you're, you know, I, I, I keep saying,

02:02:06   I'm not passing judgment. I guess I am mildly where I would,

02:02:09   I would prefer people to leave their Twitter accounts up and just stop using

02:02:14   them. But I do, I,

02:02:15   I'm not mad at you if you deleted your account because I get it,

02:02:18   but I was going to make fun of you for how many tweets you posted.

02:02:23   I looked it up and the Glen F Twitter account finished, or, you know,

02:02:27   if it is finished, at least on hiatus with 117,447

02:02:32   tweets.

02:02:33   No, sorry. Sorry, John. There's a missing piece,

02:02:35   which is I deleted all my tweets from before 2018.

02:02:37   Oh, then no wonder it makes sense.

02:02:40   Yeah. There's like a half a million. And the reason was people were, yeah,

02:02:43   so I'm honest about that. I've posted times.

02:02:46   I first did below it before 2018 and later before 2019 because I used to use

02:02:51   Twitter more like IRC, right? Especially the early days.

02:02:54   And then there was that you remember there's an extended point where anything

02:02:57   you said was being used, was being harvested by,

02:03:00   and especially with gamer gate and other stuff. There was a point where I'm like,

02:03:04   I am just going to get rid of it because otherwise I have to go through a half a

02:03:08   million tweets and figure out if there's something that can be misconstrued or,

02:03:11   you know, this is what happened to not to defend.

02:03:13   I don't want to read open bean dad gate, but that's what happened to John Roderick.

02:03:17   It's not that John Roderick was starving his child.

02:03:20   He was telling a funny story and people went, Oh, this guy's a terrible father.

02:03:24   Let's make him the main character today and went insane. And it was, you know,

02:03:27   he's a storyteller. A but B people went back to find old bad tweets.

02:03:32   And what they found was he said things sarcastically that out of context sounded

02:03:36   like he was horrible and he's, he's not a horrible person. I don't, you know,

02:03:40   I mean, we all know people in comedies, a Seattleite,

02:03:42   it's he said things that maybe were too in a particular way.

02:03:46   If you read them sounded more negative than they were,

02:03:48   but they're absolutely in keeping with this sort of sarcastic tone.

02:03:52   So if you read them in sequence, you'd be like, Oh, this is a joke. I get it.

02:03:55   That's funny. He's making fun of those people. So at some point I'm like,

02:03:58   I do not need to spend the rest of my life dealing with things that I said in an

02:04:02   unguarded fashion. So I'm just going to delete them all. So I probably had,

02:04:06   like I said, at least half a million tweets that aren't on that number.

02:04:09   So this, that, thank God, because I looked it up my, my,

02:04:13   my account with, and I've never, I've never done a mass deletion.

02:04:17   I'm at 85,048 tweets. And I was like, wait, I'm in the ballpark of Glenn.

02:04:21   Why am I making fun of him? Yeah. All right.

02:04:24   No, I did a fundraiser for Planned Parenthood when I hit 500,000 and raised a few

02:04:27   thousand dollars from somebody. And I let someone, I was like,

02:04:30   you can be my 500,000th tweet. Tell me what to tell to tweet.

02:04:34   And then I think they didn't or something, but they still did any of the money.

02:04:37   So that was, that was, there was a point when I hit that number.

02:04:39   I've been, I signed up for Mastodon in 2018.

02:04:43   I've been talking about it the last few episodes of the show,

02:04:45   but the more I'm using it,

02:04:48   the more happy I am with it.

02:04:50   And the more I'm like my skepticism is fading away and all the better.

02:04:55   The ivory beta, right?

02:04:57   I do have the ivory beta from TapBots.

02:05:00   I mean, I just got in recently and I,

02:05:03   I've been using different Mastodon clients like Mammoth and the Mastodon,

02:05:07   the branded one and the, you know, and the website interface and all that.

02:05:10   And I got ivory and I was like, Oh yeah. I mean, no, those are terrible.

02:05:14   They weren't really meant for this kind of volume. And,

02:05:16   and there's all kinds of things that, you know,

02:05:19   Mastodon is evolving super rapidly. It's open source. There's people, you know,

02:05:22   working on checking in a bazillion new features and ideas and improving things.

02:05:26   So it's going to evolve super fast,

02:05:28   but these apps evolve less rapidly than the code base. So that said,

02:05:32   I started using ivory and I'm like, Oh yeah,

02:05:35   this is the warm hot tub that I've been looking for instead of the ice cold

02:05:39   bath.

02:05:40   I do not mean to, I, and I get it.

02:05:42   I don't know if they've filled up the entire 10,000 test flight

02:05:47   capability yet, but I know that when TapBots opened it up,

02:05:51   I think it was a couple of days before Christmas,

02:05:53   they opened up like another thousand slots and they said when they were going to

02:05:57   do it. And I think the thousand slots seriously went like 15 seconds. Yeah.

02:06:01   I think they have multiple, I don't know if they can permission.

02:06:04   I think they have multiple 10,000 slot things are doing something and they,

02:06:09   I think it was 10,014. No, that must not be right. It must be a thousand,

02:06:13   but it, you know, it's an out, it's a 0.1 alpha folks. It's not, it's, well,

02:06:18   they're doing it, but they're, they're,

02:06:19   they're rapidly progressing towards something that I think that they can ship as

02:06:22   a 1.0. It's very, I think it's surprisingly close. And I,

02:06:26   and I think, you know, I,

02:06:30   maybe I'll have the TapBot guys on.

02:06:32   I would actually be enjoyable for me to talk about this with them.

02:06:34   But I think though that years ago they abstracted

02:06:39   out the Twitter specific-ness of Tweetbots in the app.

02:06:43   Remember app.net, which was ahead of the app.net. I did too.

02:06:47   And it was ahead of its time,

02:06:49   but in anticipation of app.net possibly becoming a thing,

02:06:53   Tweetbots architecture was sort of abstracted away from being,

02:06:59   assuming that it was Twitter specific.

02:07:02   And that obviously was to their benefit to prepare for making a

02:07:07   Mastodon client. They can't,

02:07:09   my understanding of the Twitter PI rules is that

02:07:14   any third party Twitter client,

02:07:17   and everybody knows from years ago that Twitter has gotten as of 2018,

02:07:22   kind of funny about third party clients. And I, I'm still not,

02:07:26   even with Elon Musk,

02:07:27   I'm not entirely convinced that they're going to,

02:07:30   I wouldn't be surprised if by the time this episode airs, if you know,

02:07:33   somebody kicks the plug out of the wall on the API server, right?

02:07:37   But I, they can't,

02:07:39   the rules of the Twitter API for third party clients is that you're not allowed

02:07:44   to integrate with others competing services.

02:07:47   So a Tweetbot or Twitterrific can't build in

02:07:52   Mastodon support in the same app because that's a vice.

02:07:55   So they have to make a separate app as opposed to the

02:08:00   actual Fediverse of Mastodon and anything else that uses these open

02:08:05   protocols, right? Which, you know, Mastodon is truly in the spirit of the open web.

02:08:10   And if there's or like Matt and Reese's micro dot blog,

02:08:14   which is not Mastodon, but which uses and talks to the same API's.

02:08:19   So micro dot blog, if you're logged in,

02:08:21   you can follow people on a Mastodon site and get them into your micro dot blog

02:08:26   feed.

02:08:27   I saw a chart the other day. Someone's like, look,

02:08:29   if you want to know what the Fediverse is,

02:08:30   and reason reason some of us keep harping on about it is, you know,

02:08:33   the Fediverse is not Mastodon,

02:08:34   but by that we don't just mean that there's something like Mastodon is too

02:08:38   limiting or it's a name of a project. They said,

02:08:41   here's what it really looks like.

02:08:42   And they posted one of those things that are usually horrifying because they're

02:08:45   like, here's what the the ad services universe looks like.

02:08:48   And there's 4,700 companies occupying niches. Instead,

02:08:52   this was like 50 projects all kind of,

02:08:55   and they're all open source and some may have commercial aspect,

02:08:58   but they're all, you know,

02:08:59   they're all basically people trying to make this thing happen in different ways.

02:09:02   And so, you know, Matt and Reese's microblog.org is in there and microblog,

02:09:05   I'm sorry, I just said this thing wrong. Anyway,

02:09:07   my commitment and recess thing is in there and Mastodon, but then they were like,

02:09:10   again, like 48 other things, not all of which were about microblogging.

02:09:14   Some are different aspects of services interaction and how they all kind of fit.

02:09:18   And I was like, Oh, this is,

02:09:20   this is going to be exciting because we haven't even scratched the surface of a

02:09:24   mass usage in any one of those areas.

02:09:27   So this could actually cause a blossoming of all of these related Fediverse

02:09:30   things that we haven't even, you know,

02:09:32   haven't even risen to consciousness because we're not switching over from a

02:09:35   failing or faltering, whatever you want to call it, you know,

02:09:38   major social network. They have all different purposes.

02:09:41   Yeah. The note that I jotted to myself a day or two ago in anticipation of

02:09:46   talking to you about it is that what's occurred to me,

02:09:48   the more I'm using ivory and Mastodon and less

02:09:53   using tweet bot and Twitter is that Mastodon

02:09:57   just has the feeling of being a frontier. And,

02:10:03   and I'm suddenly recalling now deep in middle age,

02:10:07   how comfortable I am on the frontier of, of,

02:10:11   of new things like this. Right. Yeah. You know,

02:10:14   and going all the way back to getting on the internet

02:10:19   at the very beginning in the days when usenet was the thing,

02:10:23   you know, it, I didn't,

02:10:25   the whole thing seems so new that it didn't even occur to me to think of it as a

02:10:30   frontier because it was so obviously a frontier and BBSs, right.

02:10:34   Which weren't really connected to the internet, but it,

02:10:37   it all felt new and exciting and,

02:10:40   and things were happening where it's like, Oh man, all of a sudden,

02:10:44   a 14,

02:10:45   four modem is actually something I can afford.

02:10:48   And now I can download binaries at a reasonable speed from the BBS.

02:10:53   And it doesn't take, you know,

02:10:55   four hours to just download an app or a game or something that somebody shared.

02:11:00   And just the, I don't know,

02:11:02   it's exciting and fun to be somewhere where it's fluid

02:11:08   and new and changing and evolving rapidly.

02:11:13   And so far, everybody is just,

02:11:18   it, it, it's just a good community spirit.

02:11:21   Yeah. I, it's, it's not, I mean,

02:11:23   it's resistant to a certain kind of jerk or

02:11:28   troll. It's not immune to it obviously because of the instant based distribution.

02:11:32   And because moderators,

02:11:34   because they're not beholden to shareholders or other interests,

02:11:37   they're sort of beholden to the users that they host or to nobody.

02:11:41   They can make these independent decisions.

02:11:43   And that's the fear of balkanization is moderation is a hard problem.

02:11:47   It's very hard at scale. It's very easy to piss people off.

02:11:51   And it's also extremely easy to take positions that are at odds with,

02:11:55   you know,

02:11:56   some or most of the people on the instance you're running. So that's,

02:11:59   I know that's a big fear if you read, you know,

02:12:00   you read tech dirt like I do, I'm sure. And Mike Masnick just, I mean,

02:12:03   he and his colleagues there write tremendously good things about why moderation

02:12:08   scale is difficult and their fears about that relative to Mastodon plus some

02:12:13   liability issues. EFF has written a great roundup of things.

02:12:16   EFF about if you're a Mastodon, I'm saying EFF anyway, there's a,

02:12:20   there's a good article floating around now. I'm being erratic about if you're,

02:12:24   if you're a Mastodon host,

02:12:26   what it means for you in terms of making sure you comply with the MCA.

02:12:29   If you run an instance that is and have other people on it,

02:12:32   like how you comply with the MCA rules and which are relatively easy to comply

02:12:36   with. But if you don't, you could wind up in a pickle if somebody, you know,

02:12:40   wanted to post something that could be problematic and someone can't reach you

02:12:44   and so forth. So there are some risks, but I think, you know, this is,

02:12:49   you've seen this, I'm sure the last week or so there's been a quote,

02:12:52   tweet turmoil raging on Mastodon because Mastodon does not offer a quote,

02:12:57   tweet function, a quote, toot function or whatever we're going to call it.

02:13:00   This was my next topic. My, my final thing to ask you about. So explain,

02:13:06   explain to people who aren't sure what the controversy is.

02:13:08   So, so on Twitter, when Twitter news quote tweeting,

02:13:11   before quote tweet existed as a thing on Twitter, people did it vernacularly.

02:13:15   You would copy a URL and you'd say, blah, blah, blah, blah, blah,

02:13:17   my opinion and paste it in, or you just paste in the URL sometimes. I mean,

02:13:21   I had a, sorry, there was a retweet feature for a long time. So retweet,

02:13:25   you just select it.

02:13:26   But if you wanted to comment on or add an image or do anything,

02:13:29   you would just make a new message and paste in the Twitter URL.

02:13:32   And depending on your client or whatever,

02:13:34   it would show the previous URL in line.

02:13:36   Then Twitter created that as an actual feature.

02:13:38   So you could quote tweet and it would sort of structurally support it.

02:13:42   But it was designed in such a way that even people who like, you couldn't,

02:13:45   you couldn't quote, tweet restricted accounts,

02:13:47   but there were other ways in which like if you were blocked from seeing somebody,

02:13:50   you could see there a quoted tweet version of it and there were abuse vectors

02:13:55   and so forth. There's a, the huge debate raging is,

02:13:58   so Mastodon has not been designed with a quote tweet feature.

02:14:01   And that's partly because, you know,

02:14:03   each Mastodon message is essentially hosted on the instance on which you're

02:14:08   located. And there's like 4,000 Mastodon instances now,

02:14:11   at least or last time I checked. So, you know, when you're retweeting something,

02:14:15   it's all centralized. When you're re-masted, you're re,

02:14:18   you're re boosting something or whatever you would call it in the Mastodon world.

02:14:21   It's a question of how they gets drawn in and, and structured.

02:14:24   So there's apparently some code in the Mastodon project that would enable a form

02:14:29   of re boosting and it's not rolled out. And there's a structural issue,

02:14:33   which is that some people maintain that retweeting is a very,

02:14:37   or sort of quote tweeting, I should say specifically has,

02:14:40   is an easy way to brigade people and people who are,

02:14:42   and so it's not like it's exclusively by any means used for that purpose,

02:14:46   but people feel like it's a very easy way to point a lot of people at somebody

02:14:50   who said something and create main characters of the day. Like the main,

02:14:54   you know, quote Ryan Broderick of Garbage Day, I love his stuff.

02:14:58   He's got a podcast, Content Minds. And, and when he talks about Twitter,

02:15:02   the main character of the day or the internet's main, you know,

02:15:05   main character, whatever, that is a flaw of Twitter, not a feature, right?

02:15:09   We shouldn't have a Bean Dab. We shouldn't have a Has She Landed,

02:15:11   Has Justine Landed yet?

02:15:12   Like all these things typically target people with the wrong kind of attention.

02:15:16   It's rarely solitary. It's almost always an exaggerated character, whatever.

02:15:21   So Mastodon is not designed with that feature yet. It's possible to add it.

02:15:25   It could be added.

02:15:25   And the raging debate is whether Mastodon is failing as a service,

02:15:31   even though it's still, you know,

02:15:32   just getting all this usage is a few years old because it isn't offering quote

02:15:35   tweeting and whether there's a backlash where people are like,

02:15:38   we never want to add it or it has to be in such a way that an instance could

02:15:43   turn it off so that somebody at another instance couldn't quote tweet someone an

02:15:47   instance that had it disabled or it should be disabled on a message by message

02:15:51   basis. So I think this is, the great part is this is vibrant,

02:15:54   excellent discussion about abuse vectors and whether this could cause problems.

02:15:58   And the downside is like, Oh God,

02:16:00   we have our first like conversation that takes over the whole timeline.

02:16:03   So I've been muting people selectively for like three days, right?

02:16:07   This is built in muting and features Mastodon was designed with muting and

02:16:10   blocking features that are really good. And I'm like,

02:16:12   if people are going on and on about quote tweeting, I'm like,

02:16:15   I'm going to mute them for three days and see if the fever passes because I

02:16:18   don't care about it enough that I want every other message of my timeline to be

02:16:22   talking about quote tweet. So at some level Mastodon has survived,

02:16:26   has succeeded in becoming a mass medium because everyone is talking about the

02:16:30   same thing. Yeah, it's,

02:16:33   I'm of the opinion that don't think quote tweeting is

02:16:37   inherently problematic as a vector for,

02:16:41   as you said, brigading.

02:16:42   But I appreciate the fact

02:16:47   that the Mastodon at a community level is being,

02:16:52   the community is evolving from a conservative

02:16:57   viewpoint on those issues. So I,

02:17:00   I think that quote tweeting should be a Mastodon feature.

02:17:04   I think it's too useful.

02:17:05   I think the idea that it's inherently problematic in terms of the way

02:17:10   that it can lead to pile-ons and dunking is overstayed,

02:17:14   but I appreciate that they're,

02:17:16   they're going from a let's be careful viewpoint going forward.

02:17:20   And the reason though, I think it's futile to not have it

02:17:24   is it's that, that line from Jurassic park,

02:17:30   you know, that life will find a way, you know, like, like,

02:17:33   like with the idea that, Oh, we're only going to have female dinosaurs.

02:17:37   So therefore there will be no breeding out in the wild and, ah, you know,

02:17:40   life finds a way something's going to happen and people find a way. Right.

02:17:45   And it's the way that Twitter itself evolved, where even,

02:17:48   even mentions at,

02:17:50   at username mentions were a, a,

02:17:54   an idiom that the users on Twitter came up with that did nothing

02:17:59   in the API APIs. They weren't even hot links, you know, originally,

02:18:03   it was just became a convention that Twitter users had,

02:18:07   where I would say at Glen F and say something in response to a

02:18:12   tweet of yours.

02:18:13   And then the third party clients would pick up on those things through a search

02:18:18   just for the string. It wasn't a magical feature.

02:18:21   And retweeting started like that all sorts of almost everything on Twitter,

02:18:27   yeah. Hashtags is probably the,

02:18:30   maybe the best example where it wasn't a built-in feature of Twitter at all.

02:18:35   It was just, well,

02:18:36   here's something that you can do because it's just text in a tweet,

02:18:40   just use the pound sign and put a string,

02:18:44   a unique string of characters that is human readable after it.

02:18:47   And then third party clients could do it. So without quote, tweeting,

02:18:52   Mastodon clients would evolve to do it anyway. And I saw,

02:18:56   I even saw in the discussion people are having where there's clients that are

02:19:00   talking about doing things.

02:19:01   And I think there's already people outside of the actual client software who

02:19:06   ginned up their own solutions with scripting or shortcuts,

02:19:10   where you take a screenshot of the tweet,

02:19:13   you want to quote tweet and a link to

02:19:17   the, or not tweet, but to whatever they call them, right?

02:19:21   A Mastodon post, take a screenshot of it,

02:19:24   get the URL that points to the original and then pre-populate a new

02:19:28   post from your account with the URL to the original and a screenshot of the

02:19:33   original. And then you can do your quote,

02:19:36   tweeting and it's effectively a quote, you know,

02:19:38   people are going to find a way to quote, tweet,

02:19:40   whether it's a built-in function or not, but if they make it a built-in function,

02:19:44   then they can do things like what you mentioned, where you could say,

02:19:49   I would like to make this tweet unquote tweetable, because I don't,

02:19:53   I don't want to hear from people about it.

02:19:54   I realize that what I'm about to say people might think is controversial because

02:19:58   you know, whatever the issue is, or I, you know, or you just don't want,

02:20:01   you just don't want to be bothered with it. You could just say,

02:20:04   make this one unquote tweetable,

02:20:06   but as long as quote tweeting is a built-in feature,

02:20:09   then there's not going to be any demand to build those work around features to

02:20:13   get around it.

02:20:14   I totally agree.

02:20:15   And you saw Twitter even evolve where they were like,

02:20:17   you could set who can reply to this tweet,

02:20:20   which I thought was a great feature and underutilized. So, I mean,

02:20:23   I love the idea. I love the idea that there's that massed on is coming out.

02:20:27   I think structurally they're coming at it from a position of having learned from

02:20:31   gamer gate and learn from Twitter and learn from mass abuse attacks.

02:20:34   So the system as a whole seems to be designed for least harm

02:20:39   as its starting point.

02:20:41   And so anything that's added has to be fit within that framework. And so if,

02:20:44   you know,

02:20:45   maybe there's going to be six months of conversation about quote boosting or

02:20:48   whatever, or maybe there'll be three weeks of it, but when it rolls out,

02:20:52   I'm much more confident with it that it'll be a consensus driven thing that has

02:20:56   more options. So like, even when you post a message in massed on,

02:21:00   there's the quote, the content warning link,

02:21:02   which some people just treat like a subject line, which is also fine.

02:21:05   Some people ignore, there's a lot of debates about how that gets used. You know,

02:21:08   should it be used to warn about racism or is racism something we should talk

02:21:11   about openly without content warnings? You know, this is an interesting debate.

02:21:15   It's good one to have. It's good one to expose.

02:21:17   There's the sort of four levels of post visibility.

02:21:20   You can post something that doesn't get indexed or isn't listed.

02:21:24   And there's like a lot of really,

02:21:26   or even you add an image and it presents you with this ability to even in the

02:21:31   native, you know, the, the website based client, the web app,

02:21:33   you can click and extract text through OCR from the image you've uploaded.

02:21:38   So if you're posting a screenshot and I've been trying to really obsessively put

02:21:42   at least some text into every image I post because it's a cultural norm among

02:21:47   the early Western adopters.

02:21:49   But I also think I like to do anything in which accessibility is incorporated by

02:21:53   design. I want to support and uphold it. And it also, I think it makes you,

02:21:57   I think it makes things better for everyone if you use accessibility features.

02:22:02   So yeah, I mean, I think the community as it stands,

02:22:05   even with millions of people onboarding has still vibrant but not seemingly

02:22:09   horrible debates about the direction of it,

02:22:11   but the direction will be set by people involved in an open source project that

02:22:15   is, you know, it's got aspects of, you know, some people making decisions,

02:22:19   but also a lot of consensus driven thing. And what's not underlying it is a,

02:22:23   we have to increase engagement. We need more people posting every day.

02:22:26   We need to squeeze more ads in. I don't think we've had anything like that.

02:22:30   Maybe ever. That's really been driven in entirely by community needs,

02:22:34   except maybe usenet. I mean, that's the last thing that's really like this,

02:22:37   isn't it? Yeah, I think so. And I, you know,

02:22:39   I was just reading how usenet is still a thing and you know, it, in fact,

02:22:44   it's growing in some ways. I mean, it, it, I don't know.

02:22:48   It's just a Mastodon is so clearly of the spirit of the open web.

02:22:52   And it, it thrilling to me to see it take off.

02:22:56   And I'm happy to say that my skepticism,

02:22:59   my initial skepticism that it was going to be able to work or scale, um,

02:23:04   it, you know, is wrong. I've never, I'm,

02:23:06   it's one of those things where I'm happy to be wrong about,

02:23:08   I hope it stays that way. I kind of feel,

02:23:11   and this is one of those things that I hope doesn't sound elitist,

02:23:14   but I kind of feel is fundamentally true that maybe having a

02:23:18   slightly higher bar to sign up and get

02:23:23   started compared to Twitter is for the better for,

02:23:28   for those of us who are happy to step over that bar,

02:23:31   that one service for

02:23:34   300 million people like Twitter is, is two, you know,

02:23:40   it's actually not good to be on a platform with 300 million other people.

02:23:44   Well, someone could start an unfederated Mastodon that was instance that was,

02:23:48   or not unfederated, but it could be loosely federated. Someone could have a,

02:23:51   you know, not only fans, that's the porn thing. I'm sorry. What's the,

02:23:53   what's the celebrity things, right? There's some celebrity services,

02:23:57   but they could not, what's the one where you call in people,

02:24:00   you can get a custom message, a custom video from somebody. Oh, what is that?

02:24:04   Oh, that's a different kind of custom video. It's something I forgot,

02:24:09   but that kind of thing, you know,

02:24:10   someone could set up a Mastodon instance that's meant for a lot of people as very

02:24:15   easy onboarding and it's designed to be a distributor of news and celebrity

02:24:20   information and the celebrities all go there to post stuff because it's safe and

02:24:23   maybe it's federated, but it's federated unit or actually,

02:24:26   so celebrities tweets go out, but people are members of it,

02:24:29   aren't federated out to the rest.

02:24:30   So you could have 10 million people on a,

02:24:33   on a server someplace doing mostly read only where,

02:24:36   which is what most Twitter usage, I mean, of 300 million users on Twitter, you know,

02:24:40   you've seen those big head long tail curves,

02:24:41   a tiny fraction posts more than like a message a year.

02:24:44   And most people use it in most people use Twitter in a way that you and I have

02:24:49   no, you know, most people it's a reading mechanism.

02:24:52   It's a reading medium with very little conversation and Mastodon is very much a

02:24:57   conversation medium and the early people are very much talking.

02:25:00   So that could evolve too for sure.

02:25:02   Cameo is what you're thinking.

02:25:04   Cameo. Thank you. Yeah.

02:25:06   I was very, I was very, I have not paid for a cameos from anybody,

02:25:10   but when Rudy Giuliani was on, I was very,

02:25:12   I was very tempted to get, get, get,

02:25:15   get some Rudy messages to a bunch of my friends.

02:25:17   You don't want to get a Danny DeVito message for your birthday. Are you sure?

02:25:20   No, I want to say I want to send other people, Rudy Giuliani.

02:25:23   Oh, that's interesting.

02:25:24   But anyway, that's about it for me. Tell me about shift happens. And I,

02:25:28   you got this in the show notes. So tell me about this.

02:25:30   Oh, I just were like, oh, this is a, this is a book.

02:25:33   Marching Wishery is is the author of this this tome,

02:25:36   it's 1200 page two volume book set with a third extra

02:25:41   volume that he started working on in 2016.

02:25:44   It's about the history of keyboards from typewriters to like the present day.

02:25:48   Boring. I have no interest whatsoever. Keep me away from this.

02:25:53   Duh.

02:25:53   Nobody listings a show has any interest in keyboards or typewriters.

02:25:57   Absolutely aware of it. And yeah, it is, it's going to be catnip for a lot of people.

02:26:01   But so marching started on this in 2016 is kind of a outgrowth of some

02:26:06   essays he's writing at medium. And then he wandered in by accident.

02:26:10   He was on the wrong day in the wrong place and he wandered into a museum that

02:26:14   should have been closed in Spain and then did not speak the language.

02:26:19   People there were speaking Catalan. It doesn't speak Spanish or Catalan.

02:26:22   And he got finally got pushed upstairs and it was the most glorious collection

02:26:26   of typewriters in the world just about.

02:26:28   And this led him on this journey to write this book.

02:26:30   And so a couple of years after that, I'd been, I'd met him on Twitter and I'm like,

02:26:34   Hey, you got an editor for this book? Cause I I'm literally looking at this.

02:26:37   Sounds great, whatever. So for, for five years plus,

02:26:41   I've been involved with him on this editing and development journey and he's

02:26:45   been at it for seven and it's going to be great.

02:26:47   We're going to go to Kickstarter in February and he's got a site up at

02:26:51   shifthappens.site just launched literally today while we're recording this.

02:26:56   We're just getting it finished. And John, let me tell you,

02:26:58   getting a book printed like this is a Kickstarter project and then we'll do

02:27:01   pre-orders after if we reach the goal and it'll ship in like August,

02:27:04   September of this year.

02:27:06   Like the book is pretty close and we kind of need to go over the top.

02:27:08   But it started like almost 400,000 words.

02:27:12   We got it down to under 300,000. There are so many great stories in this thing.

02:27:16   We're like, all right, so a year ago we've been getting printing bids.

02:27:19   We're sort of going onto it and then everything went out of control, right?

02:27:23   Everyone's experienced this,

02:27:24   but like shipping containers from China went up to like $25,000 up from like

02:27:29   5,000 and paper prices go through the roof and inflation is whatever.

02:27:32   This is before Russia invades Ukraine, right? And we're like, God,

02:27:36   you're going to have to delay this a year because the supply chain is so crazy.

02:27:38   I'm not sure if we, if we raise the money in February or March, 2022,

02:27:42   if we could ship it in the same calendar year and you know,

02:27:45   from both holiday sales issues and tax reasons,

02:27:48   you kind of want your revenue and your expense to be in the same calendar year.

02:27:52   If you're a cash based business, which most businesses are anyhow.

02:27:56   So that's the Ukraine invades or Ukraine is invaded and so forth.

02:27:59   But at one point in this last year, a few months ago, we're like, okay,

02:28:02   so if we order the paper from this factory in Gmund, Germany on,

02:28:06   on Lake Gmund, the Gmundze,

02:28:09   then we'll get it shipped over land into a container and then have it loaded on

02:28:15   boats and shipped to Maine where we need to, we were,

02:28:19   we were really going to order like tens of thousands of pounds of paper from

02:28:24   Germany and have it shipped over the Atlantic at one point.

02:28:26   And fortunately we were able to find a very good domestic U S paper that was

02:28:31   close enough, but holy cow, it has been hilarious logistics part.

02:28:36   So we finally have like a printer, we have a paper, we have a bid that works.

02:28:40   We have,

02:28:40   we've just got a press test with images printed to test profiles and they're

02:28:47   using stochastic screening, which is the new hotness. Everyone does this now.

02:28:51   Half tone screening is, is a thing of the past, John.

02:28:55   Does it make you feel old? Yes.

02:28:56   Stochastic screening is so amazing. I mean, it's been in use for years,

02:29:01   but it's gotten better and better. And we got some samples a year ago.

02:29:04   This one press up in Canada was like, first they were, they said, Oh,

02:29:07   we print such and such. I'm like, Oh, I have a bunch of those.

02:29:09   Cause I like that. And second,

02:29:11   they send a samples and I'm looking under a loop at like 10 X magnification and

02:29:15   I'm like, this cannot be, they're, they're printing green, but this is CMYK.

02:29:19   Why does it look so good in the alignment? The registration is so perfect.

02:29:23   So with stochastic screening, it's, you know, it's like dithering.

02:29:26   It's like really good algorithmic dithering,

02:29:29   but at a level where you're like at thousands of an inch dots, it's like,

02:29:34   you know, the finest, almost finest laser writer or a laser output dots. Right?

02:29:38   So when you have fine type,

02:29:40   like you have the letters in a shop window or letters on a, on a,

02:29:44   someone's typing a letter and you have a picture of a typewriter, you can zoom,

02:29:47   and this is half tone.

02:29:48   You can zoom in and you can actually read the words on the typewrited letter

02:29:53   and like an image that's like three by three inches.

02:29:56   And the typewriter is a fraction of that. So the screening is,

02:29:59   this is not limited to the printer we're using, but it is like,

02:30:03   it is like a digital revolution in the sense of like when we went from, you know,

02:30:07   wax and knives to, to desktop publishing and,

02:30:11   and early desktop publishing was sort of crummy, but it was fine. It was good.

02:30:14   It was good enough.

02:30:15   Stochastic screening is such a leap for me as someone I co-wrote some of the

02:30:20   editions of real world scanning and half tones, as you may recall.

02:30:24   We had a little bit about stochastic screening in the second and third editions.

02:30:27   And I'm like, Oh my God, it is. It's like, you're not, you're not like, Oh,

02:30:31   are we going to get a moiré? And Oh,

02:30:32   are they using open rosettes at a 90 degree? And you're like,

02:30:35   you just send the images with the profile to the printer.

02:30:38   And we got them back in printed form off the press. They did one signature.

02:30:42   So 16 pages aside and folded it for us to look at.

02:30:45   And we're going to use these as we'll be sending some of these out to people and,

02:30:49   and you look at it and you're like,

02:30:50   this looks as good as it looks on screen because you're not dealing the half tone

02:30:54   dots. It's not that they're super fine. They just don't exist. Anyway,

02:30:58   it's one of the most it's, so it's, I'm excited about it with this book.

02:31:01   It's not like a,

02:31:02   it's not super new so people who've been getting stuff printed listing this will

02:31:05   be like, Glenn has lost his mind.

02:31:07   I was using stochastic screening 10 years ago and it was great then,

02:31:10   but it's exciting to be involved with a project.

02:31:12   Well, I think every time you're on the show,

02:31:14   some of the people think you lost your mind, Glenn. Well, that's for sure.

02:31:16   Well, you know that I'm going to order the damn thing and you know, I'll be,

02:31:20   I'll be up for a Kickstarter link come February. But in the meantime,

02:31:25   people can go to shift shift. Don't forget the F that's right.

02:31:29   Shift happens dot site and get the, get the preview. But I,

02:31:33   this is a great, I just a great story. Oh, sorry. Oh, well,

02:31:35   I just suspect that there's a lot of people listening who are going to be

02:31:38   intrigued.

02:31:39   I think, yeah, I think there's a, we think we know our audience and it doesn't,

02:31:43   the book doesn't pander, but like at the last minute, just a few weeks ago,

02:31:46   I get an email from Marcin and he's like,

02:31:48   I just discovered a Nobel prize winning physicist got obsessed about using lasers

02:31:53   to erase typos off typewriters. And I'm like, what?

02:31:56   And so he had a rewrite a chapter about, it's just,

02:32:00   the story is insane. But the great part is he's like,

02:32:03   I want to get a picture of this thing. And we're like, well,

02:32:05   can we get the picture? Like, well, what if we contact?

02:32:08   So he was able to go to the institution that had the documents,

02:32:11   get original materials. So he's doing, you know, real research.

02:32:15   But in another case there was some photo he was trying to get and I'm like,

02:32:18   well, that guy's dead. But you know,

02:32:20   maybe this person who knew him is the estate is the, it's the,

02:32:23   we call it the executor of his estate.

02:32:25   So Marcin's like he emails that guy and that person puts him in touch with the

02:32:28   fellow's daughter and gets permission to use this wonderful photo in the book

02:32:32   that we otherwise wouldn't have been able to use.

02:32:34   But it's that kind of process where like, you're just, he's,

02:32:37   he's going to have some stuff in there that's never been seen before,

02:32:39   but there's also these, just these stories where you're like,

02:32:42   this couldn't have happened. Like, no,

02:32:44   this is a real thing related to typewriters or a keyboard and it's,

02:32:47   it's a hoot.

02:32:48   Well, I look forward to it. I thank you for your time.

02:32:51   I hope you have a good rest of the holiday week. Let me,

02:32:54   Happy new year. Yeah. Let me give a shout out here to our sponsors. One last,

02:32:59   one last message here. We've got the Squarespace where you can make a next move,

02:33:02   go use code talk show for 10% off your first order collide.

02:33:07   The cross platform end point end point security solution for teams that value

02:33:12   privacy and transparency and Linode where you can instantly deploy and

02:33:17   manage an SSD server in the Linode cloud and new accounts get a hundred dollars

02:33:21   credit. Thanks, Glenn. Happy new year. Thank you. Happy new year to you.