PodSearch

ATP

170: Casey Beats John, 29–29

 

00:00:00   That is the challenge of pixel art, isn't it?

00:00:02   To make a picture look good using pixels.

00:00:04   Or we can just rotate the pixels 45 degrees

00:00:06   and pretend we're in some crazy land.

00:00:08   Hey everyone, just me for a second.

00:00:11   We just put up our 2016 t-shirts, finally.

00:00:15   They're only available until June 3rd.

00:00:17   And they won't be delivered in time for WWDC

00:00:19   because we took too long to decide,

00:00:21   but these are awesome,

00:00:22   and I think they're gonna be worth the wait.

00:00:24   So see for yourself at ATP.fm/shirt.

00:00:28   Thanks a lot.

00:00:29   #CaseyWasRight? Maybe? It's been stunning watching the fallout from Fast Food Gate 2016,

00:00:41   which I just coined moments ago. A lot of people have been on my side. I wouldn't

00:00:46   say it's overwhelming, but the fact that I think the feedback I've seen has been

00:00:50   roughly 50/50, I'm counting that as a win for me.

00:00:54   I don't think that's how winning works, exactly.

00:00:57   Roughly 50/50, and I declare victory.

00:01:00   No, that went—John, if there's one universal truth to the internet, it's that the internet

00:01:06   believes that John Syracuse can do no wrong.

00:01:08   That is not a universal truth.

00:01:09   That is fact.

00:01:10   You're projecting.

00:01:11   And so—perhaps.

00:01:12   That's neither here nor there.

00:01:16   The internet is of the belief that John Syracuse can do no wrong, and parts of the internet

00:01:22   have said John Syracuse has done wrong. And actually, it's funny because some people,

00:01:28   and this is not a joke, some people have written in and said, and I don't have an exact verbatim

00:01:32   quote in front of me, but something along the lines of, "I thought John could never be wrong,

00:01:36   but he's wrong about this food conversation," which has made me so happy.

00:01:42   So to clarify, John being wrong means that people think Subway is better than Sbarro's? Is that what

00:01:50   yes or is not an utter unequivocal abomination. Well as in all things like the the actual

00:01:56   points that people respond to were only tangentially related to points made by the actual show.

00:02:02   The most solid ones were I felt like for people who were just saying I like or don't like

00:02:07   a particular restaurant that's fair that's fair right and because then you give all right

00:02:10   like we were talking about restaurants and you could say I love Subway or I hate Subway

00:02:15   I love Sbarro I hate Sbarro right but the ones who came away with the conclusion that

00:02:19   was that the comparisons were made that were actually not made. For example, comparing

00:02:24   Subway and Sbarro and what you would rather eat in a show in which I said I, you know,

00:02:28   or McDonald's even, in a show in which I said I eat Subway way more than McDonald's and

00:02:31   yet you have people saying that that meant that McDonald's was better than Subway or

00:02:35   or Subway was better than McDonald's. Anyway, lots of people latched onto their own particular

00:02:40   competitions. But I would say in the realm of Subway is good, Subway is bad, Sbarro is

00:02:44   good, Sbarro is bad, where they didn't try to compare the two of them, it was about 50/50

00:02:49   It's like the, which I just googled, the 1968 Yale vs. Harvard football game, which

00:02:53   was listed in the newspaper.

00:02:55   I've got to find this.

00:02:57   All right, here it is.

00:02:58   The famous headline, which they bury in the last sentence of the paragraph, the headline

00:03:02   after this football game in 1968 was Harvard beats Yale 29-29.

00:03:07   Because it was a game that ended in a tie.

00:03:10   So this is, yeah, Casey beats John 50-50.

00:03:14   I'll take it.

00:03:15   I'm good with that.

00:03:16   You're not going to hear me complain.

00:03:17   Anyway, I would just point out that literally all these issues are a matter of taste in

00:03:23   all senses of that word.

00:03:24   So there can be no victory unless one of these food items is going to kill you, which I don't

00:03:30   think either one will.

00:03:31   That's true.

00:03:32   These issues are a matter of taste, but my taste is better than your taste.

00:03:34   Well, that's how taste works, doesn't it?

00:03:36   But anyway, it's not as if there is a definitive answer, you know what I mean?

00:03:40   Oh, goodness.

00:03:41   But I will take the "John loses to Casey 50/50" or "Casey beats John 50/50."

00:03:48   Or you can do one to one.

00:03:50   Out of 50/50 we're just using it as percentages, but since this was a football score, it's

00:03:53   point, so maybe it's one to one.

00:03:55   Yeah, I'll take me beating you with a tie.

00:03:59   That is good in my book.

00:04:00   Anyway, just so you know, since you don't read my @ mentions, every time I say anything,

00:04:04   somebody says, "Boy, I usually agree with what you say, but you're really wrong about

00:04:07   X."

00:04:08   That happens all the time.

00:04:09   That's like 50% of my replies are people telling me that they usually agree with me.

00:04:11   with me, but they disagree with me about "insert whatever." And whatever you can

00:04:16   think of has been there.

00:04:17   Yeah, but to be fair, people start from the position of "Oh, Jon is right about this."

00:04:23   Whereas without turning this into analog, I'm not as convinced that people feel that

00:04:27   way about me.

00:04:28   Well, then you're spared all the tweets about how everyone is so disappointed in you

00:04:31   every time you have an opinion on anything. Usually I agree with everything you say, Casey,

00:04:34   but I'm very disappointed in your opinion on the Windows XP bumper sounds, or cars,

00:04:39   or the restaurants you want to go to, or your choice of, you know, any other thing you possibly—anyway.

00:04:46   Yeah, you're right, those replies do show a sort of the baseline assumption. But on

00:04:52   the other hand, if that baseline assumption doesn't exist, I don't get those tweets.

00:04:56   There's that. But all I'm going to say is I will take your struggle, which is oh,

00:05:00   so real over mine. Anyway, but no, I'm really happy that we kept, and we/Marco kept that

00:05:08   conversation in the show because I thought it was really funny. And as much as the bickering

00:05:12   back and forth between listeners and between us went on, I still appreciate the fact that

00:05:19   everyone seemed to take it for what it was meant to be, which was, yes, we were killing

00:05:23   each other, but it was all in good fun. And we got some good feedback about the segment

00:05:27   in general, which made me really happy. And like I said, Kacey beats Jon 50/50, so I'm

00:05:33   good with that.

00:05:34   massive overwhelming agreement that it was fun listening to us argue in the aftershow.

00:05:38   I think one person didn't like it and Port Marco had to say, "You were literally the only person."

00:05:42   He was. We got literally one complaint.

00:05:45   Yeah, it was funny. And that person, I believe, was like, "I'm sure I'm not the first person

00:05:50   to say this, but that was not good." And as it turns out, individual, you were the first person

00:05:56   to say that and the only person to say that. Anyway, I thought it was a lot of fun. And even

00:06:01   even though I wanted to kill both of you,

00:06:03   particularly Jon, 'cause I think Marco kind of

00:06:05   ended up as Switzerland by the end of the conversation.

00:06:07   - Yeah, I don't know why you focused all your anger on me,

00:06:10   'cause I re-listened to that thing,

00:06:11   and Marco was the one constantly poking and prodding you

00:06:13   and egging you on.

00:06:14   I was the one trying to bring it back to like

00:06:16   Siri and Veeve and stuff, and then Marco's like

00:06:18   throwing these little bombs then walking away.

00:06:20   It was totally Marco's fault.

00:06:22   It was not me, go back and listen to it.

00:06:24   - Well, but I also was not really 100% on either of your

00:06:27   sides, but I was mostly--

00:06:29   - You're just an instigator.

00:06:30   - Yeah, but I was actually leaning mostly towards Casey's side, not that this stuff

00:06:34   was very good, but that I thought that Subway was less terrible than Sbarro. Now I know

00:06:40   at least like when we're all at WWDC in a few weeks, we have to go to that mall food

00:06:46   court.

00:06:47   - They don't have any of those brands there. They all have weird stuff. Here's the one

00:06:51   we can put in the after show, WWDC box lunch versus any fast food.

00:06:58   those box lunches aren't bad. Sometimes they're not so good.

00:07:01   Sometimes, sometimes they are. Yeah. Well, the box, the problem with the box lunches

00:07:06   is that they basically all just taste like whatever the salad dressing was that they

00:07:12   used to coat every ingredient that day. Yeah, that's probably fair. And the ingredients

00:07:16   themselves are often made of cardboard. Yes. So you're basically tasting very chewy cardboard

00:07:21   that tastes like salad dressing. That's not a bad, that's not a bad summary. Yeah. Although

00:07:26   So apparently at Google I/O, which we're going to talk about quite a bit later, they don't

00:07:31   serve you lunch, it seems like.

00:07:33   Well, they don't even serve you a roof.

00:07:35   Yeah.

00:07:36   Sit outside in the sun.

00:07:39   90 degrees?

00:07:40   Like lizards on hot rocks.

00:07:42   Oh, goodness.

00:07:43   I can't, I can't handle this already.

00:07:47   So Aline, a friend of the show, Aline Sims, wrote in and said In-N-Out does not use frozen

00:07:53   burger patties.

00:07:54   I don't remember when or how that came up.

00:07:56   I said that. We regret the error.

00:07:58   Fair enough.

00:07:59   Sorry, In-N-Out. I've only been to In-N-Out like four times, so I've seen them cut the

00:08:02   fries, so I know they're taking potatoes and running them through a machine and cutting

00:08:05   the fries. But I lumped them in with the—it tastes more like a fast food chain, like the

00:08:09   people who get frozen patties, like McDonald's and Burger King. But In-N-Out does not do

00:08:13   that.

00:08:14   Excellent. Yeah, In-N-Out is very good, but I think as we said the last episode, I wonder

00:08:18   if the reason I like it so much is just because I can never have it. And if it was somewhere—if

00:08:23   If it was like Five Guys where I could consume it all the time, I'd be like, "Well, actually,

00:08:26   like Five Guys," and I'd be, "Eh."

00:08:28   Yeah, I had In-N-Out two times in one week, and by the second time, I was like, "Hmm,

00:08:32   all right, novelty's wearing off."

00:08:34   Yep, completely agree.

00:08:35   That's what I did when I was in California, and I felt the exact same way.

00:08:38   Man, this is no fun when we agree with each other.

00:08:40   So yeah, screw Sbarro.

00:08:42   Give it time.

00:08:43   We'll get there.

00:08:44   All right, tell me, Jon, about tech support ads being banned from Bing.

00:08:49   A couple shows back, I talked to my mom,

00:08:51   falling for a tech support scam.

00:08:53   Where, by the way, the consensus seems to be

00:08:55   that what these tech support scams really want

00:08:56   is not to steal your information

00:08:58   or put ransomware in your machine,

00:09:00   although those are all definitely possible

00:09:01   and I'm sure have happened.

00:09:02   But percentage-wise, it seems like the thing

00:09:05   that most of them are doing is trying to scare you

00:09:07   into thinking your computer has something wrong with it

00:09:09   and getting you to sign up to a monthly fee

00:09:10   for them to essentially do nothing.

00:09:12   So they make your computer seem like it's haunted

00:09:14   and say, "Oh, you have a serious problem here.

00:09:15   If you pay us $5 a month,

00:09:17   we'll make sure your computer stays clean."

00:09:19   So it's a slightly different kind of fraudulent scam

00:09:22   than the kind that installs a rootkit or malware.

00:09:24   But you can never know which one of those things you're doing.

00:09:27   And my mother hung up on this person

00:09:31   before finding out exactly what the scam was.

00:09:33   And so wiping her computer was the best thing to do.

00:09:35   But anyway, the story recently was that Bing, Microsoft's

00:09:39   still existing apparently competitor to Google search,

00:09:43   is now banning all third party tech support services

00:09:47   from Bing ads, like the ads they serve

00:09:49   with the search results,

00:09:50   just because so many of them are fraudulent.

00:09:52   So there's an entire category of business

00:09:54   that cannot advertise on Bing.

00:09:56   Now again, advertising on Bing

00:09:57   maybe is not the biggest thing in the world,

00:10:00   but Google has done similar things in trying to find,

00:10:03   trying to get businesses that are fraudulent,

00:10:05   stop them from buying keywords or ads,

00:10:07   because as we all know in these search results,

00:10:09   a lot of the times the first few items are actually ads

00:10:11   and not legitimate search results,

00:10:13   and most people I've found are not good at distinguishing

00:10:16   what's an ad, what's not, even though they're clearly labeled or in boxes or whatever.

00:10:21   And I have no doubt that the thing my mother clicked on was not actually a search result,

00:10:25   but an ad because I know she doesn't know the difference between those two things, just

00:10:28   one of the top hits.

00:10:29   All right, so anyway, it's sad that this entire, what could be an entire legitimate category

00:10:33   of businesses of like, "Oh, you have problems with your computer?

00:10:36   There's a market need.

00:10:37   We can help you with your problems."

00:10:38   Instead, they're all just scams.

00:10:39   And so they're just banned completely.

00:10:41   Pretty crappy.

00:10:42   Yep, and a lot of people wrote in with regard to our conversation about podcasting last

00:10:47   episode and many, many tinfoil hats were worn as emails were sent in saying, "Oh, obviously

00:10:53   Apple is the source of that New York Times podcasting story." It's so clear. I didn't

00:11:00   think that was the case, but I don't know, Marco, did you have any thoughts on that?

00:11:05   I mean, it didn't paint Apple in a very good light, so I don't think they would have been

00:11:10   in the source of that story.

00:11:12   Also, things that are Apple-controlled leaks

00:11:15   tend to read a certain way.

00:11:17   Also, I can't recall a time

00:11:19   that the New York Times was publishing those.

00:11:21   It seems like Apple's controlled leaks in recent years

00:11:25   have gone to the Wall Street Journal and Bloomberg, usually.

00:11:28   So I would not, I think it's pretty unlikely, basically.

00:11:33   - Yeah, I completely agree.

00:11:34   - Yeah, and especially since the idea was that,

00:11:37   like we were saying in the past show,

00:11:39   all these things that they're suggesting would just give Apple a tremendous amount of power

00:11:42   and that's bad for podcasters. And that was what was leading people to say, "Oh, well,

00:11:46   if it gives Apple tremendous power and it's bad for podcasters, stop thinking about it

00:11:49   as, isn't it weird that podcasters would say that? Think about it as, oh, this is exactly

00:11:53   what Apple would want." But if Apple wanted to grab power in the podcast market, A, it

00:11:59   could have done that years ago and B, it doesn't need to leak anything to do it, it would just

00:12:02   have to do it. So I don't see the purpose of that leak, even if the dastardly Apple

00:12:06   did want to own podcasting, which they totally have seemed like they don't, and they don't

00:12:10   need the New York Times to do that.

00:12:11   Michael Scott Yeah, it seems like, you know, what we already

00:12:14   knew, which is that Apple is a very big player in podcasting, but that it's not really

00:12:21   like top of their radar. It's not like a huge priority for them. They have much bigger

00:12:25   things to deal with, and I don't see that changing. You know, like seeing what Apple

00:12:31   has to deal with their other product lines, various market pressures, various internal

00:12:37   and external needs. Podcasting is just not very high ranking on that list and I don't

00:12:44   see that changing for the foreseeable future. It might be way down the road but I really

00:12:48   wouldn't assume that it's going to suddenly become a big thing for them.

00:12:54   We are sponsored tonight by Ring, the Ring Video Doorbell. Go to ring.com/atpnow. There

00:13:00   There are so many reasons to have a video doorbell.

00:13:02   You can actually see who's ringing your doorbell and you can respond from your phone, whether

00:13:07   you're home or not.

00:13:08   So you can pretend like you're home to maybe ward off burglars, you can tell a package

00:13:13   delivery person, "Hey, leave it on the doorstep, I'm in the shower," even if you're at work,

00:13:17   and then you can have a package left.

00:13:19   All sorts of good reasons why you'd want a video doorbell.

00:13:22   There's also incredibly advanced motion detection here, so that even if somebody doesn't ring

00:13:27   your doorbell, Ring can turn itself on and alert you that somebody is at your door or

00:13:33   looking around your door or looking around your porch or wherever you have it aimed.

00:13:37   And this can actually help you both deter and prevent home break-ins. And they've actually

00:13:42   run studies, they've actually been able to decrease the rate of home break-ins in neighborhoods

00:13:47   that have a bunch of Ring video doorbells installed. And home break-ins usually happen

00:13:51   during the day. That's when people aren't home. So this can really help out a lot. Installing

00:13:55   ring is very easy, it takes only minutes, it can either work with your current wiring,

00:14:00   or they have a built-in rechargeable battery model.

00:14:02   So Jon, you have one of these, right?

00:14:03   Well, I installed mine, they sent me one of the "just the camera" ones, that's not the

00:14:07   doorbell thing.

00:14:08   Well that's interesting, what is that?

00:14:09   It's just wrapped up with a mounting bracket that you can stick anywhere and aim up and

00:14:12   aim down, so you don't have to put it where your doorbell will be, and it doesn't have

00:14:15   a button on it, it's basically just a camera.

00:14:17   And I put it, I guess what, facing my driveway, because that's what I wanted to do with these

00:14:21   things.

00:14:22   Have you caught anybody sneaking around your car yet?

00:14:23   No, just us every time we come home and the kids go out and stuff like that.

00:14:29   So I'm still observing it.

00:14:31   I mean, it works fine.

00:14:32   I'm amazed at how long the battery is lasting.

00:14:33   I guess it's basically off or asleep the entire time, and then when you trip the IR sensor,

00:14:37   it turns on.

00:14:38   I think it's still at 90% battery.

00:14:39   It's been out there for five days, and it sends alerts to my iOS devices when there's

00:14:43   motion and I can look at it and I can see my daughter scraping her bicycle against the

00:14:48   side of my car.

00:14:49   It's great.

00:14:50   I kind of feel bad though for my kids because think about it, when you were a kid, like

00:14:54   you could get away with a lot of stuff if your parents didn't see it.

00:14:56   Now I literally have a video camera catching them, you know, scraping their bike against

00:14:59   the side of my car.

00:15:00   I feel kind of bad.

00:15:01   They can't deny it.

00:15:02   I'm like, "Look, I've got video.

00:15:03   Let's come look at it together."

00:15:04   See how the bike's not supposed to touch the car?

00:15:07   Bad.

00:15:08   Yeah, so I feel bad.

00:15:09   But yeah, that's their—that's the non-front door product because, again, after my iPad—iPod

00:15:15   being stolen out of my car, what I really wanted was to see what the deal is.

00:15:19   And so far, no one has come to visit my car, as far as I've been able to determine.

00:15:23   Also my iPod has not been stolen from my car.

00:15:25   So thumbs up so far.

00:15:27   Yeah.

00:15:28   Maybe just the presence of the camera will deter people.

00:15:29   Kind of like those fake security signs that you put on your lawn or on your windows when

00:15:33   you don't actually have a security system.

00:15:34   But this is a real camera.

00:15:36   So it is mostly filming my family now, but it's fun.

00:15:39   I thought a lot about what you were saying about getting away with things, because especially

00:15:44   when it came to like driving. I did terrible things in my 1994, well actually it was Dad's

00:15:50   1994 Saturn SL2 that really was not designed to do the terrible things I did with it.

00:15:55   - Hold that thought, our next sponsor's automatic.

00:15:58   - Yeah, fair enough. But suffice it to say, if Find My Friends or whatever was a thing

00:16:03   or Glimpse was a thing back when I was 16, I would have had a very different childhood

00:16:08   than I did.

00:16:09   So our listeners get free F-Shattered FedEx shipping when you go to ring.com/ATP.

00:16:14   That's ring.com/ATP for free fast shipping.

00:16:18   With the Ring Video Doorbell, you're always home.

00:16:20   Thanks a lot to Ring for sponsoring our show.

00:16:22   [Music]

00:16:23   All right, so this was discussed a bit on the latest under the radar, which I almost

00:16:32   accidentally called "under the weather."

00:16:34   But App Store approval times...

00:16:35   You're always making fun of me for being sick.

00:16:37   I know, seriously.

00:16:38   I actually meant to ask you if you were sick right now. Are you sick right now?

00:16:43   I think it's on its way out.

00:16:44   Okay, so on episode 27 of your Under the Weather program, you discussed fast app review, and

00:16:51   it's worth at least quickly touching upon here. App store review times, which typically

00:16:55   hovered at about a week, are now really, really fast. So Cable Sasser tweeted earlier tonight,

00:17:03   Logan sent a non-Panic app to the Mac App Store at 10.30.

00:17:07   It was in, I'm assuming that's IM,

00:17:09   it was in review at three, rejected for a crash at 5.30.

00:17:12   They fixed it, submitted at 6.30, approved at eight o'clock.

00:17:16   So in the span of, what is that, 10 hours?

00:17:19   They made two submissions and got their app approved.

00:17:22   That's impressive.

00:17:24   - And in fact, I can also confirm that this morning

00:17:27   I submitted an Overcast update,

00:17:29   and it was approved and in the store nine hours later.

00:17:33   I didn't listen to the "Developing Under the Weather" perspective podcast about this, but

00:17:39   when I think about it, a lot of people are asking like, "Oh, does this have to do with

00:17:42   Phil Schiller?

00:17:43   Maybe it does."

00:17:44   So on and so forth.

00:17:45   Like, how is this possible?

00:17:47   And there are so many ways that it could be possible, but the ones that spring to mind

00:17:52   to me are, you can use heuristics to assess the risk of each application submission and

00:18:01   and what you can mix into those heuristics are,

00:18:03   how many apps does this developer put in?

00:18:06   Have we had any problems before?

00:18:08   Is this a entirely new app

00:18:10   or is this an update to an existing one?

00:18:11   I mean, they can go down to the point

00:18:12   where they're doing like binary diffs or something,

00:18:14   but there's a whole mess of heuristics you can do.

00:18:17   And of course they're just hiring more people,

00:18:18   but this dramatic decreased from like days to hours.

00:18:23   I don't think you can do that

00:18:25   by even like hiring 10 times as many people.

00:18:28   They must be having better rules about like,

00:18:32   how much do we have to scrutinize this?

00:18:35   Or how much can we automate?

00:18:36   I mean, increased automation is another thing,

00:18:38   but again, it seems like it would have to,

00:18:39   you know, come out of nowhere.

00:18:41   When a developer can, who's never had any problems,

00:18:44   this is what we've always been talking about.

00:18:45   Your developer who's never had any problems,

00:18:47   who's conscientious, who's submitting an app

00:18:49   that looks fairly straightforward,

00:18:51   to get them through in like this, you know,

00:18:54   one day turnaround time where you pass it,

00:18:56   it gets rejected, you fix it, you pass it,

00:18:57   and it goes back out again,

00:18:59   there's just not enough time in there to do like a,

00:19:02   you know, to give it a huge amount of scrutiny

00:19:05   and go through every single screen and, you know,

00:19:07   do all this stuff.

00:19:08   Even automated testing tools could take, you know,

00:19:10   you could run through automated testing

00:19:12   for a certain amount of time

00:19:13   and maybe that wouldn't even fit in.

00:19:14   This is almost like they're saying,

00:19:16   this one's probably fine, right guys?

00:19:19   Run the fast automated tests,

00:19:21   give it a once over for two seconds and let it sail through,

00:19:23   which is what developers have been asking.

00:19:24   They're like, and again, this wasn't a panic app,

00:19:26   but Panic was saying, look, seriously, Apple,

00:19:29   are we burying malware in our applications?

00:19:31   You're never gonna catch it if we do anyway.

00:19:33   So you might as well just accept the reality

00:19:36   that the only thing you can really do is assume good intent

00:19:40   and punish after the fact, right?

00:19:42   Because if you say, oh, we don't like that,

00:19:43   that's not a secure way to do it.

00:19:45   We wanna stop that stuff from getting to the store

00:19:46   in the first place.

00:19:47   A, that's not possible, and B, it just punishes everybody

00:19:50   for the possibility that they might suddenly turn one day

00:19:52   and Panic becomes infected by malware,

00:19:54   even through no fault of their own.

00:19:55   like say they get hacked and someone sticks Malware into the map, app, there's only so

00:19:58   much of that you can detect in review.

00:20:00   So you know, the calculus may be, you know, develop these heuristics so that most normal

00:20:07   developers get their, we do less, we do less review basically for most normal developers,

00:20:12   and we'll catch them after the fact, because history has proven that you catch tons of

00:20:16   stuff after the fact anyway.

00:20:17   Tons of things get through, even obvious things that should have been caught.

00:20:20   So why pretend that your entire system is predicated on the idea that we must catch

00:20:25   bad things before they get through the store at all costs and then take a day or a week

00:20:29   or whatever to do it.

00:20:30   So this is definitely awesome and I have to think that it has been, that it's happening

00:20:36   because they're just getting, they're working smarter not harder as they say, as the evil

00:20:41   bosses say.

00:20:43   Whatever it is, it's dramatic.

00:20:45   And yeah, we talked about it a lot on Under the Radar last week.

00:20:49   the gist of it basically, I mean, for those of you

00:20:51   who don't know, App Review used to take roughly a week,

00:20:54   and throughout the entire history of the App Store

00:20:57   going back to 2008, there have been a couple

00:21:00   of ups and downs here and there, but for the most part,

00:21:02   it's been pretty consistent at taking about a week.

00:21:04   When App Review goes from taking a week to a day or less,

00:21:08   it makes a lot of App Review problems less severe,

00:21:14   and it makes, it basically lets you iterate faster

00:21:17   in your software.

00:21:18   And so, yes, it does create the potential

00:21:21   to kind of play fast and loose and ship more bugs.

00:21:24   But it also gives you the ability to fix bugs faster.

00:21:27   And so I have to imagine it's going to lead overall

00:21:29   to better quality software.

00:21:32   And it has certainly led to better developer morale

00:21:36   and better developer feelings towards the platform.

00:21:38   And as the economics have gotten more challenging

00:21:41   over the years, I think that's something that Apple

00:21:44   is right to be apparently focusing on because

00:21:48   you know, the app store is, it's pretty easy

00:21:51   to become bitter after a while in the app store

00:21:54   when you're trying to sell something

00:21:54   or trying to make some money 'cause it just gets

00:21:56   harder and harder every year as there's

00:21:58   increasing competition from everywhere.

00:22:00   And so to have signs that Apple is trying to make

00:22:03   our lives better as developers is very promising.

00:22:06   And this kind of thing is a huge improvement

00:22:10   to being an iOS developer.

00:22:11   It's an improvement that I don't think

00:22:12   any of us were expecting to ever get.

00:22:14   And all of a sudden, it's just kind of here.

00:22:16   - It's gonna be on the slide of WWDC, I would imagine.

00:22:18   But yeah, there are still remaining problems.

00:22:21   The main one I can think of is you've been developing

00:22:24   an app for a while, releasing updates to it

00:22:26   on a regular schedule, and then along the way,

00:22:28   if you're lucky, just a routine update,

00:22:30   but if you're unlucky, a bug fix update gets hung up,

00:22:33   as they say, something fundamental about your app

00:22:36   is against the rules.

00:22:36   And you're like, I've been releasing this app for a year.

00:22:38   I have, this is like the 17th update,

00:22:40   and all of a sudden, the major feature of my application

00:22:42   as a violation of the guidelines,

00:22:43   and that's preventing me from getting this bug fix update

00:22:45   or this routine update out the door.

00:22:47   Fast review doesn't help that,

00:22:49   because no matter how fast you iterate on that,

00:22:51   you still, the second problem is,

00:22:53   can I connect to a human being who understands the words

00:22:56   that my mouth is making right now?

00:22:58   Like, nothing fundamental has changed about my app.

00:23:01   This app has existed for a long time.

00:23:02   Are you telling me that this app

00:23:05   is no longer welcome in the store?

00:23:06   Tell me now, I will cease development.

00:23:08   Or are you telling me you don't understand something basic

00:23:11   out my app and I need to explain it to you." And those type of situations with

00:23:14   like the wall of silence and just trying to send it back again and again and

00:23:17   getting computerized responses, fast iteration time doesn't help with that

00:23:20   kind of frustration because then you're like blocked on something like "I don't

00:23:22   understand, I'm trying to fix a crashing bug and you're telling me my application

00:23:26   that's been on the store for a year is now like illegal for some..." you know, so

00:23:30   that frustration can still exist with fast iteration time but boy like Marco

00:23:33   said, having the time get faster, whole messes of complaints and the things that

00:23:39   people like gnash their teeth about the App Store,

00:23:42   just go away when the timescales shrink to a single day.

00:23:46   Then it's like a mild annoyance versus like,

00:23:48   I have to plan my entire business around allowing

00:23:51   for a week to a month of review time,

00:23:53   which is just, just destroys your, you know,

00:23:56   your ability to like just go to market to compete,

00:23:59   to serve your customers.

00:24:02   Like if there's a problem with the 1.0

00:24:04   and you got to wait a week for the 1.01,

00:24:06   Oh boy, I really hope this does stick.

00:24:09   - What will be promising is if Apple publicly

00:24:12   acknowledges this in any way at all.

00:24:15   - WWC, man, you think they're not gonna brag about this?

00:24:17   How can they not, this is gonna be a graph.

00:24:19   - Right, but if they don't mention it at all,

00:24:22   it could plausibly be like a fluke.

00:24:25   - It was an accident, someone was just hitting

00:24:27   the approve button right here,

00:24:27   like the little drinking bird from The Simpsons.

00:24:30   - I understood that reference.

00:24:31   - Me too.

00:24:32   - I told you it was from The Simpsons though.

00:24:34   - Yeah, but I remember.

00:24:35   - Again, I keep giving you the references,

00:24:36   - I'm like, oh, is that from the Simpsons?

00:24:38   - Well, you had, I mean,

00:24:39   we know what drinking bird is separately.

00:24:41   - I know, I know, I know.

00:24:42   - So you kind of had to.

00:24:43   - It just sounds weird to me.

00:24:44   (laughing)

00:24:45   - Golly, you're like a show business mom, man.

00:24:48   You're impossible to please.

00:24:49   - No wire hangers.

00:24:51   - That one I didn't get.

00:24:53   - Nope.

00:24:54   - That's not a show business mom though, is it?

00:24:55   Is that, that was a mommy dearest,

00:24:57   but I don't think it was a show biz angle on that.

00:24:58   - Yeah, it went right over my head.

00:25:00   And everything is back to normal again.

00:25:02   - Our next sponsor this week is Automatic.

00:25:06   Now, chances are your car has not fully kept up

00:25:09   with technology.

00:25:10   This is where automatic comes in.

00:25:12   Automatic is a small adapter that turns any car

00:25:14   into a connected smart car.

00:25:17   Just plug in the automatic into the same port

00:25:19   that your mechanic uses to diagnose engine problems.

00:25:21   Usually it's in the footwell area,

00:25:24   and this opens up a world of possibilities.

00:25:26   Now, automatic lets you keep track of your fuel mileage,

00:25:28   vehicle health, lets you expense business trips with one tap

00:25:32   and link your car to the connected devices

00:25:34   that power your life.

00:25:36   So what does that check engine light actually mean?

00:25:39   Automatic can tell you through your phone

00:25:41   before you even have to go to the shop.

00:25:42   It can also integrate with your Nest thermostat

00:25:45   to know when you're home,

00:25:46   and it can provide the answer

00:25:47   to one of life's most common questions.

00:25:49   Dude, where's my car?

00:25:50   Automatic works on nearly every car made after 1996.

00:25:55   It takes only minutes to install

00:25:57   and connects to your iPhone or Android device over Bluetooth.

00:26:00   This gives you real-time performance data,

00:26:03   all the diagnostic stuff I mentioned earlier.

00:26:05   It can even use intelligent coaching

00:26:07   to maximize your fuel economy and reduce wear and tear.

00:26:09   So for example, you can say,

00:26:11   if I am getting fuel economy below this target

00:26:14   that I wanna hit, beep at me or tell me

00:26:18   so that I can adjust the habit I'm doing.

00:26:20   If I just gave my teenage son Casey the car for the night

00:26:25   and Casey decides to drive super fast,

00:26:27   I can have a report of that.

00:26:29   - 1994 Saturn SL2, it never goes fast.

00:26:33   It just goes quicker than the roads intended, that's all.

00:26:36   - All right, they also have a web app dashboard

00:26:39   that provides granular information,

00:26:40   lets you import and export data,

00:26:42   gives you plain English translation

00:26:44   for your check engine light.

00:26:45   They also have this incredible app platform.

00:26:47   Now, when you have smarts and you have connectivity,

00:26:50   you can have integration.

00:26:51   So I mentioned earlier the Nest integration,

00:26:53   there's also integration with UR Mechanic,

00:26:54   FreshBooks, and more.

00:26:56   You can file business expenses with popular apps

00:26:58   like Concur and Expensify, and it works with IFTTT

00:27:01   for even more possibilities to integrate your car

00:27:04   into your digital life.

00:27:05   It supports Apple Watch, it supports Pebble Watches,

00:27:08   and all this is available for no monthly fees

00:27:11   and no subscriptions.

00:27:12   And they never sell your data, that's not their business.

00:27:14   Their business is selling you the automatic device.

00:27:17   So the automatic device is normally 99.95, 100 bucks,

00:27:22   but when you use our special offer code, ATP0315,

00:27:26   you'll save 20%, so that makes it just 80 bucks

00:27:29   with code ATP0315.

00:27:31   So go to automatic.com/ATP and use offer code ATP0315

00:27:36   to save 20% off the regular purchase price.

00:27:38   It gets to you for just 80 bucks.

00:27:40   That's automatic.com/ATP.

00:27:42   Thanks a lot for supporting the show.

00:27:44   - All right, so Google I/O's keynote was today

00:27:51   and I actually watched it, which is the first time

00:27:55   I think I've ever watched a Google I/O keynote,

00:27:56   But now that I'm in this whole brand new world where I work

00:28:00   alongside Android developers, and they were really into it,

00:28:03   and they did what I usually do for Apple keynotes, and we all

00:28:07   piled in a conference room, and they put it on the big

00:28:10   screen, and we watched it together.

00:28:12   And it was pretty good, all in all.

00:28:15   And we'll go through each thing in a moment.

00:28:17   But I was impressed.

00:28:19   I had heard rumblings and descriptions of past Google

00:28:23   bios that they were meandering and long and often boring, much like the tail end of the

00:28:29   last WWDC keynote, and a total disaster, kind of like the end of the last WWDC keynote.

00:28:37   Well, so I'm curious. I was talking about this recently with Tiff. Which do you think

00:28:43   was the worst moment in recent Apple keynote memory? Was it the Apple Music introduction

00:28:49   the U2 album "Finger Touch" thing.

00:28:52   - Well "Finger Touch" was nothing compared to the

00:28:55   end of last year's WRC, are you kidding?

00:28:57   "Finger Touch" was a brief moment, kinda silly,

00:28:59   excusable by celebrity, the other was just in terminal

00:29:02   watching Eddie Q vamp and then having to have Drake

00:29:05   be up there and just ramble and it was bad.

00:29:09   - But so keep in mind though, I think contributing

00:29:11   to the other side of the argument though is like,

00:29:13   the "Finger Touch" was part of a very, very awkward moment

00:29:17   That then also led into this giant thing nobody wanted, which is like everybody getting this

00:29:23   album pushed into their...

00:29:25   It's fine.

00:29:26   It was fine.

00:29:27   You like you too!

00:29:28   It doesn't matter, and even if you don't, it was fine.

00:29:31   It was a fumble, but who cares?

00:29:33   They gave you free stuff and it took up space in your phone and they gave you a way to opt

00:29:36   out of it and it was silly, but you're talking about the keynote itself.

00:29:39   Finger touch is nothing.

00:29:40   I don't know, man.

00:29:41   I'm kind of with Marco on this.

00:29:42   I think it was an approximately equivalent amount of awkward.

00:29:46   It's just that the density in the finger touch was considerable.

00:29:49   If we had to make you watch one of those things over again, if you had to watch the finger

00:29:53   touch, you'd have to watch the entire Eddie Q and Drake segment, which you'd choose.

00:29:56   Would I have to watch the YouTube performance before the finger touch?

00:29:59   Uh, uh, maybe.

00:30:02   I think it's included.

00:30:03   I don't know.

00:30:04   If you're including the musical performance, there's been a long line of musical performance

00:30:07   that people don't care about.

00:30:08   I was trying—I put all these notes in here about Google I/O because I was trying not

00:30:11   to talk about the showmanship or compare it to Apple because the outdoor thing, the set

00:30:16   decoration on that stage was not good.

00:30:18   Well, hold on.

00:30:19   Can we paint a word picture, please?

00:30:22   I don't want to paint the word picture.

00:30:23   I got to bring up the – I thought I had the video out here somewhere.

00:30:25   I don't know what their decorating theme was.

00:30:27   Was it like –

00:30:28   Neither did they.

00:30:29   – material chaos?

00:30:30   Was that the theme of the stage?

00:30:31   Because I can see the material design colors kind of.

00:30:35   I can see that theme.

00:30:37   And you kind of saw it echoed in their slides.

00:30:38   But the set design for the stage just was not aesthetically pleasing and didn't enhance

00:30:43   -- they had these -- enhance their presentation.

00:30:45   They had these bright colored lights underneath the screen.

00:30:48   Like, I just want to focus on your content.

00:30:50   The content I thought was pretty good.

00:30:52   And I really liked a lot of the things they announced, but the set duration was great.

00:30:55   But anyway, my point is, I don't want to talk about the set decoration or how polished the

00:30:59   presenters were or whether or not Drake wore a vintage Apple jacket while he said nothing

00:31:04   of interest or if Eddy Cue danced.

00:31:06   You have to give Google that.

00:31:07   - Oh, he did. - That other people tried

00:31:08   to dance, right? - He did.

00:31:10   - Anyway, I wanna talk about the products

00:31:13   'cause I thought there was a lot of cool stuff.

00:31:15   - Yeah, I mean, there was Siri, there was the Echo,

00:31:18   there was FaceTime. - Oh, don't even try,

00:31:19   don't even try to say they're just doing things

00:31:21   that other people are doing

00:31:22   'cause that is a ridiculous argument.

00:31:25   The only thing I'd give you that on is split screen

00:31:27   'cause I was really expecting them to have something more

00:31:29   than exactly what Apple has, split screen,

00:31:31   picture in picture, but anyway,

00:31:32   we're getting ahead of ourselves.

00:31:33   - Well, but yeah, that's not an exact Apple rip.

00:31:36   I actually, I give them an okay on that.

00:31:39   But we will go in the order of the show notes,

00:31:42   and it begins with Google Home,

00:31:44   which is the Amazon Echo/Alexa product, but Google.

00:31:49   - See, this actually, like, I'm,

00:31:52   this might be really good.

00:31:54   It's obviously a big risk, you know,

00:31:55   because first of all, it's Google doing its own hardware,

00:31:58   and their record for that is about as good

00:32:01   as Amazon's record is for Amazon doing its own hardware.

00:32:04   So like they have had some things that have worked.

00:32:06   Some of the Nexus's have been decent,

00:32:08   but a lot of their stuff has been flops.

00:32:11   And so it's a huge wild card.

00:32:15   - Let's separate the flop from as in

00:32:16   they don't sell a lot of them,

00:32:18   from is the hardware actually good?

00:32:21   Because I think Google has probably done multiples of both.

00:32:24   Like I think that ball thing,

00:32:25   I don't know if it ever shipped or whatever,

00:32:26   that was like not a good idea.

00:32:27   But a lot of the things they make don't sell a lot,

00:32:31   but it's not like they're bad.

00:32:33   I mean, I don't think they sell, I don't know, even like the Chromebook Pixel.

00:32:36   How many Chromebook Pixels did they sell?

00:32:37   I don't know.

00:32:38   But the hardware wasn't terrible, right?

00:32:41   And –

00:32:42   No, not at all.

00:32:43   I thought the hardware looked great.

00:32:44   And the same thing – and, you know, the tablets and all the Nexus phones, especially

00:32:47   like the Nexus phones didn't sell that many, you know, I think Google thought they were

00:32:51   going to sell more than they did and then the other vendors took over.

00:32:54   But the hardware they actually make, I don't think is bad hardware.

00:32:58   have some vague hope that this little squat little weebl that's their Amazon Echo competitor.

00:33:05   Everything is a weebl to you! It's shaped like a weebl. They wobble, they don't fall down, right?

00:33:10   How many times have you made that joke on our shows?

00:33:12   But people need to know about the weebls. Please don't make me tell you again about the weebls.

00:33:17   But no, this is literally weebl shaped, right? It's fat on the bottom and it's like... Anyway.

00:33:24   I have some faith that the hardware will be okay.

00:33:28   I don't think that's been the problem with Google's hardware products is that the hardware

00:33:32   is bad.

00:33:33   I feel like they've just fallen down another route, like the Google TV stuff.

00:33:37   Maybe that ridiculous remote, was it made badly?

00:33:40   No, it was just a bad idea.

00:33:42   So if it's a good idea, then I feel like they can make it reasonably well.

00:33:47   I think I have seen no reason why this device shouldn't sell just as well as the Amazon

00:33:54   Echo, unless the Amazon Echo has already tied up that entire market, which is possible.

00:33:57   But other than that, the people who bought the Echo, why wouldn't they buy this too?

00:34:00   It's like the same thing.

00:34:02   Yeah, I mean, really, this kind of thing is the kind of thing that Google should be really

00:34:09   good at. And again, this isn't to say that they definitely will, but I think they clearly

00:34:15   are likely to succeed here, because it combines the types of things they're really good

00:34:20   at with the kind of market that they can get in and succeed in, which is so far mostly

00:34:25   like a geeky market, a very tech savvy market of people who have smart home stuff. The main

00:34:31   problem they're going to have is the lack of retail presence. Because we've seen,

00:34:37   Amazon is so powerful. Amazon is the place where most people would go to buy this kind

00:34:44   of thing. And I'm sure Amazon is not going to be too happy to carry it, or if they do

00:34:49   carry it, they're certainly not going to ever promote it. But I'm guessing they won't carry

00:34:52   it at all. So this is going to be interesting to see how Google can try to market this and

00:35:01   sell the crap out of this. Because from a technical perspective, Google should be better

00:35:06   than Amazon at this stuff. I don't know that they will be, but they should be.

00:35:09   - Yeah, I expect them to be better. Because the only thing I feel like they might not

00:35:14   be better at is maybe this is their first try at putting something out that has 10 microphones

00:35:19   and all sorts of noise canceling stuff.

00:35:20   So maybe Amazon –

00:35:21   Well, it was Amazon's first try and they got it right.

00:35:23   I know, I know.

00:35:24   But it's been out for a while and the quality is really good.

00:35:27   So I think basically that Amazon hit it out of the park in that particular feature, right?

00:35:31   So I don't think that's a given for your first try.

00:35:33   I think Amazon did a really good job with does this thing understand me from across

00:35:37   the room.

00:35:38   Google says they do a good job with that, but they – like Amazon had an exemplary

00:35:41   first try, right?

00:35:42   Then they have the second and third try with the dot and all that other, whatever, the

00:35:46   derivative products.

00:35:47   But I think they did a really good job.

00:35:48   But other than that, everything else about this, Google, I think, should be better and

00:35:53   will be better.

00:35:54   Like, things that you can say to it, natural language processing, backend querying, the

00:35:58   only thing Amazon's going to be better at is buying stuff, right?

00:36:00   Because they have Amazon.

00:36:01   But every other part of it, Google damn well better be better than Amazon at it because,

00:36:06   you know, Amazon doesn't have any of the strength behind it to do all the language processing

00:36:11   and search querying and the deep hooks into all that type of stuff. And the thing that

00:36:17   really blew me away about this demo was, well, A, thinking about why the hell does Apple

00:36:21   not have one of these things yet, I guess maybe it's beneath their concern. But B, when

00:36:24   they did the thing like the kids talking to it and doing the conversation and says, "Show

00:36:29   me on the TV," it's like, that is an Apple move. How can we do that? Because we make

00:36:33   a TV device. And of course our little weebol knows about the TV, the Chromecast attached

00:36:37   the TV, show me on the TV. That is the type of thing we used to have to be in the Apple

00:36:42   ecosystem for, which is, oh, I have tons of Apple crap around the house, and they all

00:36:45   know about each other, and they all just work together. And it's like AirPlay to your Apple

00:36:50   TV the first time that worked. It was like, oh, this is amazing, because I have these

00:36:52   two Apple devices, they're really easy to put video up on my TV.

00:36:55   And it's the only time that worked.

00:36:57   Yeah, I know.

00:36:58   I'm just kidding. I'm kidding.

00:36:59   But that was the first time. It was impressive the first time. And this is Google saying,

00:37:03   "Hey, we actually have stuff around here."

00:37:06   And they put the little Nest icon on there to say "Connect to your thermostat."

00:37:10   And third-party, integrate with your third-party lights and stuff like that.

00:37:13   They're not being like Apple, where it would only work with your Apple lights or whatever.

00:37:16   They're all big on third-party integration.

00:37:18   They already announced an API, which we still don't have for Siri for developers.

00:37:24   Unless Google Home, unless this is the Nest Hub that we read about in all those tell-all

00:37:28   articles about Nest, about how Nest was going down in flames and everything was breaking,

00:37:31   I really hope this is not the long delayed hub from the Nest people because a dysfunctional

00:37:36   organization is not going to make a good product.

00:37:38   But assuming this is made by a functioning organization, I think this will be an impressive

00:37:43   product.

00:37:44   And I think I might actually get one.

00:37:45   I've been holding off on the Echo just because it's a first-generation product or whatever.

00:37:49   But if this gets really good reviews, I think the reason I will buy it is because basically

00:37:52   I have more faith in Google than Amazon in terms of supporting and evolving this product

00:37:59   and being good at the more interesting, sophisticated parts of it.

00:38:03   The downside though, I mean, the counter-argument to that, first of all, there is the big privacy

00:38:07   question which we should address because a lot of people, like, you know, I was one of

00:38:12   these people. I freaked out when Amazon released the Echo and I was like, "You're going

00:38:16   to let Amazon put a speaker and a microphone in your house listening to everything you

00:38:21   say all the time?" And similarly, you know, when Google bought Nest, everyone's like,

00:38:25   you want Google to own this thermostat

00:38:27   that can look in your house and stuff.

00:38:29   And there's a lot of people, my past self included,

00:38:33   who are uncomfortable with that

00:38:34   and they're just not gonna want that product in their house.

00:38:37   And that's fine, that's your decision.

00:38:39   For me, what tipped the scale is what every Google fan

00:38:44   usually says when asked about privacy things,

00:38:46   which is basically what tipped the scale for me

00:38:47   with letting the Amazon Echo in my house was,

00:38:50   I saw how good it was at other people's houses.

00:38:53   I saw how good it was and I decided, you know what, it's worth the tradeoff. I'll take

00:38:59   the risk, I'll put this thing in my kitchen because it is so good that it's worth the

00:39:03   tradeoff. And this is the same thing that most Google people say when they say, you

00:39:08   know, if people like me ask them like, you know, why do you want Google to have access

00:39:12   to all your data and to be analyzing everything you do and selling ads against it and everything.

00:39:17   And the answer I get most commonly is, well, it's worth it to me because it's convenient

00:39:21   and these features are things that I want, so I've made that decision. And, you know,

00:39:24   I can't really argue with that. If, you know, if you decide it's worth it to you, then it's

00:39:26   worth it to you.

00:39:27   I have just to clarify where I fall on that. I don't see it as a trade off as in like,

00:39:35   okay, well, there are detriments, but there are benefits. I mean, I know the trade off

00:39:37   is there, but like, I would put it more succinctly for myself is that basically, I trust Google.

00:39:43   Like, I don't trust them with everything. I don't trust them like implicitly. I don't

00:39:46   give them all of my trust. But in the grand scheme of things, I trust Apple and I trust

00:39:50   Google to basically be companies that are trying to do the right thing and are not like

00:39:56   just inherently evil like I don't know insert the name of your favorite company although

00:39:59   it's weird that I was just about to say Oracle but don't read anything into that. Anyway

00:40:03   I mean I would trust Google before I trust Facebook. Oh yeah it's a great example right

00:40:07   or even I trust Amazon you know for the most part and again I don't trust them entirely

00:40:11   you have to be vigilant and so on and so forth don't say oh I just trust them to listen to

00:40:15   everything in my house they're never gonna do anything evil never gonna sell me the advertisers

00:40:17   is like I trust them in that I kind of,

00:40:18   I know what they're gonna do, I know what they're about,

00:40:20   but in general, still the company seems to me

00:40:23   to be mostly trying to do the right thing,

00:40:25   and as long as you understand their business

00:40:26   and how much advertising is a part of it

00:40:28   and what is probably gonna happen to all your data,

00:40:30   and the fact that the US government is probably

00:40:33   going to have hooks into all this stuff

00:40:34   and they're not gonna tell you about,

00:40:35   like as long as you go on with a clear head,

00:40:37   the bottom line is that I trust that this is not,

00:40:41   I'm not like back dooring my entire house forever and ever.

00:40:45   and I wouldn't install a device from a company

00:40:48   that I trusted less.

00:40:49   Or a company that was less competent

00:40:51   or more likely to go out of business

00:40:52   or more likely to do something in desperation for money,

00:40:56   you know what I mean?

00:40:57   There's so many things for me in favor of Google.

00:40:59   So yes, it is a trade-off,

00:41:00   but mostly what it comes down to is that

00:41:02   I trust Google enough to put something like this

00:41:06   in my house.

00:41:07   - Yeah, you know, I spoke about, I think,

00:41:11   that I am a brand new Google Photos user.

00:41:14   And the more I use this app, both the web app and the native apps, the more I like it

00:41:19   and the more it amazes me.

00:41:21   And quickly during the keynote they said, "Oh, and you know, Google Photos does some

00:41:26   incredible things."

00:41:27   And I think it's what, 200 million active users on it or something like that, whatever

00:41:30   the number is.

00:41:31   But you can even search for something like hugs.

00:41:33   And of course in the room I'm in it was half Android developers, or mostly Android developers

00:41:36   and a couple of others.

00:41:38   And so I immediately hopped on Google Photos and typed in hugs.

00:41:40   And sure enough, there are a bunch of people hugging each other, like usually Aaron and

00:41:44   me and Declan, that show up immediately.

00:41:47   Like the stuff that Google Photos does is amazing.

00:41:50   And seeing how good it is makes me wonder, man, if they're this good with photos, I wonder

00:41:57   how good they would be with this Google Home thing.

00:41:59   So I agree with you, Jon, that this, the Google Photos has kind of been my gateway drug back

00:42:05   into Google.

00:42:06   use Gmail for both work and personal mail, but that's just kind of a thing in the background.

00:42:11   I don't ever use the web app, it's all just basically IMAP to me. But man, Google Photos

00:42:16   has really changed my tune as to how I think about Google these days. And I'm giving some

00:42:20   serious positive side-eye to this Google Home thing because it's aesthetically pleasing,

00:42:27   be it a Weebl or not, it still looks good. And I'm tentatively interested in what this

00:42:34   springs in a way that I haven't really been in the echo.

00:42:37   And not to bring this back to food again, but one of the demos they did show was talking

00:42:45   to, maybe this was, yeah, I think this was in this one, talking about, I think it gets

00:42:48   into the aloe thing later, but it's all tied up in the same sort of natural language processing

00:42:54   type where they're asking questions of these products and having it do things for you.

00:42:58   And they always want to demo those because they always want to show, I think they had

00:43:00   the video of an entire family talking to the Google Home device, which is a little bit

00:43:05   overblown because the impression you get is this entire family spends its entire day living

00:43:09   in the same house, but all their words are addressed towards the inanimate object instead

00:43:14   of the other people.

00:43:15   Which, you know, you kind of get that because you have to fit in lots of examples of uses

00:43:20   in a single two-minute commercial.

00:43:23   And anyway, that aside, a lot of what the things they're doing is trying to sell the

00:43:28   a fantasy that it's like a personal assistant

00:43:30   where flight is delayed and you wanna move

00:43:33   your dinner reservations and you would just want it

00:43:36   to handle things for you.

00:43:38   And I think those make for good demos, but also bad demos,

00:43:43   because I think they raise expectations on reasonable levels

00:43:46   because every time they do something like that,

00:43:47   I see sort of like scrolling down the side

00:43:49   of my virtual screen, like all the cascading number

00:43:52   of assumptions that are in there,

00:43:53   that your restaurant has an open table reservation,

00:43:55   that it understands that, you know,

00:43:57   that the restaurant thing you're talking about,

00:43:59   that it knows about your flight,

00:44:01   that you didn't use a different email address for this,

00:44:03   that it wasn't sent to a different person.

00:44:05   So many things have to go right for that to work,

00:44:10   and they make it seem like you don't have to worry about

00:44:12   or care about those, but you do.

00:44:13   Because if you're flying on an airline

00:44:14   that doesn't have integration in the same way,

00:44:16   or you want to go to a restaurant

00:44:17   that doesn't support OpenTable,

00:44:19   or something was done on a spouse's account and not yours,

00:44:22   so you're talking to it, but it doesn't understand,

00:44:25   it doesn't know anything about that flight

00:44:26   because it's done through a different Google ID or, you know, like, so many possible things

00:44:31   can go wrong to make that not work. And when they do go wrong, even in the slightest way,

00:44:35   it's like the old, you know, give up and use tables website. It's like, if anything goes

00:44:39   wrong at all, all right, forget it. I'll just do it on my phone or I'll sit down in front

00:44:41   of the computer because there's no way I'm trying to have an argument and getting into

00:44:44   the text adventure syntax game with this thing. So it all looks magical. And the same thing

00:44:49   with like ordering food. Oh, the food will be waiting for you to get home. Food from

00:44:52   where? You want curry? It found it in your restaurant. You're just accepting the restaurant

00:44:56   that it gave you, what if you don't like that restaurant?

00:44:57   Does it learn which restaurant you like?

00:44:59   Does it like, there's still, this is so primitive

00:45:01   where it's not, I don't feel like it's going to help me

00:45:04   with my life unless I just don't care

00:45:06   where things come from, want the most generic things,

00:45:08   like I wanted to know.

00:45:10   I always get Indian food from this place.

00:45:12   Or if it's not sure, or they should ask me,

00:45:14   do you wanna get it from the place you got it last time?

00:45:16   And this is all predicated on the fact

00:45:17   that it can order from those places.

00:45:18   You probably can't because your favorite Indian place

00:45:20   has no idea what computers are and they only take cash.

00:45:22   And it just doesn't, we're not there yet.

00:45:26   And these ads make it seem like they are there,

00:45:27   but I don't know anybody except like Mike Mattis maybe,

00:45:30   who can actually live this life

00:45:32   like where you just talk into the air

00:45:35   and everything you do is exactly integrated.

00:45:37   And well, he wouldn't like either

00:45:38   because he would have to know exactly

00:45:39   where everything's coming from.

00:45:40   You can't just say, "Order me Indian food."

00:45:42   Is this safe, Casey?

00:45:43   Can we talk about Indian food?

00:45:44   You can't just say, "Order me Indian food,"

00:45:45   and they'll be waiting for you and you'll be happy?

00:45:47   Order what?

00:45:48   What do you want to order from where?

00:45:49   Like, it just, it grinds my gears to see that.

00:45:53   But anyway, I do have confidence that Google is going to be better at this natural language

00:45:57   stuff than Amazon just because they have so many smart people doing this and Amazon probably

00:46:02   has a slightly smaller number of smart people doing this.

00:46:05   Well, but Amazon has, I think, two big strengths. I mean, number one, first of all, as I mentioned,

00:46:10   they have massive retail power here to push these things. And by all accounts, the Echo

00:46:16   probably already has a decent-sized install base. So that's barrier number one for Google

00:46:22   to try to overcome.

00:46:23   You can overcome the head start the Echo has gotten

00:46:27   with a good product, but it's gonna be really hard

00:46:29   to overcome the massive retail and promotional advantage.

00:46:33   Just ask anybody who has tried to make a successful,

00:46:37   technically advanced e-reader in the last eight years or so.

00:46:42   - But Google has a big in there, because Google Home,

00:46:45   who cares if the little turd, that's now a turd,

00:46:48   downgraded from Weebill, ever sells.

00:46:50   'cause all the same thing is all powered

00:46:52   by the whole Google Assistant thing,

00:46:53   and that's gonna be on all their phones.

00:46:55   And they have a ton of phones,

00:46:57   so that is their wedge basically.

00:46:58   Like look, maybe the home never beats out the Echo,

00:47:01   but if we can make this a thing on all of our phones

00:47:04   so that the 50% plus of the world

00:47:07   that has Android smartphones gets used to using this

00:47:09   on their phones, then they'll still win.

00:47:12   Because again, the phone is gonna swamp

00:47:14   how many people bought a little silly cylinder.

00:47:16   Even if people are just shouting across the room

00:47:17   to their phones and people started

00:47:18   adding multiple microphones on phones.

00:47:19   So you're right that it is a barrier, but Google already has like a beachhead there.

00:47:24   They have a way around. Like I think Google cares less whether Google Home succeeds than

00:47:29   whether this assistive technology becomes sort of what Google is known for in the modern

00:47:32   era instead of just web search.

00:47:34   **Ezra Klein:** That's fair. Well, so the other, I think, advantage that Amazon has

00:47:39   right now is that the Echo is not trying to be that kind of like super smart fantasy California

00:47:46   you land, you know, order me a new food and it just magically does exactly the right thing

00:47:50   you want. Like, the Echo is more like a really simple command line and once you learn the

00:47:57   relatively small vocabulary and syntax that it supports, it's incredibly reliable doing

00:48:05   those things. And so, I think while it might at first have a slightly higher learning curve

00:48:11   for like day one, two, three, I think once you get past

00:48:16   the very, very initial part of it,

00:48:18   I would say the Echo's actually easier to use,

00:48:19   'cause once you figure out the kind of things

00:48:21   that work with it, those things work incredibly reliably.

00:48:25   And so that's the challenge that anybody has

00:48:27   coming into this, like I think Siri doesn't do as well

00:48:29   at that, because Siri tries to do more.

00:48:31   - Doesn't have an API.

00:48:32   - That too, Siri tries to do more,

00:48:35   but it doesn't really do any of them reliably enough,

00:48:40   and it's hard to know before you ask Siri something,

00:48:44   it's hard to know whether you will succeed or not.

00:48:47   Whereas with the Echo, you figure it out within a few days.

00:48:51   You figure out, okay, this is the kind of thing

00:48:52   that will succeed, this is the kind of thing that won't.

00:48:54   Google is obviously trying to be very ambitious

00:48:58   with the kinds of things that their thing can do,

00:49:01   that Google Home can do.

00:49:02   We will see if it works.

00:49:04   I think if anybody can do that kind of complexity,

00:49:06   it's them, so it might work.

00:49:10   To me, I think the biggest risk for buying into the Google Home ecosystem is whether

00:49:16   Google themselves will lose interest or it will fail within a few years.

00:49:23   Their track record for that isn't great either.

00:49:26   If something is not working, they're the first ones to kill it usually.

00:49:30   Just don't tell them that FeedBurner is still running because I think they forgot about

00:49:33   it.

00:49:34   Anyway, Google is, if you look at the history

00:49:39   of various initiatives they've had,

00:49:42   various big platforms they've tried to launch,

00:49:45   it's littered, it's a huge graveyard

00:49:48   of stuff they've shut down.

00:49:49   - But that's no different than Amazon.

00:49:51   - Well, that's true, but if they try this out,

00:49:54   if it doesn't get very far in the market,

00:49:56   they could choose to fight harder

00:50:00   and to keep it going or to shut it down.

00:50:02   And if you bought the wrong hardware

00:50:05   and they shut yours down, that kind of sucks.

00:50:07   It's kind of like a format war going on.

00:50:09   - But don't you think this has a big advantage

00:50:11   in that the underlying technology,

00:50:13   whether it's underlying this and Allo and all that stuff,

00:50:15   the underlying machine intelligence,

00:50:18   natural language processing, speech interface,

00:50:21   that I feel like is a core technological effort at Google

00:50:25   that is not going away,

00:50:27   like 100% guaranteed not going away.

00:50:29   Now the individual products, you're right,

00:50:31   Maybe they get to Google Home and it ends up like,

00:50:33   oh, we have the second version and it's not compatible

00:50:35   and you're stuck with some bad hardware

00:50:36   and they eventually stop supporting it.

00:50:37   I could totally see that happening.

00:50:39   That's a danger in any sort of product like this.

00:50:41   And they're, historically, like I said,

00:50:42   have not been particularly good about preserving that.

00:50:44   But if you're gonna have any faith

00:50:47   in any kind of product efforts,

00:50:48   it's not going to be like the weather balloons

00:50:50   that give you wifi or the self-driving cars.

00:50:53   It's gonna be the natural extension

00:50:55   of basically Google's core product, which is search,

00:50:58   which is taking that to the next level.

00:51:00   Also because it ties into advertising,

00:51:02   if you're gonna be honest.

00:51:03   Why would Google stick with this type of effort?

00:51:05   A, they've been doing it for years, right?

00:51:07   And B, it totally fits with both web search and advertising.

00:51:11   So I have to think that this effort will continue to go on.

00:51:15   And if these products are like,

00:51:17   that they will iterate on them,

00:51:18   that they will make future versions of them

00:51:20   and will keep going.

00:51:21   And there is still a slim chance

00:51:22   that you buy some hardware, it might be orphaned

00:51:24   'cause they bail on that and have a new iteration

00:51:26   with a new name or whatever.

00:51:27   But I really think this is not esoteric

00:51:32   or in fear of being a flash in the pan

00:51:35   simply because it just reads so much as Google to me

00:51:38   and not tangential.

00:51:39   Like I feel like this is, if you were to say,

00:51:41   what does Google look like 50 years in the future?

00:51:43   It looks less like typing things in text boxes

00:51:45   and more like a much better version of this

00:51:47   because it's just such a natural extension.

00:51:49   - Yeah, and certainly, what was it?

00:51:51   Machine learning was their version of customer sat.

00:51:55   Like if we were doing a drinking game

00:51:57   and machine learning caused you to drink,

00:51:59   you would have been under the table

00:52:00   after the first 20 minutes,

00:52:02   because that's all we heard about.

00:52:03   And I mean, Apple has its own foibles

00:52:05   in very similar ways like customer sat,

00:52:07   but machine learning was all over this keynote.

00:52:09   And I agree with you, Jon,

00:52:10   that that seems to be where they're pushing as a company

00:52:13   is trying to leverage that machine learning

00:52:16   in any possible way that they can.

00:52:18   And in some of those ways, like I've been talking about

00:52:20   with Google Photos and with Google Home,

00:52:22   I think it looks really promising

00:52:24   and really, really interesting.

00:52:27   And like it's their logo or their motto or vision statement, whatever, like organizing

00:52:32   the world's information. They've been slowly but steadily doing that over the course of many,

00:52:37   many years trying to get semantics into the information so it understands not just like,

00:52:41   oh, this word appears there and these people link to that page, but understanding what the

00:52:45   information is that it's actually looking at. That's how it can do that photo stuff.

00:52:48   That photo stuff doesn't just come out of nowhere because they did a one-year project to do it.

00:52:52   that's based on years and years of research outside Google and within, and just working

00:52:57   on it, working on their language parsing, they'd open source that big, you know, natural

00:53:01   language processing thing, and image recognition and stuff with robotics.

00:53:05   This is all about taking in information and then developing an understanding of it that

00:53:11   can be encoded by computers so the computers can act on it.

00:53:14   So it's not just text so that they understand the meaning behind things.

00:53:18   once you have even the barest meaning or the barest sort of sensory perception of like,

00:53:23   "Is this a car? Is this a hug? What language is this in? What do the words mean?" That

00:53:28   previous sentence when they said "he" in the next sentence, "What are they talking about?"

00:53:32   It's really basic stuff conceptually, but it's really hard to do for computers, and

00:53:37   they're doing it on such a massive scale that that effort and that research is going to

00:53:42   underlie all their products from here going forward. And maybe they're going to hit that

00:53:46   inflection point where suddenly it becomes acceptable and it passes the barrier from

00:53:52   tech curiosity that we're all impressed by but that only nerds really use to just a normal

00:53:57   thing that everybody does.

00:53:58   We may be getting close to that, if only because people don't like typing stuff.

00:54:03   I saw someone tweeting when the I/O thing was going on of how many people they'll see

00:54:09   interacting with their Android phones or their iOS devices purely by speech, even though

00:54:15   a computer nerd looking at them would just cringe at how incredibly inefficient it is

00:54:19   to essentially be arguing with your phone and saying things over and over again and

00:54:23   trying to type things, "Just give me your phone!

00:54:25   Let me show you how to do this!"

00:54:26   But people prefer it, even when it's totally broken and crappy and doesn't work right,

00:54:32   just because it's more comfortable for them.

00:54:33   So I think we're probably closer to the inflection point where most people use what will continue

00:54:38   to be the most massively inefficient way to do anything, especially when it has any sort

00:54:42   of error simply because it's just more comfortable for them than trying to figure out what things

00:54:47   to tap on or what to type.

00:54:50   Our final sponsor this week is Pingdom.

00:54:53   Start monitoring your websites and servers today at Pingdom.com/ATP.

00:54:58   Pingdom makes the web faster and more reliable for everyone by offering powerful, easy to

00:55:01   use monitoring tools and services for anybody with a website.

00:55:05   By using Pingdom you can, for example, monitor availability and performance on your server,

00:55:09   database or website from more than 70 global test servers.

00:55:12   They can emulate visits to your site,

00:55:14   check its availability as often as every minute.

00:55:16   You can monitor the availability of key interactions

00:55:18   such as contact forms, e-commerce checkouts,

00:55:21   logging in, searching, and a lot more.

00:55:24   I have used Pingdom to monitor the stuff I do since 2007.

00:55:28   All Tumblr, all of Instapaper, all of Overcast,

00:55:32   and market.org the entire time.

00:55:33   It is great.

00:55:34   It alerts me whenever anything is out.

00:55:36   I usually do two alert levels.

00:55:39   I do one called vitals, which is,

00:55:42   I have all the servers report their disk space

00:55:44   and free memory and everything,

00:55:46   and so then I have a low alert level for,

00:55:49   oh, you know what, one of the five web servers

00:55:52   is not responding.

00:55:53   Or this server has very low disk space or something.

00:55:57   And that just sends me emails and notifications.

00:56:00   And then I have another alert level.

00:56:01   If the whole site is down, that sets off the alarm bells.

00:56:04   That sends me text messages.

00:56:06   that is like loud notifications.

00:56:09   It's like the different level of alerts with Pingdom.

00:56:11   I've also used it in the past to do things like

00:56:14   monitor the Apple WWDC website

00:56:17   to see when WWDC was announced every year

00:56:20   because it would monitor the page for changes.

00:56:22   So you don't actually have to own the sites

00:56:24   you're monitoring with Pingdom.

00:56:25   It's great.

00:56:26   I recommend Pingdom.

00:56:27   I use it myself.

00:56:28   I have used it literally since 2007.

00:56:30   So I've used it now for nine years

00:56:33   and I will continue using it because it is awesome.

00:56:36   They used it before they were a sponsor.

00:56:37   Now they are a sponsor, that's amazing.

00:56:38   So anyway, give Pingdom a URL to monitor

00:56:41   and optional conditions to check for

00:56:43   or to just check whether something's up.

00:56:44   And when they detect an outage,

00:56:45   you are alerted immediately so you can fix the problem

00:56:48   before it becomes a much bigger and more costly outage.

00:56:51   You should not be learning that your site is down

00:56:52   from people on Twitter.

00:56:53   That's embarrassing, it's unprofessional.

00:56:55   Be the first to know with Pingdom

00:56:57   and then you can fix your site

00:56:58   before too many customers or readers see it.

00:57:01   Check it out today, go to Pingdom.com/ATP

00:57:04   for a 14-day free trial and get 20% off your first invoice

00:57:07   by using offer code ATP.

00:57:09   Thanks to Pingdom for sponsoring our show.

00:57:11   (upbeat music)

00:57:13   - Let's talk about "ello, governor."

00:57:16   - Oh, geez.

00:57:18   - You so wanted to say that.

00:57:19   I know you so wanted to say it.

00:57:20   - I did, I did.

00:57:21   - This is gonna be the week of bad podcast British accents.

00:57:24   - Hey, at least they didn't put an apostrophe before the A.

00:57:27   - That's true, that's true.

00:57:29   - Although it reminds me a lot of "ello,"

00:57:31   that weird social network that didn't go anywhere.

00:57:33   - It still spams me every few days with something.

00:57:36   - Yeah, you gotta unsubscribe for all those.

00:57:38   - I have.

00:57:39   - Anyway, so Google Ello is the iMessage knockoff,

00:57:46   but with a lot more.

00:57:48   It's not just straight--

00:57:48   - It's not an iMessage knockoff,

00:57:50   it's like a line knockoff or a WhatsApp knockoff.

00:57:52   - Yeah, that's fair.

00:57:53   - It's much more like the much more sophisticated

00:57:56   messaging apps that are used outside the Apple ecosystem.

00:57:58   It's got stickers for crying out loud.

00:58:00   - Fair enough.

00:58:03   Yeah, so actually that was a very interesting interaction. So they went, I

00:58:08   forget which person was presenting, was this the British woman that was

00:58:10   presenting at this point? In any case, whoever was presenting was talking

00:58:15   about how there are times when you want to kind of shout in a text message

00:58:21   conversation, there's times where you kind of want to whisper, and so they

00:58:25   allow you to change the font size within Allo, which in and of itself doesn't seem

00:58:30   that impressive, but I thought the interaction was pretty cool. Where on iOS in the messages

00:58:35   app, you have a little microphone icon and you can swipe up and down to like, send an

00:58:39   audio message. What they do is they if you tap and hold roughly where our microphone

00:58:45   icon is, it'll let you kind of drag a slider up and down to change the size of the font,

00:58:51   which I thought again, was a very clever interaction.

00:58:54   So many things wrong with that from a usability perspective that I can't like and from a taste

00:58:58   perspective. Like this is just such a totally a Google thing to do where

00:59:01   Someone thinks is a cool feature and it seems like they don't think quite enough about the implications both

00:59:06   Aesthetically where you can make really ugly look even the demo screenshots were looking really ugly and I was thinking to myself

00:59:12   Not there's anything wrong with ugly and garish like if that's the essay you're going for but I was thinking like

00:59:16   This is the opposite of the Apple aesthetic and it can be more fun. Like the Apple aesthetic can be boring, right?

00:59:23   You know, oh everything is all Helvetica and everything's the same size and everything is sort of clean and button-down

00:59:27   down. Like I, you know, I'm not, I'm not slamming it. I'm really not like, but I,

00:59:30   I saw it. I'm like, wow, this is definitely something that Apple wouldn't do. And

00:59:33   for the shouting, find fun, make your text bigger. It's fun to shout someone.

00:59:37   It's whispering. Not good for usability. You telling me I can make the text

00:59:41   microscopic boy. Yeah. My, my aged relatives will love to squint at that when

00:59:45   I, Oh, you don't understand grandma whispering. No, you can't. I mean, maybe

00:59:49   it's overridable at the local level, but it just doesn't make for a good demo

00:59:53   because the obvious question, which is,

00:59:56   really, I can make text so small

00:59:58   that my recipient can't even read it and has to squint?

01:00:01   That's the question in people's minds,

01:00:03   if they're looking at it from an accessibility perspective.

01:00:04   So just throw a sentence out there

01:00:06   that says the thing that I really hope is true,

01:00:08   which is like, oh, and of course the recipient

01:00:09   can set a minimum font size,

01:00:11   like whatever you have to say,

01:00:12   address the, we'll get to this later

01:00:14   with some more stuff from the knock-knock duo,

01:00:18   what do you call it, video conferencing thing.

01:00:21   Like if there's an obvious,

01:00:22   there should be an obvious question

01:00:23   in most people's minds about the feature you're doing,

01:00:25   throw out a sentence or two to reassure us

01:00:28   that the feature that you surely added

01:00:29   to account for this is there.

01:00:30   Because if you don't say it,

01:00:32   we're left to think that you haven't thought of it.

01:00:34   And even if that's not true, it gives a bad impression.

01:00:36   And in this case, I don't trust Google to think these things

01:00:38   'cause they very often do things that,

01:00:41   the company's persona is that of an individual

01:00:48   that doesn't account for the variety

01:00:52   of different kinds of people and lives that are out there.

01:00:55   And I know they're fighting very hard against that.

01:00:57   I know it's kind of unfair,

01:00:58   but a lot of times I see Google do things

01:01:00   with obvious problems that I assume

01:01:03   that they have accounted for.

01:01:04   Then they release a product and I say,

01:01:05   oh, they didn't account for that.

01:01:07   And in fact, it seems like they didn't even think of that.

01:01:09   And they'll fix it after the fact to their credit.

01:01:10   And it's not like they're trying to do something wrong,

01:01:13   but sometimes they drop the ball.

01:01:14   So I'm saying Google in your presentations

01:01:16   for the foreseeable future,

01:01:17   when you show a feature that has obvious problems

01:01:21   throughout two, Apple does this all the time,

01:01:23   show up two or three words to say,

01:01:24   oh, and of course, don't worry about this

01:01:27   because we handled it in an X and Y way,

01:01:28   like in the obvious way.

01:01:29   'Cause I don't think for me anyway,

01:01:31   they get the past of making me think

01:01:33   that they thought of that.

01:01:33   - I think that's understandable.

01:01:35   We should also talk about the predictive replies

01:01:39   or suggested replies with ding, machine learning, ding.

01:01:42   That seems weird to me.

01:01:47   And I have very conflicting thoughts about this,

01:01:48   which is actually kind of a theme

01:01:49   for most of what I thought of Google I/O.

01:01:52   But it seemed odd to me to have suggested replies

01:01:57   on a phone or a tablet where it's relatively easy to type.

01:02:02   I don't really have a problem with it on the Apple Watch,

01:02:06   partially because it's so terrible,

01:02:07   and partially because you can't exactly type

01:02:11   on an Apple Watch, and we're gonna get to that later

01:02:13   as well, hopefully.

01:02:14   But on a device where there's some reasonable approximation

01:02:19   of a keyboard, it just seems insensitive. Like one of the examples they used was like

01:02:23   a picture of a kid. "Oh, how cute" was one of the replies. Like, I don't know, that just

01:02:27   seems insensitive or cold to me.

01:02:32   So here's an example, a great follow-up from what I said, here's an example where they

01:02:35   did realize what the obvious objection to this feature would be and address it with

01:02:40   a sentence or two. So the obvious objection from the demos is like, I would never tap

01:02:44   any of those circles because all of them sound like inane things, like that's not how I communicate,

01:02:49   right? Just like you were saying, it's like, "Really? Aw, how cute am I going to send that?"

01:02:53   Or it's going to look like a form letter, or it's going to look like a canned reply, you know,

01:02:58   similar complaints about the inbox software that they have for auto-replying to emails and stuff

01:03:02   like that. That's the obvious objection to any sort of, "Hey, we'll figure out what you're going

01:03:06   to reply and give it to you in a box," right? And they addressed it immediately with a single

01:03:09   sentence, which may or may not actually address the issue in the product, but they're saying,

01:03:13   "Oh, and don't worry, we will learn from what you actually reply and suggest things that are essentially things that you have said before or that are otherwise in your voice."

01:03:21   Now, will they be successful at that? Will they just literally have replies I've had before and it'll sound dumb and none of the boxes will look right?

01:03:27   But at least they understand that they can't suggest replies without taking how you actually communicate as an input.

01:03:35   And so they said, "No, that's exactly what we're doing."

01:03:37   Now, again, they can do a bad job of it and it'll still be a stupid feature that people will want to turn off,

01:03:41   but if they do a good job of it, it's like, "Yes, that's exactly what I want.

01:03:44   I want you to watch how I type to people, and preferably how I communicate with different people.

01:03:49   How do I reply to my mom versus my wife versus my friend?" You know what I mean?

01:03:52   That's machine learning, and again, they keep leaning on that.

01:03:56   Hopefully that's what they're aiming for.

01:03:57   They're not saying, "Oh, we're just going to give you a bunch of canned replies like the Apple Watch shows

01:04:03   or probabilistically try to make a reply that would make some sense," you know?

01:04:08   Apple Watch is not learning how I reply. It has no way to learn from how I reply.

01:04:11   Apple doesn't have that information or whatever, but Google at least says, "This is what we're

01:04:16   trying to do."

01:04:17   And I really hope they do because that's the dream, right?

01:04:19   Yeah, that's fair.

01:04:22   They also have plenty of third-party integrations in LO, and I initially was really keen on

01:04:30   this.

01:04:31   So I forget the exact example they used, but say you're talking with your wife about where

01:04:34   you want to go to dinner, and so I'm talking to Erin about it, and she says, "Oh, would

01:04:37   you like pizza?"

01:04:38   And I say, "Yes, I'd love some CCs."

01:04:40   And it will try to figure out some sort of contextually relevant information, like if

01:04:46   it's something nicer than CC's, "Oh, can I get you a reservation?" or perhaps, "Here's

01:04:50   directions there," etc.

01:04:53   And at first I thought, "Oh, wow, that's super convenient, really useful."

01:04:56   But then I thought about it and I thought, "I don't know if I'd really want that in an

01:05:00   iMessage conversation," which obviously is approximately the equivalent, "with Aaron."

01:05:05   That just seems weird to me.

01:05:08   And I can't, I haven't put my finger on which one I think more.

01:05:11   Well, it's great that they kept leaning on this, and I think it is a great feature,

01:05:14   is that you don't have to leave the app.

01:05:16   This frustrates me when I'm on my phone communicating with somebody that I have to like, "Wait,

01:05:19   let me go Google that," or, you know, "Wait, let me go to this other app to do the thing."

01:05:23   Having everything in line, one aspect of it is like, do I want the other person to see

01:05:27   sort of the research I'm doing here?

01:05:29   Do I want to share these things?

01:05:31   That's questionable, debatable, although if you're trying to decide in restaurants, maybe

01:05:34   they do want to see them so you don't have to describe the restaurants or type them all

01:05:37   So I can see the deciding what to share versus what's your own, but the big advantage that

01:05:42   Google was selling and that I think is a real advantage is you don't have to switch to 15

01:05:45   different apps.

01:05:46   You can get it all done here, mostly because the power of Google is showing through inside

01:05:51   this chat application.

01:05:53   We will do all the searching and the pulling of the search results in smart ways and even

01:05:57   the ordering and the making the reservation.

01:05:59   And again, it falls back on like that's a silly scenario unless everything in your life

01:06:03   is integrated with Google and all the things you care about have third-party integration.

01:06:06   But hey, at least they have third party integrations.

01:06:08   And if third parties really care about this,

01:06:10   they can integrate it.

01:06:11   And if a restaurant really cares and wants to sell

01:06:13   to tech nerds in the San Francisco area,

01:06:15   they can integrate with it.

01:06:16   It doesn't help most of the rest of the country

01:06:18   whose favorite restaurants are not integrated in this way,

01:06:20   but at least maybe you can get directions to them

01:06:22   assuming that Google has accurate information

01:06:25   for where they are.

01:06:26   So this one is a little bit fantasy,

01:06:29   but Google is doing all the right things.

01:06:32   And I think making people not leave the app

01:06:35   is actually the right thing.

01:06:36   just difficult to draw that line of how much crap do I need the other person to see versus

01:06:40   how much stuff do I need to see to make a decision.

01:06:44   Yep, they also allow you to play what they call the emoji game in line.

01:06:49   So if you get bored talking to your spouse, you can play the Guess the Movie Title by

01:06:53   Emojis game, which strikes me as the, what do they call it, digital touch features on

01:07:01   the Apple Watch.

01:07:03   But it's not though, because this was just their sort of like "Hello World" example program.

01:07:08   The idea is that, again, third parties can make actual fun games with this API.

01:07:13   This is more like, see, it's like making an Echo server for demonstrating your simple

01:07:20   server-side concurrency framework, right?

01:07:21   It just echoes back whatever you put in.

01:07:23   It's not a real thing, it's an API, and they fully expect people to make real fun games

01:07:29   out of this.

01:07:31   with third-party products, games that lead you to third-party products.

01:07:35   There are many opportunities for both advertising and commerce integrated into this.

01:07:40   And so I think showing that, and by the way, on stage, essentially saying, "Shall we play

01:07:45   a game to your computer?" is not a good move.

01:07:47   It's not a good look, so maybe stay away from that.

01:07:50   But the big thing I think they weren't selling is that we didn't just make a chat app that

01:07:55   has some features in this version of Android.

01:07:58   This is now another platform for you to target all your stuff at.

01:08:02   You people out there who have products and applications and services can integrate with

01:08:06   this and be integrated with it in a way that people can buy your stuff and find their way

01:08:11   to your thing and play your game that advertises your thing or whatever.

01:08:15   That's the feature.

01:08:17   The actual emoji game, I don't think that was...

01:08:20   I mean, it was probably clear to developers, but if a regular person saw that, they'd be

01:08:23   like, "Oh, I'm not interested in emoji games."

01:08:25   Ignore the emoji game.

01:08:26   just a proof-of-concept, hello world type thing. Other integrations is what you're

01:08:30   looking for here. Well, fair enough. Then they made mention during the Ello

01:08:36   conversation and a couple other times about security, which I really, really

01:08:41   liked that Google and Apple seemed to be quietly colluding in the good way to try

01:08:47   to make their their platforms more secure. But that being said, I wasn't

01:08:51   entirely clear on what the security situation is in LO, it seems that there's kind of two

01:08:57   modes.

01:08:58   There's the general normal use mode, which is encrypted-ish, I guess.

01:09:06   It's encrypted enough to prevent snooping by anyone but Google, because they need to

01:09:12   have their equivalent of Slack -- or, yes, the Slackbot in there in order to offer all

01:09:20   these suggestions. And then there's an incognito mode, which is just like Chrome, in your messaging

01:09:26   app, which is apparently encrypted E2E or end-to-end. That just, I like it overall,

01:09:34   but that just seems weird to me to have to think about whether or not you want this to be end-to-end

01:09:39   encrypted. Yeah, and I wasn't clear too. I watched most of this presentation, but I didn't get a

01:09:45   chance to rewatch this segment. In the incognito mode, it doesn't hide you as the sender, right?

01:09:49   Like it's the other person still knows it's you talking, right?

01:09:52   You just initiated a new end-to-end encrypted session.

01:09:54   Is that—?

01:09:55   I think that's correct.

01:09:56   Yeah.

01:09:57   All right.

01:09:58   So anyway, yeah, it's—I think you got exactly right.

01:10:03   It's a difference between could Google encrypt this if they wanted to in theory or can nobody

01:10:09   do it?

01:10:10   And as has been discussed at length with iMessage, even though iMessage is end-to-end encrypted

01:10:14   as well, because Apple has control over the key servers, still technically if Apple wanted

01:10:19   to be nefarious, they could, you know, decrypt your conversations as they happen.

01:10:25   They don't do that, and they don't plan to do that, and they say they'll fight the government's

01:10:28   attempt to make them do that, but there are so many other ways to get to your information,

01:10:32   like the unencrypted iCloud backups of your conversations, and yeah, it's much more complicated

01:10:37   than simply putting E2E up on a screen and making people feel safe, especially with the

01:10:41   government climate the way it is.

01:10:43   But yeah, the reason Google, I think, didn't have iMessage-style encryption from the beginning

01:10:47   is because Google wants to be able to see everything that you type and it wants to have

01:10:53   a participant in the conversation who's in on it. Even if you did an end-to-end encryption,

01:10:57   one of your authenticated end-to-end recipients would be something on a Google server somewhere.

01:11:03   I don't think they're anonymizing when they go back, like Apple is when they go back to

01:11:06   the Siri stuff, because they kind of have to know things about you and know who you

01:11:09   are to do smart server-side things, because I don't think all this is happening on the

01:11:14   phone, right? There is a server-side component. So it's the tension we've always talked about.

01:11:17   like that Apple is trying to be good about privacy,

01:11:19   and I don't even want to know your stuff,

01:11:21   but if you want personalized service

01:11:23   from an intelligent agent with machine learning,

01:11:28   the intelligent agent has to know who you are.

01:11:30   I mean, I want it to know who I am,

01:11:31   but now all of a sudden this intelligent agent

01:11:33   is privy to my conversations, and it's not a person,

01:11:37   and it's essentially owned and controlled by Google.

01:11:39   So end-to-end encrypting it with the agent

01:11:42   is not really helping things

01:11:44   be worried about the government forcing Google to give it records.

01:11:49   Anyway, I think in this day and age, if there's something you don't want the government to

01:11:53   see, don't send it through Google or any other service, probably even Apple.

01:11:59   I mean, Apple's probably your best bet, but in general, you'd have to take encryption

01:12:02   into your own hands if you actually want it, which is fairly easy to do.

01:12:06   The math exists, it's out there.

01:12:08   You can get encryption software and use it yourself and communicate with somebody in

01:12:11   in a way that the government can't subpoena anybody to get, but again, you'll probably

01:12:14   just screw it up and they'll be able to social engineer something out of it anyway.

01:12:18   [laughter]

01:12:19   Oh, goodness. All right, let's talk about Duo and Knock Knock. So Duo is FaceTime, but

01:12:27   not. There's more to it than that. And then what is Knock Knock? That's the intro mode

01:12:32   in Duo, is that right?

01:12:34   Yeah, this is another one of the demos where they didn't say the obvious thing. It's like,

01:12:38   And when someone's calling you look I can see my daughter and I you know

01:12:41   It's like your your phone is ringing a center

01:12:43   It's like the FaceTime ring, right?

01:12:44   And rather than just seeing the name of the caller and maybe like a little picture of them from your contacts or whatever like, you know

01:12:51   Your your daughter is calling and you see that your daughter's name and a still picture of your daughter. What you'd instead see is

01:12:56   video

01:12:58   From the the caller side before you've decided whether you want to pick up or not, right?

01:13:05   How can you do that demo on stage and not immediately say but don't worry

01:13:10   This only happens for known contacts that are in your address book that you communicate with surely that is the case

01:13:16   To say that you have to actually say the words it takes two seconds to say the words

01:13:22   Otherwise the entire audience is going oh my god naked people are gonna be calling me constantly and I'm gonna instantly see their video

01:13:27   Because I can't stop them. It's like like do they not have women at the company like

01:13:32   The women would say I do not want to see a million guys junk as they randomly dial me and I have no choice

01:13:38   And I see video of it immediately like and again

01:13:41   I don't think it's gonna happen because surely this only works for like your favorites or your known contacts

01:13:45   Like that's got to be a feature, but you have to say it on stage because otherwise I think Google seriously

01:13:50   Do you know I don't know that really bothered me that that went unmentioned

01:13:55   I feel like even Apple would say and don't worry. This only works for your contacts. See how easy that is Google

01:14:01   I mean, and again, surely that's the case.

01:14:03   I think there have to be engineers inside Google, you know, who understand the privacy

01:14:08   implications of unsolicited video from strangers appearing on your phone screen.

01:14:14   But they didn't say it in the presentation.

01:14:15   Or if they did and I missed it and I'm sorry.

01:14:16   No, I don't think they did.

01:14:17   Boy, boy, yeah.

01:14:18   As someone put in the chat room, junk time instead of face time.

01:14:23   So the thing that bothered me about the Knock Knock, to be honest, I didn't think about

01:14:27   that, but you're absolutely right.

01:14:28   But the thing that bothered me about Knock Knock was they started from a good place,

01:14:32   which was video call--well, phone calls in general, but particularly video calls, are

01:14:36   extreme--are an extreme interruption, and you have to really dedicate your entire attention

01:14:41   to a video call.

01:14:43   And sometimes it is not a convenient time for that, but then again, on the other side

01:14:49   of the coin, what if--what if Aaron is calling me to say, "Oh my God, Declan just took his

01:14:54   first steps," or something like that?

01:14:56   So what Knock Knock does is it lets you see what this person is calling about.

01:15:00   So I can see Erin in Declan and I can see her freaking out.

01:15:03   And in their little demo reel video, they showed a guy who was holding up two ticket

01:15:07   stubs like he had gotten tickets to a concert or something like that.

01:15:10   And then they showed another person who holded up their left hand to show, I think it was

01:15:14   their engagement ring.

01:15:16   And so the problem I had with this, and maybe I'm just being a jerk about it, but it was

01:15:22   like you have to earn your right for them to pick up your phone call.

01:15:26   Like you have to earn it man. You can't just call and be smiling. You got to perform

01:15:30   You got to put on a show or I'm not picking up the damn phone

01:15:33   Or they're not gonna pick up. I don't think it's a real problem because again

01:15:36   It's only gonna be your actual people you know doing this right who's video calling you anyway

01:15:41   No one is cold video calling you or if any case you wouldn't pick up

01:15:44   It's gonna be people you know and the only time they would video call you instead of regular call you is if they have something

01:15:49   To show off or they're an adorable right right?

01:15:52   Why else are they gonna no one's gonna video call you just to essentially have a phone conversation

01:15:56   Because who wants people looking at them unless they have something to show are they in an impressive place?

01:16:01   So, I don't know

01:16:02   I don't I guess it depends on who you talk to but I don't feel like anybody who I would ever video call

01:16:08   Would deny my call because I was not sufficiently

01:16:10   Well, the other thing is any face time conversation I've ever had has almost always been

01:16:18   preceded by an iMessage conversation saying, "Hey, I'd like to FaceTime you. Are you cool?"

01:16:22   You know, it's almost to your point, John,

01:16:25   it's almost never me cold calling or somebody cold calling me saying, you know, with no previous warning.

01:16:32   But is that perhaps because there is no knock-knock-like feature for me to kind of screen that call?

01:16:38   Maybe? I do get those, by the way. I do get a surprising number of not phone calls, but FaceTime calls

01:16:45   randomly on my phone and of course on my Mac is integrated with my phone. I don't answer

01:16:50   any of them.

01:16:51   - From strangers?

01:16:52   - It's like unknown, yes.

01:16:53   - Oh, that's weird.

01:16:54   - Just unknown caller number.

01:16:55   - Really?

01:16:56   - And not a phone call and not a FaceTime audio call, an actual FaceTime video call.

01:16:58   - That's super weird.

01:16:59   - I don't know if there's like war dialers out there just trying every single number.

01:17:02   I don't know, I've never answered one of them so I have no idea what's going on but it's

01:17:04   annoying that it makes the FaceTime like bloopity-bloop sound.

01:17:08   - I mean, like I've had, like I have a few relatives who basically treat FaceTime like

01:17:15   phone calls and so they will just call me on FaceTime like, you know, unannounced, just

01:17:18   as that's their call.

01:17:19   Yeah, my parents will do that. I think that's fine.

01:17:21   But I, you know, I've never wished for this feature. I mean, I'm sure it's fine. It's

01:17:26   certainly going to be odd at first before anybody's used to it because that's not how

01:17:30   things have gone so far with these kind of video calling things. Like, so there's kind

01:17:36   of an expectation that you are not being broadcast before the person picks up. So I think you're

01:17:41   going to see a lot of people picking their nose and adjusting their hair and stuff like

01:17:44   that, like just like, you know, embarrassing prep before they think they're being watched.

01:17:48   >> Yeah, I mean, yeah, this is something that people will have to get used to, but I think

01:17:53   they will get used to it.

01:17:54   Hopefully it's an optional feature, again, something they didn't mention.

01:17:56   Like, is this optional?

01:17:57   You know, it should be for only known contacts, and for known contacts it should be an option,

01:18:01   because maybe you don't want that to be the case, exactly the way you said it, you're

01:18:04   just more used to, like, until they pick up I'm not on camera.

01:18:06   >> Yeah, a couple other quick notes about this.

01:18:09   It apparently does really, really well with graceful degradation as the connection gets

01:18:16   crappier and then whatever the opposite of that is as things get better.

01:18:22   And the thing that I'm most excited about about this is that it's multiplatform.

01:18:27   And I believe at one point or another, they had said that Elo and presumably Duo operates

01:18:34   where your phone number is kind of your user ID, if you will.

01:18:38   something that I've struggled with is Erin's side of the family with the exception of Erin is exclusively Android and

01:18:44   There have been several times where we'd like to have a video chat with her mom which her mom only lives about 20 minutes away

01:18:51   But maybe Declan is doing something funny or maybe I don't know we want to show her something in the house

01:18:57   we'd like to do a video chat and

01:18:59   We don't feel like we have a mechanism to do that now. Yes. I know that Skype exists. Yes. I know that hangouts exist

01:19:06   yes, I know that there are many other video chat apps that exist but

01:19:10   There's nothing that I personally have used that is as easy as FaceTime

01:19:16   And I hope that this duo thing will be as easy as FaceTime

01:19:20   and so we'll be able to install this on our iPhones and

01:19:22   My in-laws will be able to install it on their Android phones and we'll be able to do these very quick very simple

01:19:29   Video chats on a whim which would be really really awesome

01:19:33   And I'm really looking forward to that hopefully working.

01:19:36   Or you can just wait for FaceTime to become an open standard because they're going to

01:19:41   the standards bodies today with that.

01:19:43   It's just a matter of time now.

01:19:45   Yeah, this is another thing that's Google's strength.

01:19:48   My experience with Google Hangouts, for example, is it is the first and best sort of multiple

01:19:54   people on a video stream thing that I've ever used.

01:19:56   And is the quality great?

01:19:57   No, but they like really good video games and good iOS applications and so many good

01:20:02   They understand that the most important thing, like they understand the hierarchy of needs in a video call.

01:20:10   Number one, audio. If you have to drop the video to maintain the audio, do it, because nothing is more annoying than not being able to hear people.

01:20:17   And then number two, responsiveness of video rather than quality.

01:20:21   If you have to drop the quality to an obscene level just to keep track of someone waving their hand, then do it.

01:20:26   Like, responsiveness and understanding the audio is the most important thing.

01:20:29   And my frustration with FaceTime is often that the audio will start cutting out and makes it impossible to even communicate about the bad video

01:20:36   And the video will have higher overall quality when it's working, but when my parents' terrible internet connection starts dropping things out and it becomes like a slideshow, I wish it would just degrade to a much uglier algorithm

01:20:48   And FaceTime will do this. FaceTime will turn off the video entirely to go to the audio, but in practice it does not do that soon enough

01:20:55   It very often struggles with stuttering audio and I wish I could just like just you know drop this entirely very often

01:21:01   FaceTime calls have been derailed by saying just call on the phone because we know the phone will work and I'll be able to hear

01:21:06   your words, so

01:21:08   I gotta haven't you know, the only thing I've used video from Google is you know YouTube which of course is only Google back position

01:21:14   and Google Hangouts and I've always been impressed by

01:21:17   The performance under pretty you know dire conditions of so many people are across the country doing one big giant multi video

01:21:24   conference call thing. So if this duo thing works and because it's cross-platform, it

01:21:31   may end up being an easier, like, we'll see if it ends up being like my go-to. Like, I'll

01:21:36   probably still do FaceTime with my parents because that's just what they're used to,

01:21:39   but if FaceTime is messing up, I will have this app installed and I'll make sure they

01:21:42   have it installed too and I'll say, "Switch to this app and we'll just do A/B testing."

01:21:46   Oh, FaceTime is dying because of you read Internet weather over there? Try this other

01:21:51   app that I showed you and hopefully it will be simple enough that they can figure it out

01:21:54   and it will be connected up in a way.

01:21:56   But I'm optimistic about this as being a decent product.

01:22:00   And like Marco was ridiculing Apple for the open standard,

01:22:02   like this, Apple could have done this, but hasn't.

01:22:06   How has FaceTime gotten appreciably better

01:22:08   since it was rolled out?

01:22:09   I'm sure it has.

01:22:10   I'm sure it's gotten more reliable

01:22:11   and the algorithms have gotten better

01:22:12   and the quality is probably better at the top end

01:22:15   than Google's thing is gonna be.

01:22:17   But that doesn't matter when I can only hear

01:22:18   every fifth word that my parents are saying.

01:22:20   - Yeah, and also FaceTime notably does not support

01:22:23   more than two people on a call yet.

01:22:25   And I hope we get that at some point,

01:22:27   'cause like, you know, iChat had that before FaceTime

01:22:31   was a thing, like we had that 10 years ago.

01:22:34   So I know it's more challenging on mobile,

01:22:36   on cell connections and everything.

01:22:37   It's not an easy problem by any means,

01:22:39   but I do hope we get there.

01:22:41   - Yeah, but is Duo, I mean, Duo itself implies two.

01:22:45   I don't think that Duo is going to be more than one person,

01:22:49   is it, or more than two people, I should say.

01:22:51   I mean if nobody actually has like more than one of the person they know with an Android phone, right?

01:22:55   The other thing is supposedly e2e encrypted always enterprise to enterprise. Yeah

01:23:05   That one is either we'd encrypted always

01:23:08   Well, they didn't say I was they just said end-to-end encryption. They didn't give a qualification

01:23:11   So I'm sure this means always mostly because at this point I wouldn't say this is gonna be true forever

01:23:16   But at this point there is no Google bot equivalent that needs to see every frame of your video do something intelligent

01:23:21   But if there was sufficient bandwidth both in terms of CPU and you know and and data throughput

01:23:27   I'm sure like under unlimited conditions in a local environment

01:23:31   Google would love to have an intelligent agent watching every frame of your video and doing intelligent things based on it because you know

01:23:37   You know, it's still image recognition

01:23:39   It's you can do that on video and then video recognition where you're actually not just looking at individual frames with the actual video and

01:23:46   having something realize

01:23:48   Where you are what you're talking about being able to do hand gestures seeing your facial expressions

01:23:54   Then Google, you know would need to see your video and on the local device maybe maybe you could still be end-to-end encryption

01:24:02   But again, there's probably there's almost certainly a server component to this

01:24:05   And so I don't know how long end-to-end encryption end-to-end encryption on do only last until Google realizes

01:24:12   They can do intelligent helpful things by

01:24:15   Looking at your video, so that's probably still a ways off because of just bandwidth concerns and everything else

01:24:20   But I wouldn't expect it to be always fair enough all right so updates to Android yeah

01:24:26   I was getting towards the end of it. I was getting towards the end of this section

01:24:29   I had to skim some of this, but I tried to put some highlights in here. I'm sure there's things that I'm missing

01:24:33   Vulcan Vulcan was kind of depressing because I

01:24:36   mean Apple Apple's been so far ahead with the

01:24:41   graphics performance on their devices for so long, especially compared to like the average

01:24:45   of Android versus just like the high-end Android.

01:24:48   And then Apple, you know, did that OpenGL ES, and then they did the Metal thing, which

01:24:51   is a lower-level thing, and that was great, and I'm sure it's good for iOS game developers

01:24:56   and everything, but on the other side of the coin is the sort of open standard OpenGL Chronos

01:25:01   group thing, which is Vulkan, which is based on Mantle from AMD, and a bunch of other stuff,

01:25:06   and it's very much like Metal, and you know, all these ideas have been floating around

01:25:09   the graphics community for a while.

01:25:11   I don't like to see Apple as the,

01:25:13   it's kind of like the FaceTime situation.

01:25:15   Oh, we've got our own thing.

01:25:17   The only good thing about it is that

01:25:19   Apple's own thing is actually pretty influential

01:25:21   because Apple, a lot of games ship on Apple's platforms

01:25:25   and they make a lot of money.

01:25:26   So maybe they could quote unquote win

01:25:28   just because most of the money in game development

01:25:31   is happening on iOS.

01:25:33   But I would really rather see Apple supporting

01:25:36   the industry standards.

01:25:37   Like let's get everybody working together

01:25:40   to make the low-level graphics API

01:25:42   that everybody's going to use,

01:25:43   as opposed to Apple having its own.

01:25:45   But Apple having its own

01:25:46   is probably kind of a competitive advantage,

01:25:48   and they are so big they can probably get away with it,

01:25:49   but it just depresses me a little bit.

01:25:52   - Fair enough.

01:25:53   Let's see what else.

01:25:55   They did a lot with just-in-time compilation,

01:26:00   which is kind of exciting,

01:26:02   and even just-in-time installation, which is weird.

01:26:08   Well, yeah, the Instant Apps thing.

01:26:10   We'll get to that in a second.

01:26:10   But the just-in-time compilation is the flip side,

01:26:12   where I start to feel bad for Google,

01:26:14   that they're still dealing with the JIT.

01:26:16   And they're like, oh, we'll profile the code

01:26:19   and write the optimized version of it

01:26:21   to Flash to make a faster launch each time.

01:26:23   And I know they have precompiled apps, too.

01:26:25   But just that technical decision to go with, initially,

01:26:29   a weird Java-ish language and just-in-time compilation

01:26:32   and everything, they're still kind

01:26:34   of paying the price of like having to try to match Apple, which is shipping precompiled

01:26:41   binaries, right?

01:26:42   Precompiled, pre-optimized binaries that are smaller, that they don't have to have the

01:26:46   -- they're doing plain catch-up here.

01:26:51   And maybe long-term they're still doing the right thing, and maybe it's still an advantage,

01:26:54   but in the current stage there's lots of weird compromises that they're digging out from

01:26:58   underneath.

01:26:59   And so there's a whole section of the slides that are just things that Apple is just not

01:27:01   concerned about because they don't have these problems because they don't have a JIT.

01:27:05   >> JIM: Indeed. They have multitasking.

01:27:09   >> BRIAN KARDELL Yeah. That was the part where it's like,

01:27:12   "All right, you did pretty much what Apple did, which is not that imaginative when Apple

01:27:15   did it, and it's not that imaginative when you did it." Which part did you think was

01:27:18   novel? I mean, split screen and picture in picture is like literally like word for word

01:27:22   of the iOS, iPad multitasking. They did it on a phone, which I think is something that

01:27:27   that Apple didn't do.

01:27:28   Does Apple even let you do split screen on the mini,

01:27:31   does they? - Yep, yep.

01:27:32   - Do they?

01:27:33   I don't know what the limits of that feature is.

01:27:34   All right, anyway, on the iPhone, no,

01:27:36   but they showed it on the phone,

01:27:37   basically a top bottom split instead of from the side.

01:27:40   - Yeah, well, 'cause you wouldn't want

01:27:41   like two inch wide apps side by side,

01:27:45   that would be awkward.

01:27:46   - Yeah, well, I don't know, it depends on the app,

01:27:48   but yeah, they can scroll the text sideways.

01:27:51   But what was, and they had like the double tap

01:27:54   to switch back really fast.

01:27:55   Oh, here's the thing that killed me in this presentation.

01:27:59   I kind of understand why they did it.

01:28:00   So they showed their app switcher,

01:28:01   which looks like all apps switches look like these days,

01:28:03   a bunch of little cards that look like a web OS

01:28:05   from many years ago, showing all your applications.

01:28:08   I said by popular demand, they added a clear all button.

01:28:12   And that is essentially recognizing the fact

01:28:15   that lots of people either based on superstition

01:28:19   or bad reasoning related to the iOS multitasker,

01:28:22   or on the Android side, probably both of those as well,

01:28:26   but also on both sides, based on just the idea

01:28:30   that some people want to clean up messes

01:28:32   and that having a bunch of quote unquote,

01:28:34   open applications is visual clutter, right?

01:28:38   And they just want to clear out.

01:28:39   So the newer version of Android does two things.

01:28:40   One, they limit it to only seven.

01:28:42   So it just, they go off the end, you know,

01:28:43   they don't, they just don't show them,

01:28:45   which is kind of nice for reducing clutter,

01:28:47   but I hope that's adjustable somewhere, probably is,

01:28:50   'cause what if you want to see more than seven?

01:28:52   And the second thing is the clear all button.

01:28:53   It's like sometimes I just want to clean everything up.

01:28:55   I mean, I don't think I've mentioned this before.

01:28:57   I've seen my son do it.

01:28:58   I saw him using his iPhone and he goes into the app switcher

01:29:01   and flicks all the applications up.

01:29:03   I'm like, whoa, whoa, whoa.

01:29:03   It's like, where did you learn this?

01:29:05   Who's teaching?

01:29:06   Like, it's just, they pick it up on the streets.

01:29:09   They're kind of like, force quitting application.

01:29:11   So I had to explain to him, you know, the whole,

01:29:14   we don't talk about that much on the show,

01:29:15   but the angle that I gave to him,

01:29:17   I didn't want to explain to him about multitasking,

01:29:20   suspended processes and everything.

01:29:21   I said, look, when you do that,

01:29:24   it's gonna actually exit the program.

01:29:26   It's going to, I don't know if I said Forest Court

01:29:28   or whatever, which means the next time you launch it,

01:29:30   which will probably be two and a half minutes from now,

01:29:32   it has to launch from a completely fresh state.

01:29:35   Whereas before it was just suspended in memory,

01:29:37   ready to go the next time you used it.

01:29:39   So you think you're quote unquote saving battery,

01:29:41   which is the excuse he gave me, but you're probably not

01:29:43   because launching from a fresh state takes more energy

01:29:46   than unsuspending the application.

01:29:47   So just don't worry about and don't leave them.

01:29:49   But the other angle isn't it,

01:29:50   you have an argument about it, it's like, I just don't like seeing them there.

01:29:53   Well, so Android users apparently have the exact same problems, so they gave them a big

01:29:56   button that says "Clear All."

01:29:58   What I hope the Clear All button does is remove the graphics but do nothing else.

01:30:01   Leave them all suspended.

01:30:04   Because that's what people want.

01:30:05   They just don't want to see the squares there.

01:30:06   That's what Apple should do.

01:30:07   I mean, they'd have to add a new feature for actual force quit, but I wish flicking up

01:30:11   the squares didn't do anything except for remove them from the multitasking switcher,

01:30:15   but everything else stayed the same.

01:30:16   They just stayed suspended and were managed in a normal way.

01:30:18   And boy, this ailment, this is like the zapping the PRAM of the mobile age or rebuilding the

01:30:26   desktop.

01:30:27   I don't know how far I can go into repairing permissions for building desktops older.

01:30:31   These voodoo solutions, but this one I think there's a foundation that people just don't

01:30:34   like to see the mess.

01:30:35   So Android is giving them a clear all button.

01:30:37   We need to have a summit, like a G8 summit for people who make computer devices to talk

01:30:43   about pathologies related to the multitasking switcher.

01:30:48   - There are a lot of them.

01:30:49   - Yeah, I mean the sad part with the Apple one is like,

01:30:53   it would be nice if we didn't have to do that

01:30:55   and if they removed the ability to do that

01:30:57   so that people could stop doing the stupid thing.

01:30:59   But unfortunately, it does occasionally solve problems.

01:31:02   - Yeah, you do need a way to force quit things.

01:31:04   Like you do badly behave, you do need a way to do that.

01:31:07   So you have to, you can't take that away and just say,

01:31:09   "Oh, don't worry, we'll manage it all for you."

01:31:10   But the obsessive need to do it every time,

01:31:13   like I've seen my son do it, I've watched him,

01:31:15   it just, he just like, it's just part of his routine.

01:31:17   It's like, there's no point in that, you know.

01:31:20   And especially because you just convinced yourself

01:31:22   that if I don't do this, something bad will happen.

01:31:23   Like, I just wish that it just removed the graphics.

01:31:27   Because I think that's like 90% of it.

01:31:29   People just don't wanna see the squares.

01:31:30   They don't wanna see the little rectangles.

01:31:31   They want to feel like everything is a clean slate.

01:31:34   So fine, get rid of the rectangles, right?

01:31:36   But then, you know, when the Facebook app starts going crazy

01:31:39   you do need some way to actually kill it.

01:31:41   Or sometimes when applications get all screwed up,

01:31:43   which happens as well.

01:31:44   Like perfectly well behaved applications

01:31:45   where there's a bug in them all of a sudden

01:31:47   and every time you bring them to the front,

01:31:49   they're just not working, that happens,

01:31:51   and you need a way to get rid of them, get rid of them.

01:31:54   I don't know, it's a difficult situation.

01:31:56   - All right, I got it.

01:31:58   This is how Apple's gonna solve all their problems.

01:32:00   You get one free clear all a day.

01:32:04   And then if you want more--

01:32:06   - You're gonna want more with an in-app purchase?

01:32:08   - Yeah, you can buy one for a dollar each

01:32:11   if you want more after that.

01:32:13   Their growth issue with iPhone revenue is solved.

01:32:16   Yeah, it's an energy-based mechanic where you exhaust your energy, you have to do an

01:32:22   app purchase to get more.

01:32:26   Clear all.

01:32:27   I don't know, I wasn't there in the crowd, but I wonder if that was like a big applause

01:32:31   feature because they did say "by popular demand."

01:32:33   Someone who works at Google, please tell me that doesn't actually do anything except remove

01:32:38   the graphics, it'll make me so much happier.

01:32:41   All right, we should quickly talk about Instant Apps.

01:32:45   This is a really big, hairy, technical mess

01:32:50   that might end up being really good,

01:32:52   has a lot of potential risks,

01:32:54   has a lot of potential ramifications, good and bad.

01:32:57   We don't really know enough about it yet,

01:33:01   I don't think, to really say whether this is going to work,

01:33:06   what the risks totally are.

01:33:08   - You're worried about security risks?

01:33:09   Is that what you're thinking of?

01:33:10   - Hold on, hold on, hold on.

01:33:11   Let me kind of explain what's going on here.

01:33:12   - Yeah, yeah, explain.

01:33:14   - Chief Summarizer. - Chief Summarizer,

01:33:15   Summarizer and Chief, so.

01:33:16   - Makes it longer, why'd you?

01:33:18   You're doing it.

01:33:19   - I'm doing a terrible job, I'm fired.

01:33:21   So, the idea with Instant Apps is,

01:33:24   you're in a situation where you really want

01:33:27   to consume some content that is associated with an app.

01:33:30   So they gave an example of wanting to watch a Buzzfeed video,

01:33:33   they gave an example of walking up

01:33:35   to an NFC-enabled parking meter.

01:33:37   - Right, 'cause critically, you wanna do something

01:33:39   with an app that you don't have installed.

01:33:41   - Sorry, yes, yes.

01:33:42   So you don't have this installed, and you walk up to this parking meter, and the parking

01:33:46   meter has some NFC ID that some way, somehow, Google and Android know are associated with

01:33:53   such and such app in the Play Store.

01:33:56   So what Instant Apps will do is it will, behind the scenes, instantly download the subset

01:34:05   of the app that you need to perform the particular function you're trying to do, be that watch

01:34:09   the BuzzFeed video or pay for parking or what have you.

01:34:13   And it will do that and load it instantly or thereabouts so it's available to you.

01:34:17   So you walk up to the parking meter, you swipe your phone near it, the NFC thing kicks in.

01:34:22   It will download the parking meter app.

01:34:26   You can put money into it, you can use your Android Pay and walk away and everything's

01:34:31   good.

01:34:32   And you haven't downloaded the entire app, you've just downloaded the pieces, I'm assuming

01:34:36   the intents in the Android system that are required for the operations you're trying

01:34:41   to perform.

01:34:42   At first glance, it sounds kind of good, because I don't like downloading apps just for single-use

01:34:49   events like this.

01:34:50   And actually, I have in the past downloaded a parking meter app in the DC area specifically

01:34:55   to park in DC, and then I've never used it since.

01:34:59   And so, in theory, this sounds kind of good, but what happens then when I walk away from

01:35:07   the meter, I didn't pay close attention to what the brand of meter was or what app I

01:35:13   just got quietly installed for me, and I need to add time.

01:35:17   Like what do I do then?

01:35:19   Does it say, "Oh, you've just installed the Parkminder app"?

01:35:22   Does it allow you to download the Parkminder app from the Play Store because you've instant

01:35:27   downloaded it recently?

01:35:28   how does that all work? I'm actually less concerned about Android apps splitting themselves up into pieces

01:35:35   because they tend to be, from what I gather from our Android developers, a lot more modular, but

01:35:40   this whole, like, user interaction, I think it leaves a lot of questions in my mind.

01:35:45   Well, I mean, with the machine learning and the context thing, like, they said, "Oh, and you can add more time later."

01:35:51   I thought the same thing, like, how do I know how to add time? But,

01:35:55   Textually basic machine learning stuff that they've already got it knows what you're talking about because you just did a parking meter thing

01:36:02   And so if you just yell into your phone or type into your Google Assistant app add another minute to the meter

01:36:08   It should know what the heck you're talking about from context, and I think it can already do that

01:36:12   but I think the idea is that from this is kind of like getting more into the open doc model from a

01:36:19   phone user's perspective

01:36:21   For certain classes of things and perhaps a very large

01:36:25   Class of things they don't care so much about your app. Your app is important to you as the app developer

01:36:31   They just want to accomplish something in the world whether it's paying for a parking meter or buying a movie ticket or you know

01:36:37   making a reservation at a restaurant there in front of

01:36:39   And they just want to accomplish that task. They don't want to find the app that they need to download

01:36:45   I found myself in a position where I want to do something

01:36:48   And I've heard that you can do something with this store that I know I'm like, what is their app called?

01:36:52   Or what was the name of that app that I heard or like?

01:36:55   Just based on other contextual clues and having an open API to say from the users perspective

01:37:01   Their phone will just do this something their phone couldn't do before it something they can do and fairly quickly

01:37:06   Without them having to search the App Store for an app and download

01:37:09   I think that's a good user experience and something they could be shooting for many problems exist in it

01:37:13   Security is probably one of the big ones but like you said like is this just

01:37:18   Spamming my phone with apps that are installed as I walk past parking meters or even if I initiated

01:37:23   How do I keep track of all this crap that's getting installed? Like there's a potential downside

01:37:27   For this to be abused but on the other hand it is a very Apple like feature

01:37:31   I mean Apple is the company that dreamed up open doc of like I'm not so concerned with your application

01:37:37   I'm concerned with my task and it's like computers of the future

01:37:40   We're like, how does that thing all of a sudden know how to work with this thing?

01:37:43   It's like magic or you're like, oh, I've never used this thing before in my life

01:37:46   But it has Android integration. I don't have to do anything all of a sudden my phone can do this thing that it couldn't do before

01:37:51   And I think that's awesome

01:37:53   I'm not entirely sure Google is the company to pull that off based on

01:37:57   The amount of weird things that go on in the Android ecosystem

01:38:00   But I endorse the idea like like Chromebooks for example and like the whole

01:38:05   30-year computer in a lake it doesn't matter because everything is in the cloud. I like the idea. I like it as a goal

01:38:11   I don't know how close we're gonna get to with this particular implementation, but we'll see

01:38:15   Anything else on Android N?

01:38:18   They didn't actually announce a name,

01:38:20   they're soliciting names, they made the joke about

01:38:22   don't call it something McSomethingface,

01:38:24   but anything else on Android N?

01:38:27   - I mean, I think it's gonna be interesting

01:38:29   to see how much of this stuff plays out and develops.

01:38:33   None of us use Android really ever,

01:38:35   so I think it's funny that we just spent two hours

01:38:37   talking about this stuff, but hey, it's industry news.

01:38:40   I think it's also interesting to see,

01:38:43   a lot of the commentary so far has been

01:38:45   that this is a lot of underwhelming stuff.

01:38:48   A lot of it is kind of just matching Apple features

01:38:52   or giving their response to Apple features.

01:38:54   And that's just how the industry goes.

01:38:57   There are some years where Apple

01:38:58   borrows heavily from Google features,

01:39:00   and there are some years where the reverse happens.

01:39:01   And this is a very mature market

01:39:04   of these advanced smartphone OSs

01:39:07   where I don't think we can really expect changes

01:39:11   that are on a much bigger scale than this most years,

01:39:15   because it just doesn't, we're to the point now

01:39:18   where this stuff is mature and the low-hanging fruit

01:39:20   has all been picked and--

01:39:22   - Is that really what you think after seeing this impression?

01:39:24   You're like, oh, low-hanging fruit,

01:39:25   they're doing kind of similar things,

01:39:27   we shouldn't expect big changes?

01:39:28   That was not my impression I got of this at all.

01:39:30   - Well, I mean like at the OS level.

01:39:32   The stuff they're doing with data services

01:39:35   and machine learning, ding, ding, ding,

01:39:37   is like, that is where the advancement

01:39:39   is happening for Google now.

01:39:41   - But you don't think Instant Apps is,

01:39:42   I mean, implementation-wise, who knows?

01:39:44   And by the way, this implementation will work all the way back to Jelly Bean, which was

01:39:47   another kind of sad part of the presentation.

01:39:48   It was like, "Oh, look how good we are about backward compatibility."

01:39:50   No, it's because you can't move your install base to your most recent OS.

01:39:54   But anyway, I think that is an admirable and interesting goal that no one else is even

01:40:00   touching.

01:40:02   Where have you seen something like that before?

01:40:03   I think it's an awesome idea, and they are the first big player to say, "It's a thing,

01:40:09   and you can try to do it."

01:40:10   and lots of caveats about it, but that's,

01:40:13   OpenDoc didn't work and maybe this won't work either,

01:40:15   but from a user's perspective, I think it is significant.

01:40:18   - I guess.

01:40:19   - Like, and that's, I would call that an OS feature.

01:40:21   Like, that's, you know, what else is, like,

01:40:23   launching and, installing and launching and running apps.

01:40:25   That's significant.

01:40:27   - I don't know, it's potentially so problematic,

01:40:31   like, from just like a security and technical perspective.

01:40:33   Like, not to say it can't be done, but just the, like--

01:40:36   - Yeah, that'll get garbage dumped with, yeah.

01:40:38   Yeah, it's going to be hard to do it well and correctly and safely.

01:40:42   And it depends on support.

01:40:43   It's like Apple Pay would be crap if I couldn't use it anywhere.

01:40:46   But you know, it's like the technology and the idea behind Apple Pay could be great,

01:40:50   but if I couldn't use it in any of the stores, it would be a total failure.

01:40:53   So there is the infrastructure part of it, which is weird because you're like, "Oh, Android's

01:40:56   the majority.

01:40:57   They should be great on the infrastructure, but they're the majority, but they're not

01:41:01   always where the money is."

01:41:02   And that kind of tends to lead to what, you know.

01:41:04   So let's put it this way.

01:41:07   Well, this is an idea that I think Apple should share,

01:41:09   and I think my overwhelming impression

01:41:11   of this entire presentation is so,

01:41:14   how incredibly far behind Apple is in so many of these areas

01:41:17   and how I don't see any hope of them catching up.

01:41:19   Like all the things that Google emphasized,

01:41:21   all of their strengths, machine learning,

01:41:23   server-side stuff, all of that,

01:41:26   I just look at that and I feel like if it was at Apple,

01:41:28   like, we can't do that, we're terrible at this stuff.

01:41:30   We're so far behind them,

01:41:31   I can't even see them in the distance anymore.

01:41:33   They're this tiny speck, like they,

01:41:35   Apple is just trying to take its basic services

01:41:37   and make them reliable and have some sort of infrastructure

01:41:39   for doing things that Google was doing reliably

01:41:41   like five years ago.

01:41:43   This, all this stuff is not a glimmer

01:41:44   in anyone's eye at Apple.

01:41:45   They've got Siri and haven't been able to advance it.

01:41:47   This thing is dancing over Siri's grave.

01:41:50   Like, oh, I just, if you're an Apple fan,

01:41:53   like this makes me think is that the good old days

01:41:55   when Apple and Google were working together on the iPhone,

01:41:57   how awesome it would have been if they had like, you know,

01:42:00   divided the labor and say, Apple,

01:42:02   you make the hardware in the OS and we'll do the services

01:42:04   and together we'll make the awesome platform of the future.

01:42:06   That didn't happen, unfortunately.

01:42:09   But now we have the situation where Google is just

01:42:11   so much better at so many things than Apple

01:42:15   and that it doesn't seem like Apple's getting better,

01:42:16   not only not getting better fast enough,

01:42:18   but some of them not getting better at all.

01:42:19   And it's just, it depresses me.

01:42:21   I guess on the bright side,

01:42:22   as long as Apple continues to make good quality hardware

01:42:25   and a pretty good OS that sells a lot of copies,

01:42:28   Google will continue to be forced to make

01:42:29   like its dual application for iOS.

01:42:32   But the other features like Instant Apps,

01:42:34   we have to wait for Apple to copy,

01:42:35   and all the machine learning stuff,

01:42:37   I don't have any real hope of Apple ever copying

01:42:39   if their past history is any judged.

01:42:42   So, boy, I think this was,

01:42:45   it's not like a giant victory of Google, although Apple,

01:42:47   because again, there's things that Apple does better as well

01:42:49   and each one is obviously going to emphasize the area

01:42:51   where they're stronger,

01:42:53   but this, I feel like Google is accelerating away from Apple,

01:42:57   not just barely staying ahead in the areas

01:42:59   that it has always been ahead at.

01:43:00   - Yeah, I think that's really astute.

01:43:02   One other quick point I wanted to make, and then we should probably wrap, is they talked

01:43:07   about Android Wear 2.

01:43:09   A couple of quick thoughts about that.

01:43:11   Number one, I understand why the watch I carry on my wrist today and every day is a rounded

01:43:16   rect, but man, the circles look so much better.

01:43:21   And number two, they had a couple of different means of input, including like a swipe keyboard,

01:43:27   which strikes me as freaking terrible, but they also had handwriting where as you write

01:43:35   it scrolls to the left automatically, which I think I'd seen on like OneNote or something

01:43:40   like that in years past, but it seemed like an incredibly clever way to handle text input

01:43:45   on a watch.

01:43:47   Because you can put about a character on the screen at a time, and if it's scrolling automatically

01:43:52   you know, under your finger, it looked at a glance like it worked really, really well.

01:43:57   And I'm very curious to hear if that is implemented well. And if it is, then I want it on my watch.

01:44:04   Oh, I still believe that any time you're doing text input on a watch, you've lost.

01:44:08   I agree. But in a pinch, it would be nice to not have to use Siri.

01:44:12   You just use graffiti. Yeah, totally.

01:44:14   Just draw the same letter over the same spot. You don't need to scroll. Just keep drawing

01:44:18   the same letter on the watch face, yeah.

01:44:21   The Android Wear, what it has going for it is

01:44:23   they're trying everything.

01:44:24   - Yeah, that's true. - Right?

01:44:26   - Well, honestly, I mean, I'm not gonna get too far

01:44:28   into this now 'cause we're out of time,

01:44:29   but honestly, I think, you know, Casey, you mentioned,

01:44:32   like, man, round looks so good.

01:44:34   Android Wear can do different shapes than Apple can

01:44:37   because the design of the platform from the beginning

01:44:41   was, in a very Android way, to have a scalable system

01:44:45   that could apply to any large set of different device sizes

01:44:50   and shapes and characteristics.

01:44:52   That I think ultimately will prove to be the better way

01:44:56   for wearables to be designed,

01:44:58   for wearable platforms to be designed.

01:45:00   I don't think the Apple, the current Apple Watch model of,

01:45:04   we're just gonna make one type of watch in one shape

01:45:09   with one configuration basically.

01:45:12   Like, yeah, they allow all the different bands

01:45:14   and there's two different sizes of the same shape,

01:45:16   but it's basically one configuration of the watch.

01:45:20   And I think ultimately the Android version of,

01:45:25   you can have all sorts of different sizes and shapes

01:45:28   and different capabilities, you know,

01:45:31   I think that will ultimately prove correct for wearables.

01:45:34   Where it didn't really necessarily prove correct for phones,

01:45:36   but you know, 'cause most people just want a rectangle phone

01:45:39   with a decent sized screen that has a whole bunch of,

01:45:42   you know, hardware in it and stuff.

01:45:43   but I think watches and wearables,

01:45:45   it'll prove to go the opposite direction.

01:45:47   And I wonder, I'll be very curious to see if Apple

01:45:50   takes the watch in that kind of direction.

01:45:52   'Cause honestly, I don't think they are

01:45:55   headed that direction, and that worries me.

01:45:57   But we will see.

01:45:59   - Alrighty.

01:46:00   - Thanks a lot to our three sponsors this week,

01:46:01   Ring, Automatic, and Pingdom,

01:46:03   and we will see you next week.

01:46:05   (upbeat music)

01:46:07   ♪ Now the show is over ♪

01:46:09   ♪ They didn't even mean to begin ♪

01:46:12   'Cause it was accidental, oh it was accidental

01:46:17   John didn't do any research, Marco and Casey wouldn't let him

01:46:22   'Cause it was accidental, oh it was accidental

01:46:28   And you can find the show notes at ATP.fm

01:46:33   And if you're into Twitter, you can follow them

01:46:38   [MUSIC]

01:47:07   So somebody who shall remain nameless and who is too embarrassed to be named on this

01:47:11   particular podcast tells me, tells us, or tells me specifically, that I am wrong about

01:47:17   quitting at iOS.

01:47:19   He says it's his only salvation.

01:47:22   Salvation from what?

01:47:23   From I'm assuming his battery draining.

01:47:25   He says, he says, you want a video?

01:47:27   It's completely replicable.

01:47:28   And I thought, I told him in the chat here is that I thought I covered all bases.

01:47:34   Yes, sometimes it's a badly behaving app that you have to force quit.

01:47:38   Sometimes an application is in a weird state that the only way you can get it to work again

01:47:42   is force quitting it.

01:47:44   And some applications, even when they're working perfectly, you need to force quit right when

01:47:47   you're done with them if you don't want them sucking your battery down.

01:47:50   True.

01:47:51   All true.

01:47:52   And yet, I still say, the reflexive habit of force quitting every single application

01:47:55   every time you're done with it is crazy because you are draining your battery more because

01:47:59   you're just relaunching them fresh the next time you use them, which is going to be 30

01:48:01   seconds from now when you launch your Twitter app again.

01:48:04   What I'm against is the reflexive routine force quitting of everything and that is the habit

01:48:09   I see not the selective based on past experience that I have to force quit this application or drains my battery

01:48:14   No, that's not how people act most people. It's just like flick flick flick flick flick use flick use flick use flick

01:48:19   They're double tapping that home button like crazy

01:48:22   They cannot have a single they can only ever have one thing there when they're done with it

01:48:25   They flick it so they can go back to springboard and see nothing

01:48:27   it's you know, it's

01:48:32   He's saying it's not about that it's about apps that won't launch

01:48:34   I said sometimes you they get into a state where you bring them to the foreground

01:48:37   They do nothing and you have to force quit them to be able to launch them again

01:48:40   I understand that and yes, sometimes the OS gets so hosed that you got a reboot

01:48:44   but

01:48:45   None of this argues for you must reflexively force quit every single application on your phone every time you use it and which again is

01:48:52   Exactly what I see and it's what I see my son doing and trying to reason with him

01:48:55   It's not worked and he still does it and it's an embarrassment

01:48:59   - Okay.

01:49:01   - Good talk.

01:49:01   - No, I mean like, any reason that,

01:49:04   any like legitimate reason that you'd need

01:49:07   to force quit all these apps all the time

01:49:09   is probably either a bug or a shortcoming

01:49:13   in the operating system.

01:49:14   And so like, yeah, Apple should fix those.

01:49:16   Like, it is totally valid today to say,

01:49:20   I need the ability to force quit apps

01:49:23   because sometimes they don't launch right or whatever.

01:49:25   - Oh, you're always gonna need the ability,

01:49:27   but just for it to be a 100% blanket habit

01:49:30   is not a good idea.

01:49:32   - Yeah, I mean like the Apple Watch,

01:49:34   this is not a good example in general

01:49:36   of how to design a responsive and stable software platform,

01:49:39   but the Apple Watch has a way to force quit apps

01:49:42   without having an app switcher.

01:49:45   Most people don't know about it, but you can do it.

01:49:47   You basically hit the sleep button

01:49:49   and then you hit it again, you look it up.

01:49:51   But yeah, it has a way to do that.

01:49:55   So like it is possible to still have some kind of gesture involving the sleep/wake button

01:50:00   and holding it down in a certain way to have a force/quit method without putting it in

01:50:08   the switcher and having that be the method.

01:50:09   But I don't know.

01:50:10   I think the whole design of the multi-tasker switcher itself needs a lot of help in a lot

01:50:15   of ways, not least of which is that conceptually everybody thinks all those apps are always

01:50:19   running all the time and they need to clear them out.

01:50:21   So that's one of the many problems

01:50:23   with the current iOS app switcher

01:50:25   that is ultimately a design problem, not a technical one.

01:50:29   - Yeah, and the technical problem is when, you know,

01:50:31   an app gets stuck in that state where you have to kill it

01:50:33   'cause it won't launch again.

01:50:34   Like it will launch and you can, you know,

01:50:36   run it and use it, but it's useless,

01:50:38   or like you'll tap the icon and nothing will happen.

01:50:41   Yeah, and then like, or just memory gets filled

01:50:43   or corrupted or there's an OS bug, that happens,

01:50:45   but none of that argues for, happens all the time,

01:50:48   first of all, you've got a problem with your phone

01:50:49   and it's obviously not a widespread problem

01:50:53   because plenty of people use their phones.

01:50:54   And like, I don't force-quit anything ever essentially,

01:50:57   'cause I don't use the Facebook app, I don't use anything,

01:50:59   and I use iOS devices for years,

01:51:01   not force-quitting everything.

01:51:01   So it's obviously not an endemic bug

01:51:03   that affects every single device.

01:51:05   If it's happening to yours, who knows what's going on,

01:51:07   maybe you have a problem, but don't accept it

01:51:08   as the status quo as things working correctly.

01:51:11   And even when you do have a problem like that,

01:51:13   feel like you can target it

01:51:14   by figuring out which applications are the problem,

01:51:16   when do I need to do this?

01:51:18   Because if you can't do that,

01:51:19   You just give up and say I have to do this all the time. It's just the easiest system that works

01:51:24   Like I feel like the total failure of the product like it's not the users fault at that point

01:51:28   It's just the product is so frustrated the user that they can't that this is their tool for dealing with it

01:51:34   Which is blanket force quitting and they will never move away from it because it's the only thing that's giving them a salvation on

01:51:38   Whatever weird thing is going on on their phone or the apps they use

01:51:41   and I feel like if Apple saw you do that they would feel that they have failed you as a

01:51:46   vendor of a product because they don't want you to use it that way and you shouldn't have to use it that way and if

01:51:51   You feel like you do whether it's true or not whether you have to if you feel like you do

01:51:54   That's a that's a breakdown there

01:51:56   But anyway, I would say if you if you are listening this and you respectfully force quit all your apps

01:52:01   consider

01:52:04   Trying to target your force quitting trying to target it better and maybe if there's an application that's a particular problem

01:52:10   Get it off your phone. I know you can't tell people to get rid of Facebook

01:52:13   It's like telling people, I don't know,

01:52:15   like give up their firstborn.

01:52:17   But if that's not an option for you

01:52:20   and a Facebook is a problematic one,

01:52:22   just try reflexively force quitting Facebook,

01:52:24   but don't say reflexively force quit

01:52:26   your favorite Twitter app because it's probably fine.

01:52:28   Or maybe your Twitter app is the problem, I don't know.

01:52:29   I just, all I'm telling you is that it's not normal.

01:52:32   Like you shouldn't have to do this.

01:52:34   And I know many people who spend years and years

01:52:38   using iOS devices across multiple hardware models

01:52:40   and multiple OS models with multiple apps

01:52:43   who don't even know how to force quit, and they're fine.

01:52:45   So it is not like a fact of life on iOS, and people should not be doing it.

01:52:50   Please.

01:52:51   [BEEP]