PodSearch

ATP

115: Empathy for the Machine

 

00:00:00   Learn about the cat game.

00:00:02   It's more boring than I thought it was.

00:00:04   I thought you get to play with the cats or something.

00:00:06   I thought it was interactive in some way, but they just show up.

00:00:08   Yeah, it's just like real cats.

00:00:10   You don't really interact with them.

00:00:11   You just kind of feed them.

00:00:13   Right.

00:00:14   You're just so excited that they exist and that you exist at the same time.

00:00:17   Oh, look, a cat!

00:00:18   And that's it.

00:00:19   You don't get to play with the cats.

00:00:20   You don't get to throw them toys.

00:00:21   It's just they show up, leave you crap, and then leave.

00:00:25   Yeah, they don't really care about you.

00:00:27   They just care about your food.

00:00:29   - And you can't even interact with them.

00:00:31   - Right, just like real cats.

00:00:33   - Don't like it.

00:00:33   - Don't like cats.

00:00:34   (electronic beeping)

00:00:36   - Are we talking about this Microsoft stuff at some point?

00:00:38   - I think we should, even though in the grand tradition

00:00:41   it's like happened today and nobody knows about

00:00:43   anything about it, but whatever, we all read headlines,

00:00:45   right, it's all it takes.

00:00:47   - Yeah, we're pundits, all we need is a headline.

00:00:49   - Well, this is pretty much all you guys

00:00:50   because I was really not following any of this.

00:00:53   I was using my Apple Watch all day,

00:00:55   which you guys can't say anything about.

00:00:57   (laughing)

00:00:58   - Damn it, Marco.

00:00:59   - But I don't know anything about the Microsoft stuff.

00:01:01   So I think we have a lot of stuff to cover

00:01:03   from subsets of us.

00:01:04   - Yeah, we'll be taking turns.

00:01:06   And we might as well start, as we always start,

00:01:09   with Jon taking the turn of follow-up.

00:01:12   - Yeah, I thought we had a lot of follow-up this week,

00:01:13   but now that I read through it, it's just long,

00:01:15   but I don't think it will take a long time

00:01:16   'cause they're all small items.

00:01:18   - Famous last words.

00:01:19   - Everyone look at the timestamp.

00:01:21   - Oh, I was right last time I said that.

00:01:23   We got through it quickly, I think we will today too.

00:01:25   - Okay.

00:01:26   - So the first one is from Jon Tall,

00:01:28   giving us more color, as they say in Apple earnings calls,

00:01:32   about the Seagate, the bad Seagate hard drive thing

00:01:37   that we talked about.

00:01:38   And this is from 2011.

00:01:41   And he says it was actually a bug in their firmware.

00:01:44   And specifically, it was a bug in a counter

00:01:47   that kept track of how many times the drive had spun up.

00:01:50   And the counter would overflow,

00:01:51   and it would make the drive think that it was brand new,

00:01:55   that it had never been spun up.

00:01:57   And the reason the firmware would use this count is,

00:02:00   apparently when it's brand new,

00:02:01   it spins up really quickly.

00:02:03   And when it gets older,

00:02:04   it takes a little bit longer to spin up.

00:02:06   And the reason the hard drive wants to know

00:02:07   how long it takes to spin up,

00:02:08   is it wants to know when it's safe

00:02:10   to put the heads on top of the disc.

00:02:11   If you, we should put the link

00:02:13   to the Wikipedia hard drive page,

00:02:14   or maybe a how stuff works page or something.

00:02:17   If you don't know how hard drives work,

00:02:19   they have these little heads that look kind of like the,

00:02:21   what does it call the, in a phonograph case,

00:02:24   the tone arm or whatever.

00:02:26   - Well, the stylus. - Something like that.

00:02:27   Well, the stylus is the needle,

00:02:29   but I think you are talking about the tone arm, I think.

00:02:32   I don't even pay that much attention

00:02:34   as much as I have this reputation

00:02:36   for being obsessed with vinyl, I'm not even sure.

00:02:38   - You know, hard drives sound better than SSDs.

00:02:40   - Oh my God.

00:02:41   For people who don't know how hard drives works,

00:02:43   trying to explain it as an analogy for records,

00:02:46   which they probably might have never seen,

00:02:48   and even I who grew up with records

00:02:49   don't know what the heck the parts are called.

00:02:50   Anyway, it's a little arm that goes over the disc,

00:02:53   and it's got a little reed head on it

00:02:55   there's usually top and bottom of the disk and it's got a little read head that reads

00:02:58   a little, you know, the magnetic thing that reads little blips on the disk of magnetic

00:03:03   poles or whatever.

00:03:06   That head is really, really close to the disk.

00:03:08   That head cannot touch the disk because if it touches the disk it ruins it because the

00:03:12   disk is spinning very fast and it's made of, I don't know, some kind of metal, glass, whatever

00:03:17   material in the head is actually pretty hard and if it touches it, that's a head crash

00:03:20   that will put a big gouge in the disk and you basically ruined it.

00:03:23   and that's where the term crash in computers comes from.

00:03:26   - Is it where it comes from?

00:03:27   - Yeah, it's hard drive heads crashing into the bladders

00:03:30   and therefore killing the drive and you lose your data.

00:03:33   And originally when you say my computer crashed,

00:03:36   that was what you meant.

00:03:37   - I'm not sure if that's true, so listeners,

00:03:39   you can come in and tell us whether that's crazy.

00:03:41   All I know about is the bug that was the moth

00:03:42   that flew into the big machine and got caught

00:03:44   in the trip as the original bug.

00:03:45   But anyway, so it wants to know when is it safe

00:03:48   to bring the heads onto the disks,

00:03:50   'cause when the disk is off, the heads are off on the side.

00:03:52   They're not over the discs at all.

00:03:54   They're parked.

00:03:55   Exactly.

00:03:56   In the parking lot.

00:03:57   Why does it matter how fast the disc is spinning?

00:03:59   Can't you just bring the heads out whenever the heck you feel like it?

00:04:01   No, because the thing that keeps the heads from touching the disc is a tiny cushion of

00:04:05   air between the head and the disc.

00:04:07   And it's super tiny.

00:04:08   Like they always show the diagram that shows like a human hair next to the gap.

00:04:11   And the human hair dwarfs the gap between the heads and things.

00:04:14   So they have to be really really close, but can't touch.

00:04:17   And so it's really important that the disks be spinning fast enough for the heads to be

00:04:23   able to safely move on to it.

00:04:24   And if that counter overflows and the mechanism, the hard drive thinks that it's a brand new

00:04:29   disk and it's probably spun up right away, it sends the heads out, they crash into the

00:04:32   disk and it kills them.

00:04:33   And so this is a firmware bug that causes a hardware failure that affected a lot of

00:04:38   Seagate drives, apparently.

00:04:40   All right.

00:04:41   We will put a link to HowStuffWorks that on page 95 of the slideshow.

00:04:47   What page was that?

00:04:49   On page, I think it's seven of that slideshow, they show what Jon is describing.

00:04:54   So we will have that in the show notes.

00:04:56   I wanted to jump in and do a very brief piece of follow-up.

00:05:01   I am talking to everyone on Aaron's MacBook Air, the one that went for a little swim last

00:05:06   week.

00:05:07   So this is going to be a short show?

00:05:08   Yeah, it's going to be a real short show.

00:05:11   It actually, knock on my glass desk,

00:05:15   has been doing just fine.

00:05:17   I'm not expecting that to continue on forever.

00:05:19   I'm expecting it to just, bzzt.

00:05:22   - Well, I guess we lost Casey.

00:05:24   - (laughs) I'm just kidding.

00:05:25   - Anyway.

00:05:26   - Yeah, moving on.

00:05:27   But no, all kidding aside, it is fine.

00:05:30   I am surprised by that.

00:05:31   I have heard several conflicting reports

00:05:33   as to how much Apple would charge me to fix this

00:05:36   if it does eventually fry.

00:05:38   I've heard a couple hundred dollars.

00:05:41   I've heard, "Well, maybe a genius will take pity on you."

00:05:44   I've heard, "$800 and anywhere in between."

00:05:48   So hopefully I'll never have to find out,

00:05:50   but as of tonight on Wednesday, what is it?

00:05:54   The 28th, 29th, or something like that, 29th.

00:05:58   Wednesday, the 29th of April,

00:05:59   it is still working much to my surprise.

00:06:01   So I'm very happy about that.

00:06:03   - Oh, you made that website though,

00:06:05   like the Death Watch website, right?

00:06:06   - Yeah, I did.

00:06:08   I did actually make a Death Watch website.

00:06:10   And what I did was I found a site somewhere

00:06:14   that would create a launch DP list for me.

00:06:17   And that launch DP list would then just launch cURL

00:06:21   and tickle a URL on my website, but it's very unreliable.

00:06:25   And so part of the reason I haven't publicized this link

00:06:28   is because it reports Aaron's Mac as being broken

00:06:33   and dead way more often than it actually is broken and dead.

00:06:38   - Yeah, I went to it and I saw that it was dead too.

00:06:40   And I said, "Well, for a second I was upset, and then I said, you know what, that's probably

00:06:44   just a monitoring failure."

00:06:45   Exactly.

00:06:46   And so the reason I haven't given this to the internet is because as much as I know

00:06:50   everyone would be doing it out of the goodness of their hearts, and I'm not being sarcastic,

00:06:53   I mean that, I would be getting hourly, if not minute-by-minute reports that Aaron's

00:06:58   computer is dead, I must check on it, go look right now, oh my God, it says it's dead.

00:07:02   When in reality, it's probably user error with me and LaunchD, but one way or another,

00:07:07   It's just not a very reliable mechanism for monitoring the Mac.

00:07:11   So there is a website for it.

00:07:12   I will not be sharing the link because it's not a very good link and it's not worth paying

00:07:16   attention to.

00:07:18   But it is alive.

00:07:19   And did you, you said that you went to a website to generate a launchd file.

00:07:24   Is that right?

00:07:25   So I don't know anything about launchd and I was originally going to use cron.

00:07:28   I was going to set up a cron job, but I can never get cron syntax right.

00:07:32   And so I was figuring out, okay, well, what do I have available to me?

00:07:36   You know, can I just do a scheduled task sort of thing?

00:07:38   And maybe there's an easier way of doing this that I just didn't think of or stumble upon but one way or another

00:07:43   I ended up going to this website that generated this just silly

00:07:48   Humongous p-list for launch D. And again, I don't really know what I'm talking about, but

00:07:56   apparently what launch D does is it you give it this p-list and you say on the

00:08:02   fifth minute of the zeroth hour

00:08:05   go run this command. On the 10th minute of the 0th hour, go run this command. On

00:08:09   the 15th minute of the 0th hour, and this continues through all 24 hours of the

00:08:13   day. And so because of that, this plist was enormous. Maybe that's not how

00:08:18   it's designed to be used. If it's not, it doesn't really matter. To be honest, I

00:08:22   don't really care, but I tried to do that real quick and I thought it was working,

00:08:26   but not so much. I put a link in the show notes to an app that will make launchd

00:08:31   P-lists for you because there are much more verbose.

00:08:35   And Casey's saying that he didn't know cron syntax

00:08:38   off the top of his head is the most Windows developer thing

00:08:40   I've ever heard him say.

00:08:41   (laughing)

00:08:43   I know it and I, yeah.

00:08:44   Yeah, so anyone who has ever done server side development

00:08:47   on any platform other than Windows knows how to write cron.

00:08:50   - Hi, John.

00:08:51   Well, and I find Launch D's syntax completely inscrutable.

00:08:55   - Yeah, now that's why there's like, yeah,

00:08:56   I don't think Casey's crazy to get help doing that

00:08:59   because it is like cron is easy.

00:09:00   It's just a bunch of items and they're in a weird order and you memorize it and if you

00:09:04   forget man5 crontab to refresh your memory, whatever, to see if this version supports

00:09:09   the /5 syntax or not.

00:09:11   Anyway, the lunchd stuff is way more complicated and I bet only the people who develop lunchd

00:09:16   at Apple have memorized all the options and all the different things you can add there.

00:09:20   And even if you're going to do it quote-unquote by hand, you'd use Apple's plist editor because

00:09:24   who writes plist by hand, right?

00:09:26   Animals.

00:09:27   Yeah.

00:09:28   There's the thing it's lingon or whatever is named after the the barrier the fruit

00:09:32   I have a version of it somewhere on here

00:09:34   Be careful because you can screw up your system with that like it lets you see I think all the p-lists for like the system

00:09:40   Wide things and if you're like what is this? I don't understand this I can delete that right just you know be careful out there

00:09:44   Does it taste like IKEA? Yeah, it's nice. I will put the

00:09:49   Site I used in the show notes as well as well

00:09:52   It's launched zero with calm and again

00:09:55   All it does is you say, "When do you want this thing to run?"

00:09:57   And it generates this humongously verbose launch DEP list.

00:10:00   - Photos. - Photos.

00:10:03   - Yep, this is from Andrew Woods.

00:10:05   He is making sure that we know that Photos

00:10:08   does a whole bunch of stuff in the background

00:10:10   even after it's done importing.

00:10:12   Said that an activity monitor on his 2010 MacBook Pro

00:10:15   showed it pegging all four cores,

00:10:18   all four virtual cores for a long time.

00:10:20   So his recommendation was to check activity monitor

00:10:23   Wait until things have settled down and things really are idle before judging his performance as I think I said in the last show

00:10:29   I did like I'm looking at activity monitor

00:10:31   I'm making sure that it's that I don't see the cloud deep process still grinding away or whatever. You know, I

00:10:36   Let it I really let it get into that steady state over the course of many days where I didn't see anything in top

00:10:42   Or whatever and then just try to do this down. So I was judging it

00:10:44   After it had gone to an aisle state

00:10:46   It's even worse

00:10:47   Of course if it's in the middle of doing face recognition or even uploading to iCloud or anything like that

00:10:52   but that's something if someone if you are trying photos for the first time and you're not accustomed to launching activity monitor and

00:10:58   Sorting by CPU and seeing if anything is grinding away

00:11:01   There will be a period of time after the import is quote unquote done when things are still going on

00:11:06   So you should let it stew

00:11:08   See if you ran I stat menus you would know immediately

00:11:13   Whether these things were taking up all your CPUs or not

00:11:16   No, because it doesn't tell you which process is hurting your CPU for all you know

00:11:19   - Oh, it could be kernel_task,

00:11:21   and then that doesn't help you much, does it?

00:11:23   I see nothing about it.

00:11:24   - No, you just click on the thing

00:11:25   and it shows you all the top processes.

00:11:27   - Yeah, I just click on the dock icon

00:11:29   and it shows me all the top processes too,

00:11:30   but including their names and I can sort them.

00:11:32   And when I'm done looking at it, I quit the app,

00:11:34   so it's not there updating every three seconds.

00:11:36   - Oh, Jon.

00:11:37   - Scanning the entire process table

00:11:38   and the operating system,

00:11:39   acquiring and releasing kernel locks

00:11:41   over and over and over again with nobody looking at it.

00:11:44   (laughing)

00:11:47   This is called empathy for the computer.

00:11:49   (laughing)

00:11:50   - That's what you're trying to instill

00:11:51   and programmers should all have empathy for the machine

00:11:53   because it makes you write code that isn't stupid.

00:11:56   (laughing)

00:11:58   - We're done, I don't even know where to go from here.

00:12:00   - Thanks a lot to our three sponsors this week.

00:12:03   That's a real thing though, empathy for the computer.

00:12:05   I keep saying that and people look at me like I'm crazy,

00:12:07   but don't you guys feel that to some degree

00:12:09   when you write code that like, I don't know,

00:12:11   just something as simple as like hoisting an invariant

00:12:13   out of a loop, why are you doing that?

00:12:16   You just do it like, 'cause you have to sort of,

00:12:18   put yourself in the position of the computer

00:12:21   and of the different machinery

00:12:22   and you just feel guilty about making some process

00:12:24   be repeated over and over again

00:12:27   when you know the result is going to be the same.

00:12:29   It's like you're just wasting everybody's time.

00:12:30   You're just wasting battery.

00:12:32   You're wasting, I mean, it's negligible, it's stupid,

00:12:34   it's premature optimization, blah, blah, blah.

00:12:35   But if you have no empathy for the computer,

00:12:38   then you just do everything in a ridiculous way

00:12:40   and your entire program is just one giant gray haze of,

00:12:44   there's no hotspot.

00:12:45   That's just like a complete soup of super slowness.

00:12:49   So I think empathy with the computer is an important skill that programmers should acquire.

00:12:54   So why do you program in a scripting language?

00:12:57   Because you have empathy with the computer, but you don't want to be flipping toggle switches,

00:13:01   right?

00:13:02   You don't want to be setting your bits and magnetic core memory by hand, right?

00:13:05   It's a balance, but no matter what level you're programming at...

00:13:08   There's nothing between those two.

00:13:10   Well, no, what I'm saying is like, it's even more important when you're in a higher-level

00:13:13   programming language, right?

00:13:14   Now, to some degree, you don't have control over some things.

00:13:17   But if you know that, for example,

00:13:18   if you're in a high-level programming language

00:13:20   and function calls are relatively expensive compared

00:13:22   to how fast it is to, say, jump to an address in an assembly

00:13:25   program, you're like, do I really

00:13:27   need to get all the arguments into the right registers,

00:13:31   to put the address I want to jump into the right register,

00:13:33   to jump to that register, to do all the stuff for the return

00:13:36   value, put that into the right place, return, pop the stack,

00:13:39   make a new stack frame, do all that stuff.

00:13:41   Do I really need to do that, or could I inline that here?

00:13:42   Like, that is empathy with the compiler.

00:13:44   and at every level that works.

00:13:46   And in a higher level language,

00:13:47   you have to know which things are slow

00:13:48   in your high level language and which things,

00:13:50   which things can you avoid and which things can't you.

00:13:52   And in any language, it doesn't matter what the language is,

00:13:55   high level, low level, medium level,

00:13:55   there's some things that just apply all the time.

00:13:57   Don't do something over and over again

00:13:59   when the result is gonna be the same.

00:14:01   Don't make copies when you can pass around references

00:14:04   if your language supports that concept.

00:14:07   It doesn't really matter how it's implemented.

00:14:08   You could have, "Well, I can make a copy,

00:14:09   "but under the covers I know the language is smart enough

00:14:11   "to do copy on write and it's really efficient,"

00:14:13   or whatever, but I don't know.

00:14:14   I guess it comes down to somewhat knowing

00:14:16   how your language is implemented,

00:14:17   but I think it's more important

00:14:18   when you get into higher level languages, not less, right?

00:14:21   Like, because the higher level languages have the potential

00:14:23   to put something innocuous that,

00:14:25   if you knew how the language is implemented

00:14:27   or know what is particularly slow in this language,

00:14:29   you'd be like, "Oh man, really? Seriously?"

00:14:30   Whereas when you're down at assembly,

00:14:32   you might feel some empathy, but in general,

00:14:34   almost everything you do is fast,

00:14:36   there's very little you can write a line at a time

00:14:37   that's gonna make things super slow.

00:14:41   - Wow, this took a turn, I was not expecting.

00:14:45   - Our first sponsor this week is Backblaze.

00:14:48   Go to backblaze.com/atp.

00:14:51   Backblaze is unlimited and unthrottled online backup

00:14:56   for just five bucks per month for a whole computer

00:14:58   or anything plugged into it.

00:15:00   Backblaze is awesome.

00:15:02   I use them myself, I've used them long before

00:15:05   they were a sponsor.

00:15:06   My wife uses them, my mom uses them,

00:15:09   everybody I know who I can convince to use them,

00:15:12   which isn't very hard, everyone uses a back blaze,

00:15:16   except Casey, and it is so great because,

00:15:20   you know, online backup in general,

00:15:21   this protects you from so many great,

00:15:23   so many problems that you might not think might affect you,

00:15:27   but when you start thinking about

00:15:28   your current backup strategy,

00:15:29   like if your backup strategy is a time machine drive

00:15:32   connected to your computer,

00:15:33   then anything that happens to the environment of that

00:15:38   can disturb and kill both of them.

00:15:40   So things like a fire, a flood, theft,

00:15:43   power surges, stuff like that,

00:15:44   all those things can take out your computer

00:15:47   and anything that's plugged into it.

00:15:48   Online backup is great for just peace of mind.

00:15:51   Just like to have that final fail safe

00:15:53   in your backup strategy,

00:15:54   if you know that's gonna be there no matter what.

00:15:57   And Backblaze is by far my favorite

00:16:00   of the online backup services.

00:16:02   So they just released a new version, 4.0.

00:16:05   This is now, it now has multi-threaded backing up,

00:16:09   which allows for even faster upload speeds.

00:16:11   And this especially helps if you are somewhere

00:16:14   in a part of the world where you're far from

00:16:17   one of their CDN nodes and, you know,

00:16:22   because of latency with uploading tons of small files

00:16:24   and everything, having multi-threaded things,

00:16:26   being able to run amongst these things in parallel

00:16:28   really helps upload speeds if you're around the world.

00:16:32   Backplays is huge, they've been around forever.

00:16:34   They have over 150 petabytes of data backed up.

00:16:38   Over 10 billion files have been restored.

00:16:41   Not just backed up.

00:16:42   Over 10 billion files have been restored by their customers.

00:16:45   They know the backups work.

00:16:46   They are tested constantly.

00:16:48   Backblaze client runs native.

00:16:51   This is native code on Mac and Windows PCs.

00:16:54   This does not use Java or anything else.

00:16:56   It is real native code.

00:16:58   So it's fast.

00:16:58   It's efficient.

00:16:59   You don't have to install any garbage runtimes.

00:17:02   You can access your data in Backblaze.

00:17:04   anywhere you are, if you log into a web browser

00:17:07   or if you use their app on iOS or Android,

00:17:10   you can log in and pull a file off that,

00:17:12   maybe you're on a trip and you wanna get a file

00:17:14   off your home computer that's backed up to Backblaze,

00:17:16   you can just go do that, you can restore one file

00:17:18   onto your device.

00:17:20   About 25% of their restores are for just one file

00:17:24   to do this exact use case.

00:17:26   It's very, very easy and if you need everything back,

00:17:28   everything you've backed up, if you have a total loss

00:17:30   and you need everything backed up,

00:17:32   You can download it.

00:17:33   You can even order a USB hard drive with everything on it.

00:17:37   So stop putting this off.

00:17:39   Look, backing up is so important.

00:17:41   And with online backup, it is easier than it's ever been.

00:17:44   I really cannot recommend this enough.

00:17:46   You really, really should have online backup.

00:17:48   And if you're gonna have online backup,

00:17:49   the one I recommend is Backblaze.

00:17:51   So stop putting it off.

00:17:53   Start getting back up today.

00:17:54   Backblaze.com/ATP.

00:17:56   Five bucks a month per computer for unlimited space

00:18:00   and unthrottled speeds.

00:18:02   There's no add-ons, no gimmicks, no additional charges,

00:18:04   no upsells.

00:18:05   Five bucks per month per computer.

00:18:08   Unlimited space.

00:18:09   That's all you need to know.

00:18:10   Risk-free, no credit card required trial.

00:18:13   Go to backblaze.com/atp.

00:18:15   Really, check it out.

00:18:16   Risk-free trial, just go today.

00:18:17   Backblaze.com/atp.

00:18:19   Please do this.

00:18:20   You will thank me someday when you need it.

00:18:22   Trust me, it is so, so worth it.

00:18:24   Backblaze, thanks a lot.

00:18:26   - All right, so John, tell us about hard links.

00:18:28   - Yeah, there's a little bit of confusion

00:18:30   from people about hard links.

00:18:31   I don't think I did a good job of explaining it last time.

00:18:33   I just assumed everybody knew.

00:18:35   Did we put the hard links link in last week's show notes?

00:18:38   I forget.

00:18:39   - I don't remember doing so, but I'm not 100% sure.

00:18:42   - Anyway, we'll put it in this one.

00:18:43   So there was a tweet for someone saying,

00:18:45   about that photos.app creating hard links

00:18:47   instead of an entirely new library

00:18:49   and then sent me a screenshot showing the photos library

00:18:52   right next to the iPhotos library saying,

00:18:53   yeah, I thought you said it wasn't gonna make

00:18:55   a whole new library, but look,

00:18:56   this here's a whole new library.

00:18:57   I'm like, yeah, but like the other one is full of hard links.

00:18:59   And he said, really?

00:19:00   It says each one is 45 gigabytes.

00:19:02   So that's the thing about hard links.

00:19:03   We'll put the link in the show notes

00:19:05   so you can read this long thing

00:19:06   for one of my old OS X reviews about it from back in 2007.

00:19:09   But the short version is, if you've got a file on disk,

00:19:12   you can think of it as a blob of data

00:19:14   that's sitting on your disk somewhere,

00:19:15   and it has a name associated with it,

00:19:17   and that's how you can find that data.

00:19:20   This name is, you know, you go through this name,

00:19:22   foo, whatever, I don't know.

00:19:25   Foo.txt, it leads to this set of bytes.

00:19:27   That name attached to that set of bytes can be said to be a link.

00:19:32   It's how you get to those bytes on disk.

00:19:34   Oh, well, I look them up by going foo.txt, right?

00:19:37   You can make another name that points to that same set of bytes on disk,

00:19:41   called bar.txt.

00:19:43   Bar.txt is one way to get to that same blob of bytes,

00:19:45   and foo.txt is one way to get to that same blob of bytes.

00:19:47   Still just one blob of bytes,

00:19:49   two different names that point to it.

00:19:51   Those are hard links.

00:19:52   So now that set of data on disk has two hard links.

00:19:56   Every file can be said to have one hard link,

00:19:59   conceptually in the Unix prompts,

00:20:00   not in the HFS push implementation,

00:20:01   which we don't really want to talk about.

00:20:03   But anyway, conceptually,

00:20:05   every file in disk has one hard link,

00:20:07   and that's like the name of the file.

00:20:09   If you make a second hard link,

00:20:10   we call that second one, oh, that's the hard link.

00:20:12   The first one was just a file, right guys?

00:20:13   But the second one is a hard link to the file, right?

00:20:16   Really, if you want to think of it correctly conceptually,

00:20:20   the first one is also a hard link.

00:20:22   There's a little number if you type ls

00:20:24   at the Unix command, a shell prompt,

00:20:26   The ls command will show you a number of links to this thing.

00:20:29   Most files have a one next to it.

00:20:30   If you make a hard link to a file, it'll go up

00:20:33   and you'll see two, that's what this article shows.

00:20:35   And since they're pointing to the same bucket of data,

00:20:40   this gets more complicated.

00:20:40   If you use an editor that edits the file in place,

00:20:44   that does not do what many editors do,

00:20:45   which is make a complete copy of the file

00:20:47   and then rename it on top of the file

00:20:48   or do some other update.

00:20:49   If you use an editor that actually edits the file in place

00:20:52   and you edit through either one of those names,

00:20:54   you will be editing the same,

00:20:56   like it's just one bucket of bits.

00:20:57   If you change it and then you look at the contents

00:21:00   of foo.txt and bar.txt,

00:21:02   they both point to the same blob of bits.

00:21:03   If you edit it through bar.txt, it changes the same bits.

00:21:06   You could edit bar.txt and then show the output

00:21:09   of what's inside foo.txt and it will show your edits, right?

00:21:12   That's how hard links work.

00:21:13   If you get info in any of those files,

00:21:16   if you list the file with ls,

00:21:18   if you get info in the finder,

00:21:19   they'll all say, how big is it?

00:21:21   They'll say, well, how big is the big bucket of bits?

00:21:23   Oh, it's like, it's two megs, it's 25 megs, it's five gigs.

00:21:26   They'll both say that.

00:21:28   The tricky part is that if you have,

00:21:30   say you had a hard drive that was like 100 megs

00:21:32   and you had a 99 meg file in it,

00:21:34   you make 75 hard links, that 99 meg file,

00:21:37   and every single one of those 75 files will say,

00:21:40   I'm 99 megabytes, I'm 99 megabytes.

00:21:42   It's like, how can you have this many 99 megabyte files?

00:21:44   The disk is only 100 megabytes.

00:21:45   How could they all fit?

00:21:46   It's just one set of 95 megabytes.

00:21:48   Each hard link does not add any data.

00:21:50   So when Photos.app makes a new Photos library,

00:21:54   and you see it right next to your iPhoto library,

00:21:56   inside both of those libraries are a bunch of,

00:22:00   one of them has the quote unquote originals in it.

00:22:01   In HFS Plus there really is an implementation difference

00:22:03   under the cover.

00:22:04   And the other one just has hard links.

00:22:05   But if you get info on both of them in the final,

00:22:07   they'll both say that they're the same size

00:22:09   after they're done roughly.

00:22:10   Like, oh, I'm 45 gigs and so am I.

00:22:12   But there's only one set of 45 gigs

00:22:14   worth of photo data on your disk.

00:22:15   Now, there are files that are private to the Photos library,

00:22:17   like the metadata databases and so on and so forth,

00:22:19   but the photos themselves,

00:22:20   that's what it's making hard links to.

00:22:22   Those are the things that take up

00:22:22   the majority of the room.

00:22:24   So yes, it is very confusing.

00:22:26   And if you try to do the math,

00:22:27   if you just get info on these libraries

00:22:29   and try to do the math and say,

00:22:30   "Okay, so I've got a 45 gig library

00:22:32   and a 45 gig library.

00:22:33   That means I'm using 90 gigs of space

00:22:35   and my total drive space,

00:22:36   like the math won't add up."

00:22:39   That's just how hard links work.

00:22:40   And to make things more complicated,

00:22:42   as I said in the last show,

00:22:43   as you edit files with the Photos app,

00:22:46   it will take something that was a hard link

00:22:48   and change it into a copy because it doesn't want to mess with your iPhoto library.

00:22:52   So if you make a modification to the actual photo file, which I don't even know if it

00:22:55   happens, maybe if you change the geotag data, which as Casey said you can't even do at this

00:22:58   point, but if you were to change that locally, it will make a copy of it and they will slowly

00:23:04   diverge from each other.

00:23:05   So they do that to keep them separate.

00:23:06   That's not a property of hardlinks, the application itself is doing that.

00:23:10   The application of course itself isn't free to do anything like that.

00:23:14   That has nothing to do with the nature of hardlinks or anything.

00:23:16   As I said, as you see in the demo link, if you have two hard links to the same blob of

00:23:20   data and you go through either one of them to edit that data, they both see the changes.

00:23:23   The photos application itself prevents that from happening sort of manually by making

00:23:27   a copy when it needs to.

00:23:29   So it's confusing.

00:23:31   Hard links, this talks about sim links too, which are a different thing, photos and creating

00:23:35   confusion for people who now can't tell how much free space they have on their disks anymore.

00:23:39   All right.

00:23:40   So how about shared albums?

00:23:42   Yeah, this was a bummer.

00:23:44   Lance T Hildebrand said, "Even if you try to do that crazy system," I said last time

00:23:48   of like throwing all your photos into the family photo library that's shared with your

00:23:51   whole family, apparently you can only have 5,000 photos in any shared album.

00:23:55   So that wouldn't have worked anyway.

00:23:57   Which kind of makes sense, like you shouldn't be using shared albums as a way to share your

00:24:01   whole library, there should just be a way to share your whole library and have it read/write,

00:24:04   but there isn't, and unless you have fewer than 5,000 photos you can't.

00:24:09   You can't even like manually do it.

00:24:12   How about what goes on with a MacBook Air with 7800 photos?

00:24:17   - Yeah, we talked about performance last time

00:24:19   and it was like, well, I'm using a 2011 MacBook Air,

00:24:22   maybe it's just slow because I don't have a lot of RAM,

00:24:24   I have a slower computer.

00:24:26   So, Mucho Spanish says that he has a very small library

00:24:31   on a 2012 MacBook Air, just 7800 photos

00:24:34   and keywording is still slow.

00:24:36   And then our friend Will Haynes says,

00:24:38   very slow for per photo operations

00:24:41   and he's on a maxed out iMac 5K.

00:24:43   Also, he couldn't sync iCloud at all.

00:24:45   - I just am not really sold on why this is something

00:24:48   I need in my life right now.

00:24:49   Like the idea of being able to get to any of my pictures

00:24:53   in the photo chooser, especially on iOS,

00:24:56   that sounds appealing in principle,

00:24:58   but hearing how much it slows down your iOS device

00:25:02   anytime you go to the photo chooser,

00:25:03   that sounds freaking terrible.

00:25:05   And just a lot of this,

00:25:07   I'm not saying I'll never use photos

00:25:08   by any stretch of the imagination,

00:25:09   but it just does not sound like it's something

00:25:11   I want in my life right now.

00:25:13   - I don't know, it's working great for me.

00:25:15   The only problem that I'm seeing with it

00:25:17   is what many people have reported,

00:25:19   and Jon, I think you said this too,

00:25:20   which is that the photo picker takes a long time

00:25:23   to come up for anything that lists the photos.

00:25:26   So the biggest time I see this is

00:25:29   after I've taken a new picture on the iPhone,

00:25:32   when I want to tap on it to bring it up

00:25:34   to view the picture I just took,

00:25:36   there's like a five or 10 second delay

00:25:38   where just nothing happened.

00:25:39   you think the phone is frozen, and then it comes up.

00:25:42   That's the only time I'm really seeing issues.

00:25:44   Besides that, which is annoying,

00:25:47   but not that major of a problem,

00:25:50   it's just a frequent annoyance,

00:25:52   besides that I've had no problems,

00:25:54   and that's the kind of thing that I would imagine and hope

00:25:56   that they would be working on,

00:25:58   like they would notice that and work that out

00:26:00   in the next version of iOS.

00:26:02   - Yeah, these are just tweets talking about the slowness.

00:26:05   Like, slow is not a measure of anything.

00:26:07   It's just like, it feels slow to them.

00:26:09   And I think especially with single photo operations

00:26:12   or like doing something like keywording that you think,

00:26:14   I'm not asking you to run some crazy filter or blur

00:26:18   or there's no image effects,

00:26:20   there's not a massive amount of data.

00:26:21   It's like, please just associate a keyword

00:26:24   with these three photos, right?

00:26:26   These people are reporting that it is quote unquote slow

00:26:29   because they think that the thing that they're doing

00:26:31   shouldn't take as long as it takes.

00:26:32   And I agree with them, but we don't have any measures.

00:26:34   So you never know, like Will Haines says,

00:26:37   It is very slow on perf auto operations

00:26:39   on a maxed out iMac 5K.

00:26:41   Is that because he just expects it to be really fast

00:26:43   on an iMac 5K?

00:26:44   His operations may be five times as fast as mine,

00:26:47   but he has greater expectations

00:26:48   'cause he's got a brand new computer

00:26:50   and I'm willing to accept some slowness.

00:26:51   But bottom line is, perception wise,

00:26:53   I think it feels slow and some other people seem to agree

00:26:57   and I'm depressed that even people with small libraries

00:27:00   and even people with much newer, fancier computers than me

00:27:02   also are feeling that operations feel slow

00:27:05   that shouldn't be slow.

00:27:06   Maybe somebody didn't have enough empathy for the computer when designing it.

00:27:10   [laughter]

00:27:11   It didn't.

00:27:12   It didn't.

00:27:13   That's kind of a design thing.

00:27:14   It was like, "Why might it be slow?"

00:27:15   Well, I can tell you from experience that once you get into large number of items in

00:27:20   a SQLite database, SQLite's performance characteristics, especially for inserts into

00:27:27   a database or a table that has thousands or millions of records, you can just run that

00:27:31   experiment yourself.

00:27:32   SQLite database with a simple scheme of start inserting stuff and see and do a timer and

00:27:37   say how many rows can I insert per second and watch that performance slowly slowly get

00:27:42   worse.

00:27:43   So as you get into the millions of rows things get pretty grim.

00:27:46   There are things you can do to help with that but anyway if photos is using SQLite databases

00:27:51   anywhere and it's implementation which I can imagine it is and if that has anything to

00:27:54   do with keywords which I can imagine it might that might explain some of the slowness in

00:27:58   large libraries.

00:27:59   I'm just guessing.

00:28:00   And so the empathy with the computer is, you know, like, I didn't write SQLite.

00:28:05   And SQLite is a great database, but am I using the right tool for this job?

00:28:08   Would it be better to add them as extended attributes or as individual plist files or,

00:28:12   you know, I don't know.

00:28:13   That's more of a design thing, the empathy with the computer at the micro level.

00:28:16   All right.

00:28:18   That's macro level empathy.

00:28:20   Not Marco level empathy, macro.

00:28:22   All right.

00:28:23   What happens when you delete a photo, Jon, in the new Photos app?

00:28:27   I don't know.

00:28:28   That's the thing.

00:28:29   So last time I was talking about the prospects of using iCloud sync as a backup strategy.

00:28:35   It's like, well, if everything's all synced, if something ever goes wrong, that wrongness

00:28:39   could also sync all over the place and then your backups are hosed.

00:28:42   Because they're not really backups.

00:28:43   It's like RAID where you just have a second copy of whatever is currently in the thing.

00:28:47   So whatever is currently in the library is screwed up.

00:28:50   Whatever is currently in all your automatically synchronized copies can also be screwed up.

00:28:53   So many people pointed out that if you delete a photo in the Photos app, it doesn't get

00:28:57   deleted immediately goes into kind of like a trash type holding area and I

00:29:01   think it's kept there for like 30 days before it's purged or something which is

00:29:04   good but mostly what I was afraid of was the app going crazy and suddenly

00:29:08   thinking these photos never existed let me just get you know not as if it thinks

00:29:12   I deleted them by dragging them to the trash or hitting the delete button but

00:29:15   if it starts thinking because of some weird syncing thing that photos in my

00:29:19   library just never existed and says oh I need to I need to rectify I need to

00:29:23   synchronize and rectify my state of the world with what the state it's supposed

00:29:27   to be in.

00:29:28   Rather than it thinking that it saw a delete operation and synchronizing that, just thinking

00:29:32   that all these photos here, I don't know where all these came from, but I have no idea what

00:29:35   they are.

00:29:36   Like that's what I'm afraid of, not so much the manually deleting things or whatever,

00:29:39   but it is good to know there's a holding area where you can recover from mistakes where

00:29:42   you delete, because when you delete them it says are you sure you want to delete this

00:29:45   from the cloud and every single one of your devices it tries to tell you like this is

00:29:49   not just deleting locally, this is deleting everywhere.

00:29:51   And even if you do that, it still holds onto it in this deleted items area to give you

00:29:54   an out, which is nice.

00:29:56   I'm still a little bit afraid of it and I would never use it as a backup strategy.

00:30:00   And speaking of backups, god forbid something happened, what do you do to restore your photo

00:30:05   library from a backup?

00:30:06   Yeah, I don't know because I haven't tried that, but I thought about it like, restoring

00:30:10   like your iPhoto library from a backup, you know, it's just a bunch of files on disks.

00:30:13   If your thing gets hosed, just, you know, throw out your iPhoto library, go to your

00:30:18   time machine backup, your super duper backup, your backblaze backup or whatever, and just

00:30:21   like find a previous version of your iPhoto library, which appears to be one file on disk

00:30:26   but it's really just a pack, a bundle with a folder with a bunch of stuff in it,

00:30:28   and you drag it back in your backend business, right?

00:30:30   What happens if your photos library that you have iCloud synced,

00:30:35   like something gets wrong with it or it gets hosed or something, can you just say,

00:30:39   okay, well,

00:30:39   I'm just going to throw that library out and drag a new one from a backup and put

00:30:43   it in place and launch it? What happens? It would work fine locally.

00:30:46   I'm sure that would work locally, but what happens when you launch in an iCloud,

00:30:51   like, is it aware that you just did that? Or is it like, whoa, whoa,

00:30:54   this library and disk does not match what I expected to do,

00:30:57   does it forcibly make it like the old broken one was

00:31:01   or the one with the missing data?

00:31:02   How do you recover from a backup in that scenario?

00:31:05   And when I was thinking about this,

00:31:07   one more item, like you can tell me if you guys remember this

00:31:10   do you remember when iPhoto and Time Machine,

00:31:12   like the integration between them was first shown

00:31:14   that maybe when Time Machine was first announced

00:31:15   in a keynote and they tried to show you use Time Machine

00:31:17   from within iPhoto, am I just imagining that?

00:31:20   - I remember that being a thing

00:31:22   where you could use Time Machine specifically

00:31:24   within certain apps, but I don't remember anything

00:31:26   about iPhotos specifically.

00:31:28   - Yeah, whatever happened to that?

00:31:29   Like, using iPhotos specifically from within certain apps,

00:31:31   that's not a thing, really, is it?

00:31:33   - You mean Time Machine within certain apps?

00:31:35   - Yeah.

00:31:36   - There's probably some kind of API that nobody ever uses.

00:31:39   - I don't think there's a public API.

00:31:40   I think they showed it, like, you could do it

00:31:42   within photos, you could do it within address book.

00:31:44   There was a couple of apps they showed

00:31:45   that you could do it within.

00:31:46   And the reason I'm thinking of that is, like,

00:31:49   I'm trying to think, what would be a safe way

00:31:52   to take a photos library that you had iCloud synced

00:31:56   that is hosed in some way, it's damaged, it's corrupt,

00:31:58   you accidentally deleted stuff and you wanna say,

00:32:00   actually I wanna go back to the state things were

00:32:03   a week ago and you go pull your photos library

00:32:05   from your backup, which you have,

00:32:06   it's just a bunch of files on disk,

00:32:08   you put it back in the same place the other one was

00:32:10   and you launch photos and then what happens?

00:32:12   And I don't know what happens.

00:32:13   - I mean, I would imagine it's just like the import process.

00:32:16   When I imported my 25,000 photos,

00:32:18   I basically had an empty iCloud library

00:32:21   and I dragged them all into this app

00:32:23   from my previous arrangement of files,

00:32:25   and it slowly imported them and uploaded them all.

00:32:29   That I would imagine, as long as your cloud

00:32:33   photo library state can be cleared,

00:32:36   which I don't know if it's as easy as going

00:32:38   to the all photos view and hitting command A,

00:32:39   command delete, I have no idea, but.

00:32:41   - But do you do that before you pull the backup?

00:32:45   Like what if you can't even launch photos

00:32:47   because the library's so corrupted?

00:32:49   Like how do you get it to understand that you want it to launch and accept the on-disk

00:32:57   library that you pulled from backup as the current state of things?

00:33:00   Honestly, I have no idea, but the files are all there.

00:33:05   The library is laid out very similarly to an iPhoto library.

00:33:08   Yeah, you do have the out of just finding all the JPEGs, pulling them all out, making

00:33:12   a new empty library.

00:33:14   I think what you'd have to do, and I've done this before with address books, that's why

00:33:17   why I thought of it. A few times, AddressBook has gotten hosed in some way, and I've been

00:33:22   like, "Look, I have all these backed up," because AddressBook—I don't know if people

00:33:24   know this, but in the—not AddressBook, it's called ContactsNow, sorry—in the Contacts

00:33:28   application that Apple provides, they provide a way to make a complete backup of your entire

00:33:31   Contacts library, and please, everybody, do this at least once a year, because Contacts

00:33:36   Cloud Sync is not that reliable, and you'll be happy to have it. So I got into a state

00:33:39   where Contacts were hosed. It wasn't actually my Contacts, it was my wife's, so it was

00:33:43   more dire than you would imagine it might be.

00:33:46   And I needed to fix it.

00:33:48   And what I wanted to happen was,

00:33:50   I have a backup and I wanted to say,

00:33:53   look, iCloud, just forget everything.

00:33:56   Just start over, I'm going to import from a library

00:34:00   and I want you to believe that you have

00:34:01   zero contacts everywhere.

00:34:02   And that was super, super hard to do

00:34:04   because every time I thought I'd done it,

00:34:06   I'd tell it to import the library

00:34:07   and then it would do this crazy merge

00:34:09   with what I thought was still in the cloud

00:34:10   and hose everything again.

00:34:11   Took me so long to beat this thing into submission

00:34:13   to say you have nothing, start from zero.

00:34:15   Anyway, that's what I'm afraid of with the photos library.

00:34:17   Like I don't know what the restoration plan is.

00:34:21   Even in the scenario where you say,

00:34:22   look, I'm just gonna extract all the JPEGs

00:34:25   and then just have to convince photos that it has nothing.

00:34:27   Like just really, 'cause you can be fooled.

00:34:29   Like the contacts would be like, yep, no contact,

00:34:31   zero contact, zero contacts everywhere, everything's fine.

00:34:34   Okay, import.

00:34:35   And then it would do this crazy thing.

00:34:36   And I couldn't, you know.

00:34:37   - So real time follow up from _S_.

00:34:41   There's a, I don't think this is a KBase,

00:34:44   but anyway, there's a page on Apple's website.

00:34:45   How do I remove all iCloud photo library content

00:34:48   from my iCloud account and devices?

00:34:50   On OS X, go to System Prefs, iCloud Manage,

00:34:53   select Photo Library, then select Disable and Delete.

00:34:55   Once you turn off your iCloud photo library,

00:34:57   you'll have 30 days to download your library

00:34:59   to at least one device, and blah, blah, blah.

00:35:03   So that's how you do it.

00:35:04   - That's good that there's a button,

00:35:05   because that was always the problem with Apple's things.

00:35:07   It was like, oh, there's no button,

00:35:09   there's no place to go, it just works,

00:35:11   it's magical, but sometimes you just want,

00:35:13   like people always want, I want a button that says,

00:35:15   no, please synchronize now.

00:35:16   And this button is the, I want a button that says,

00:35:18   just forget everything, just start over from zero.

00:35:22   And my past experience was like back when they had

00:35:25   sync services and everything, that trying to get it

00:35:28   to start from zero was like basically not possible

00:35:31   for a regular person.

00:35:33   Sometimes you needed like a special application

00:35:35   that like Apple internal people would use

00:35:37   that like, because it wasn't just on your machine.

00:35:38   Like you could delete everything off your machine

00:35:40   You could find all the secret hidden files and everything, but there was stuff on servers

00:35:43   somewhere and you had to convince those servers to dump all their information and you had

00:35:46   to know like, you know, what requests to make to what servers to what authentication to

00:35:51   convince it to really, really reset things.

00:35:53   So I'm glad that Apple is learning and making this easier by putting in an actual GUI and

00:35:57   having a, you know, a knowledge base article about it.

00:36:00   Our second sponsor this week is Casper.

00:36:03   Casper is an online retailer of premium mattresses for a fraction of the price of most mattresses.

00:36:09   Now the mattress industry has inherently forced customers into paying notoriously high markups.

00:36:15   Casper is revolutionizing the mattress industry by cutting the cost of dealing with resellers

00:36:18   and showrooms and passing the savings directly on to you, the consumers.

00:36:24   Casper mattresses are very high quality.

00:36:26   They provide resilience and long lasting support and comfort.

00:36:30   It's a one of a kind combo of, it's a hybrid of premium latex foam and memory foam.

00:36:37   So it has all the benefits of both

00:36:40   without having the downside of like,

00:36:42   you know, if you have all memory foam,

00:36:44   then a lot of people don't like the way that feels

00:36:46   or it might feel too sweaty to them or whatever.

00:36:48   By combining memory foam with latex foam,

00:36:51   you really do get the benefits of both.

00:36:53   They call it just the right sink, just the right bounce.

00:36:57   And that's, you know, Casey, you have one of these.

00:36:59   I think that's a pretty good way to describe it, right?

00:37:01   - Yep, very much so.

00:37:02   I do not care for memory foam.

00:37:04   I think it feels weird and it's way too soft.

00:37:06   And in fact, I'm very, very, very anti-memory foam,

00:37:10   and the Casper mattress is wonderful.

00:37:12   It is, just like you said, just right.

00:37:15   It's like the Goldilocks of mattresses.

00:37:17   - Nice, they can run with that.

00:37:19   Now mattresses usually can cost, you know,

00:37:21   for a good mattress, something that's gonna last you a while

00:37:24   and it's gonna be really comfortable

00:37:25   and like decent and high-end,

00:37:27   you're looking at over like $1,500 for decent ones.

00:37:31   Casper mattresses cost, they start at just $500 for a twin,

00:37:36   all the way up to $8.50 for a queen and $9.50 for a king.

00:37:40   These are incredible prices.

00:37:41   I mean, I've now bought a couple of mattresses

00:37:43   so far in my life, and these are incredible prices

00:37:47   for mattresses this good, really.

00:37:49   So Casper understands that buying a mattress online

00:37:53   can have consumers wondering how this is all possible.

00:37:57   Now, to help you with these fears.

00:37:59   So first of all, buying a Casper mattress

00:38:01   is completely risk-free.

00:38:03   They offer free delivery and free returns

00:38:05   within a hundred day trial period.

00:38:07   So you can try it, you can sleep on it for 99 nights,

00:38:10   and then on the hundredth day decide,

00:38:12   you know what, I don't like it,

00:38:14   then they will arrange to have it freely picked up

00:38:17   and returned back to them.

00:38:18   It is really that simple.

00:38:20   I mean, you can't beat that.

00:38:22   If you just lie on some beds in a showroom

00:38:24   for a few minutes,

00:38:25   you're not really learning anything about them.

00:38:27   You're not really gonna learn

00:38:28   what it's like to sleep on them,

00:38:29   what it's like to live with them for a long time.

00:38:31   With Casper, you can try it for a hundred days.

00:38:33   It's totally risk-free.

00:38:35   Now this is obsessively engineered, shockingly fair price, and these are also made in America,

00:38:43   so this is all good news here.

00:38:45   And again, these prices are amazing, $500 for a twin up to $950 for a king.

00:38:50   Now you can also get $50 off by visiting casper.com/ATP and using coupon code ATP at checkout.

00:38:58   So really check out Casper Mattresses, they really do ship it to you in a little tiny

00:39:01   box and then it expands by sucking all the air out of the room into itself.

00:39:04   It's pretty cool.

00:39:05   You'll wonder how they can ship a king size mattress to you in this box, but it works

00:39:11   and it's pretty incredible.

00:39:13   So check it out.

00:39:15   Casper.com/ATP for 50 bucks off.

00:39:18   Good mattresses.

00:39:19   Thanks a lot, Casper, for sponsoring our show once again.

00:39:22   And speaking of sponsors, how are the shirts doing?

00:39:24   Oh, I haven't checked today, but they're doing well.

00:39:27   We have, at this point, yesterday we had already matched and slightly surpassed our goals,

00:39:33   our sales from last year, so thanks everybody.

00:39:36   - And we have now sold what, five or six white shirts?

00:39:39   - I think six, right?

00:39:40   - Six, double the number from last week.

00:39:42   - Woo!

00:39:43   - So thank you everybody, you're running out of time

00:39:44   if you wanna buy one of these shirts,

00:39:45   I think you have what, like six days left

00:39:47   or something like that?

00:39:47   - Yeah, but by the time you hear this podcast,

00:39:50   most likely we will not be talking about it

00:39:53   on the next podcast because the sale will be over.

00:39:54   So if you want a shirt, now is the time to go to the website

00:39:58   and get it, atp.fm/shirt.

00:40:01   - So let's talk about Apple Watch experiences.

00:40:03   starting with the grooves.

00:40:05   - I don't know if that's where we're gonna start.

00:40:07   Mark is not looking in the notes,

00:40:08   so he has no idea what we're talking about.

00:40:09   But I wanted, this item is here because,

00:40:13   you know, one of us has an Apple Watch

00:40:15   and we wanted to hear about it,

00:40:16   but I put a sub item in there

00:40:17   because someone tweeted at me that

00:40:19   after wearing Apple Watch for three days,

00:40:21   they have a groove starting to show up on their wrist,

00:40:24   right where the little heartbeat monitor thing presses in.

00:40:27   I think this person may have their watch

00:40:29   on a little bit tight.

00:40:30   - Oh goodness, I would say so.

00:40:33   I remember back when I wore watches that I would get sometimes marks from the band or

00:40:37   whatever.

00:40:38   I don't think this is an Apple Watch problem.

00:40:39   I think this is a watch band, possibly too tight problem.

00:40:43   But it did remind me that Marco had said that he wasn't sure he was all that excited about

00:40:48   the little dome on the bottom, like feeling that rubbing into his wrist.

00:40:51   So I guess we can start there because Marco actually does have a watch unlike me and unlike

00:40:56   Casey.

00:40:57   Yours hasn't come yet, right, Casey?

00:40:58   Mine is, I don't remember when the shipping update happened, but it is now due to arrive

00:41:05   anytime between the very, very, very end of May and the Thursday I'm in San Francisco.

00:41:10   Nice.

00:41:11   Which is the most first of first world problems.

00:41:13   Well, I've just taken off my watch, which I already have because I'm luckier than Casey

00:41:17   and woke up earlier.

00:41:18   I've just taken it off after 14 hours of wearing it so far today, and there's absolutely no

00:41:24   sign of any dimple on top of my wrist where the thing pushes in. There's no markings from

00:41:30   the band or anything. So yeah, I think Luca Soldani, however you say his name, sorry,

00:41:37   is probably just wearing it too tightly. Maybe you have the wrong band for you. Most likely

00:41:42   it's just set too tightly.

00:41:43   Wrist-shape problems, too. Look at that little bone bump on his wrist there. Possibly another

00:41:48   Italian like me with weird wrist-shape wrists.

00:41:51   It also appears that he's wearing it on his right hand.

00:41:53   I don't know if that makes any difference.

00:41:54   Probably not, because the dome is centered in the watch,

00:41:57   so probably is not the issue.

00:41:59   One thing I did find with the tightness of the bands,

00:42:03   I ordered it initially in order to get it on day one.

00:42:06   The only configuration I could get on day one

00:42:08   at the time that the store showed up for me

00:42:10   was the Milanese loop.

00:42:12   I was interested in it,

00:42:13   but I wasn't planning on ordering it on day one.

00:42:15   But anyway, I got that, and so I had that for a few days.

00:42:18   Since then, I actually met up with a nice guy in Connecticut

00:42:23   who wanted the Milanese loop and had the one I wanted,

00:42:26   which was the black leather classic buckle,

00:42:28   the one that just looks like

00:42:29   every other leather watch strap ever made.

00:42:31   I traded with him for that.

00:42:33   So I've now had a couple days with each of those.

00:42:35   So a couple days with Milanese,

00:42:36   couple days with classic buckle.

00:42:37   I like them both a lot, first of all.

00:42:41   They're both really nice bands.

00:42:42   The classic buckle is extremely lightweight

00:42:46   and probably the softest in the lineup

00:42:48   besides maybe the sport bands.

00:42:49   It is just a very comfortable band.

00:42:51   Milanese is a lot better than I thought it would be

00:42:55   when I ordered it.

00:42:56   It really is extremely nice.

00:42:58   It really kind of splits the difference

00:43:01   between formal and informal.

00:43:02   It is not too formal or too blingy or too flashy.

00:43:05   It really is extremely nice.

00:43:08   I was very happy with the Milanese.

00:43:09   It does not pinch your arm hair,

00:43:11   what a lot of people think that Link bracelets do.

00:43:15   The Milanese does not.

00:43:15   The only thing you gotta be careful of

00:43:17   when you're putting it on is the part that overlaps back on itself so it's like sandwiched

00:43:20   together, you can get arm hairs pinched in that when you put it on. So just be careful

00:43:25   of that and you're fine. The Milanese also I think is the coolest on you because like

00:43:32   the air can flow through it and it's not leather sticking to you or the fluoroelastomer sticking

00:43:37   to you the way that does. So I would say it's the coolest. The reason I didn't want to keep

00:43:41   it is because I was getting some skin irritation late in the day after wearing it for a long

00:43:47   time and some people mentioned this is an issue with metal for them a lot and I don't

00:43:51   think it's a nickel allergy based on some quick research but just some skin irritation

00:43:56   I was getting it would just itch a lot like crazy at the end of the day and that doesn't

00:44:01   happen with leather so I'm glad I switched but they're both really nice bands I would

00:44:06   have been very happy with the Milanese if not for that issue. So overall I can recommend

00:44:10   I will say though, I have had an issue with both of them in how tightly I wear it.

00:44:18   Because if you wear it super tight, I don't find that very comfortable.

00:44:23   If you wear it a little bit loose on yourself, not so it's like rolling around, but just

00:44:27   so it's more comfortable, it makes it harder to feel the taptic taps.

00:44:32   And I've had that issue a lot where I just miss the tap.

00:44:37   It does tap me and I just don't feel it.

00:44:39   And then later on I looked down and I see my little dot and I said, "Oh, I missed a

00:44:43   notification."

00:44:44   And I know that there's been an issue, the WSA reported earlier today, there's been an

00:44:47   issue with faulty Taptic engines that might be the reason why there was a short stock

00:44:55   in the watch launch.

00:44:57   I don't think mine is faulty.

00:44:58   It seems to work just fine.

00:45:00   It just isn't extremely strong.

00:45:03   It's like moderate strength.

00:45:05   And so to fix that problem, and I have it on the highest strength setting, it's still

00:45:08   not that strong. I have, there's a setting that that's that basically plays like a strong

00:45:14   vibration before any taptic alert. So like vibrate and then tap tap. I've enabled that

00:45:21   for most of the time I've had it and that has helped a lot. That makes me feel every

00:45:24   time but without that I will frequently miss them and I think that just because I'm wearing

00:45:28   it a little bit looser than than than I'm supposed to be because I don't like the feeling

00:45:32   of it digging into my wrist so strongly but it doesn't seem unreasonably loose. Again,

00:45:36   Again, it can't get misaligned, it can't roll back

00:45:39   on the sides of my wrist, it's not that loose.

00:45:41   So, I don't know.

00:45:43   - Have you done your heart rate or whatever?

00:45:46   'Cause that's the reason I think people want to have it

00:45:49   on tightly is they want accurate readings

00:45:51   from the biometric monitors.

00:45:53   And if you're not ever looking at that,

00:45:54   then you don't really care how loose it is

00:45:56   except for feeling the vibrations.

00:45:57   But have you ever looked at your heart rate

00:45:58   and seen if it's reasonable or crazy?

00:46:01   - I have not, but it is tracking my activity a lot

00:46:04   and it does register the exercise time.

00:46:08   So it has, in its built-in rings of activity,

00:46:11   it has the, it has, you know, calories,

00:46:13   which are related to steps, I think, and the standing,

00:46:16   and then it has the middle ring,

00:46:18   which is the one I always fail to get.

00:46:20   The middle ring is time spent exercising,

00:46:23   and that is like, you know, time at an elevated heartbeat.

00:46:26   And that seems to be measuring the actual real-life time

00:46:29   I'm spending at an elevated heartbeat.

00:46:31   It seems to be measuring that fairly accurately for me

00:46:33   throughout the day.

00:46:34   I don't actually use the workout app to measure

00:46:37   my heart rate as I work out because I don't work out.

00:46:40   So I'm not a good person to ask about that.

00:46:43   But it seems like I'm not wearing it so loosely

00:46:46   that that would be a problem.

00:46:47   - So the lump thing isn't bothering you,

00:46:49   but basically you are intentionally making it

00:46:51   a little bit looser so it doesn't bother you?

00:46:53   - I'm wearing it at the looseness level I'm wearing it at

00:46:57   not because of the lump thing,

00:46:58   but because of just the entire strap.

00:47:00   The lump thing is not the limiting factor

00:47:02   in the comfort here for me.

00:47:03   I thought it would be.

00:47:05   - It's just that you don't wanna feel the strap

00:47:07   like squeezing your wrist basically.

00:47:08   - Exactly, yeah, that's all it is.

00:47:10   And so I found with the classic buckle,

00:47:13   there are two holes that work for me.

00:47:15   Oh God, that's gonna be a terrible title.

00:47:16   (laughing)

00:47:18   Yeah, like one of them is like the really snug one

00:47:20   and one of them, oh God, I'll just stop talking.

00:47:22   - Come now, we are adults, we can talk about the watch.

00:47:25   - Okay, so anyway, yeah, so I found one that works for me.

00:47:29   Oh, the Milanese is also really nice

00:47:30   because you can make smaller adjustments

00:47:33   than any other band in the lineup.

00:47:35   So if you want to be like a little bit tighter,

00:47:38   a little bit looser, you can align it to any of those rows

00:47:42   of the woven links.

00:47:43   You can just like slide it over by one row.

00:47:45   So you can really--

00:47:47   - But it's like the toaster dials, you never know.

00:47:49   How do I get it exactly how it was last time?

00:47:51   Oh, I don't know, all the rows look the same.

00:47:54   - I assume before having it

00:47:56   that that would annoy the crap out of me.

00:47:57   And I did fidget with it throughout the day.

00:47:59   I would frequently change it,

00:48:01   like make a little micro adjustment throughout the day,

00:48:04   like to tweak it a little bit.

00:48:05   I didn't feel like that was annoying to me

00:48:07   as much as I would have thought.

00:48:10   I don't love about the classic buckle.

00:48:11   I don't love how, you know, just like any other watch,

00:48:14   you kind of need to like set it down to put it on.

00:48:17   You gotta like hold your hand against a table

00:48:19   to put it on, right, you know,

00:48:20   or like hold it against your leg or something like that.

00:48:22   Like you can't quite do it freehand very well in midair.

00:48:26   The Milanese does not have that problem.

00:48:27   I also don't like with the classic buckle that,

00:48:30   you know, just like any other watch strap.

00:48:31   So the little loop that holds the tail,

00:48:34   like the excess tail that holds it on the band,

00:48:38   I keep slipping out of that,

00:48:39   and it keeps catching on things.

00:48:41   Not crazy about that.

00:48:42   So I might end up giving the Leatherloop another try

00:48:45   at some point in the future,

00:48:46   but so far I'm very happy with this overall.

00:48:48   - How many times you taking it on and off a day?

00:48:51   - The Milanese, because I had that first,

00:48:53   I was a little more paranoid.

00:48:54   I was taking it off whenever I'd go wash the dishes

00:48:57   or anything like that.

00:49:01   that I've had the leather one for a couple days, I'm a little more confident in it, a

00:49:04   little more familiar with it, so now I'm just keeping it on all day. I took it off just

00:49:08   now just so I could look at my dent or lack thereof on my wrist from the follow-up. But

00:49:15   yeah, I'm just leaving it on all day now, so it's not that big of a deal.

00:49:19   So how is it affecting your life, if at all?

00:49:23   So again, I had not before this, I had not worn a watch since high school. I, like so

00:49:30   many people, especially among geeks like us, like so many people, I thought, "What do

00:49:36   I need a watch for? I have my phone in my pocket. That always tells the time." Turns

00:49:41   out, having a watch to tell time is actually really nice. I know the watch people are all

00:49:48   like, "Oh, God, we've been trying to tell you that for years," but yeah, that was

00:49:52   news to most of us geeks who have been brutally pragmatic and like, "Well, we have this

00:49:58   clock in our pockets all the time. Yeah, it turns out the move from pocket watches to

00:50:03   wristwatches a million years ago, there was good reason behind that. It's actually more

00:50:07   convenient.

00:50:08   I don't think I've ever looked at my phone to see what time it is, at least not during

00:50:12   the day. Because like the geek, I think the geek groove for people to tell the time is

00:50:20   you look slightly to the upper right of your field of vision because it's in the menu bar.

00:50:25   Right, right.

00:50:26   staring at all day as a screen with the times in the menu bar.

00:50:28   I cannot think of a time that I have looked at my phone

00:50:32   during work or during home.

00:50:34   Like, basically, any place I'm sitting in my house,

00:50:36   there is a clock in my sight lines intentionally,

00:50:40   so I know what time it is.

00:50:41   And when I'm sitting at my computer,

00:50:43   there is a clock in my sight lines all the time.

00:50:45   And none of those scenarios do I ever

00:50:46   reach for my iOS device, which is also probably in reach,

00:50:50   to check the time.

00:50:50   But many people on the go, like when you're outside,

00:50:53   you don't have a clock in your sight line.

00:50:54   They do use their phones.

00:50:55   I think Serenity Koval tweeted earlier that someone had asked her what time it was and

00:50:59   she fished out her phone to check the time while wearing the watch.

00:51:02   So old habits die hard.

00:51:03   Well, that's the thing is I was just about to say that even though I am more often than

00:51:09   not in front of either an iOS device or a computer, no matter what I'm doing, if I need

00:51:15   to look at the time, my first gut reaction is to look at my wrist.

00:51:20   And of course, there are times I'll look at my iOS device, I'll look at the upper right

00:51:24   hand corner of the screen but generally speaking or maybe the cable box like you were saying

00:51:28   John but generally speaking the first place I look to find the time is my wrist.

00:51:32   The cable box come on.

00:51:34   Yeah I forgot I forgot with whom I'm speaking but um but yeah so I stopped wearing a watch

00:51:40   for a few years I'd worn one pretty much all my life and then after after I got an iPhone

00:51:45   I was like you know what this is redundant and silly and I don't need to worry about

00:51:48   this and then I think like the battery my watch died or something like that and so it

00:51:52   wasn't worth fixing it, and so I just used my phone.

00:51:54   And then about a year or two ago,

00:51:55   I started wearing a cheap watch again,

00:51:57   and I'm glad because I like having the time on my wrist,

00:52:00   and I'm excited probably after WWDC

00:52:05   to have more than just the time on my wrist.

00:52:07   So we'll see what happens.

00:52:08   But I am feeling quite smug that Marco has decided

00:52:12   that this watch thing isn't so bad

00:52:13   because I do believe I tried to tell you, Marco,

00:52:15   that having something dangling on your wrist

00:52:17   really is not the end of the world.

00:52:18   - Yeah, I wonder how easy it is to rewire these things.

00:52:21   Like for the people who have the ingrained habits of going for their phone in their pockets

00:52:25   just the time, how long will it take them to retrain for like the look of the wrist?

00:52:29   And I'm also thinking like for the wrist move for Apple watch wearers, I imagine a lot of

00:52:35   them will become accustomed to a slightly exaggerated motion to really ensure that the

00:52:39   watch activates because, you know, sometimes the worst is if you bring it up and it doesn't

00:52:43   activate so like they may be training themselves into that habit.

00:52:46   And the reason I think about these habits and how easy they are to change is I know

00:52:50   Basically, since for half my life at this point, when I'm wearing my contact lenses,

00:52:55   I often take my finger and push my glasses up the bridge of my nose when there are no

00:53:00   glasses there.

00:53:01   Basically, I'm poking myself between my eyes with my finger.

00:53:05   And I've had contact lenses since, I guess, middle school, basically as young as I could

00:53:10   possibly get them.

00:53:11   So I am not new to the experience of not wearing glasses.

00:53:14   Nevertheless, every time I am wearing my contact lenses, I will at least once push my glasses

00:53:19   up the bridge of my nose that are not there.

00:53:22   - Yeah, that's understandable.

00:53:23   And when I went from being a not watch-wear

00:53:27   back to a watch-wear, it only took me a few days,

00:53:29   maybe a week to start looking at my wrist again

00:53:32   to find the time.

00:53:34   - Yeah, I mean, I thought this would be a longer adjustment

00:53:36   for me so far, and so far it really hasn't been

00:53:38   a long adjustment.

00:53:39   It was, yeah, a couple days that now I'm looking

00:53:41   at my wrist for the time and I'm enjoying it.

00:53:44   And the wrist detection thing where you try to detect

00:53:46   when you move your wrist up and turns the screen on.

00:53:49   I've had mixed success with that.

00:53:50   That still needs some work.

00:53:52   Overall, it works for me about three quarters of the time,

00:53:57   but that's like 25% failure rate is not good.

00:54:01   So that could still use some improvement.

00:54:03   And it certainly exacerbates the issue

00:54:07   that the screen has to turn off so quickly.

00:54:10   Oftentimes, I will try to do the motion to show the time,

00:54:15   and it will, it seemed like it's misinterpreting it

00:54:17   as first it thinks I flipped my wrist

00:54:19   and then it thinks I didn't,

00:54:21   and so the screen's on for a split second

00:54:23   then it turns back off again,

00:54:24   'cause it like cancels it.

00:54:25   - You're looking at it wrong, Marco.

00:54:27   - Yeah, exactly, yeah.

00:54:28   So I don't, that part, it could use some work,

00:54:32   but overall, I really do like it a lot.

00:54:35   I like the way it looks.

00:54:37   I like the way the watch face looks.

00:54:39   I've looked at other watches,

00:54:42   ever since Apple made a watch,

00:54:44   ever since I've all announced they were going to make a

00:54:45   watch, I've been looking around at what other people think are

00:54:49   really good watches and and every time I pass like a watch

00:54:52   store in the mall or something, I'll like I'll look

00:54:55   in the window and just kind of look at the watches there.

00:54:57   Look in the airport stores that are all crazy and

00:54:59   everything and when I look at other watches, I have never

00:55:04   really seen anything that I thought was really good

00:55:07   looking and yeah, I know there are some out there. I'm sure

00:55:10   everyone's going to then send me their the picture of the

00:55:12   watch that they think is good looking and say,

00:55:14   "Oh, you didn't see this one."

00:55:15   And yeah, I mean, you might be right,

00:55:16   but I never looked at any other watch

00:55:18   and said, "I really want that."

00:55:21   And this one, I look at this on my wrist,

00:55:24   the way I've set up, I use the utility face

00:55:28   with a fairly minimal set of complications on it.

00:55:32   And the way this looks to me is really nice.

00:55:35   Like, I really enjoy just the way the watch face looks.

00:55:38   Just as a timepiece, I enjoy it.

00:55:41   And the other functions it does are really great too,

00:55:44   but even just that part, I'm extremely satisfied with,

00:55:48   because I just like it.

00:55:50   For whatever reason, I just like it.

00:55:51   I will often look at the time

00:55:52   just to enjoy how the watch face looks.

00:55:56   I will say the complications,

00:55:58   which are the little features you can add

00:56:00   to show the temperature or your calendar events

00:56:02   or whatever in various areas of the faces,

00:56:04   I share a similar opinion as,

00:56:07   I believe Jason Snell who wrote about this,

00:56:09   and Casey used it as well.

00:56:11   The complications are a mixed bag.

00:56:13   I think many of them are very useful.

00:56:16   Most of them I would have some design tweak to.

00:56:19   In general, I'm like 90% satisfied

00:56:24   with the watch face availability that I have,

00:56:26   but that last 10% is kind of a killer.

00:56:28   If I prefer a digital watch face,

00:56:32   I think I'd be a little bit less satisfied

00:56:34   'cause the digital options are not that great.

00:56:36   One of my favorite concepts of the watch face

00:56:38   the solar one which shows the big like arc of the sun throughout the day because

00:56:42   I like knowing how much sunlight I have left so I know when like when the latest

00:56:45   I can walk my dog and be in the daylight basically and that watch face is

00:56:49   frustrating because there's no complication so if you want to also show

00:56:52   the temperature on screen or also show anything else on screen with that face

00:56:57   you can't do it there's no there's no customization of that face furthermore

00:57:00   if you're on that face and I think one or two others I think maybe the solar

00:57:04   system. Anyway, if you're on that face, if you hit, if you scroll the wheel slightly

00:57:10   on the digital crown while you're trying to hit it to go to the home screen, any slight

00:57:15   scroll is interpreted as you're moving the sun around in that watch face and it cancels

00:57:20   the button push. And so you're like stuck in the watch face. And that, I found that

00:57:25   very error prone when I was using that face where I kept accidentally scrolling slightly

00:57:30   when I touched the wheel and not being able to go immediately to the home screen that

00:57:34   I wanted. So that was frustrating. But for the most part, once I found what I wanted,

00:57:38   which was the utility phase, I'm very happy with it as a watch. It is really quite nice.

00:57:45   Again, like you, Casey, and like Jason, I would make some small changes to the complications.

00:57:51   Like the empty state for things like the timer, or when there's an event, the calendar thing,

00:57:59   when you have nothing for the day, it says, "No events," in these big letters. Why can't

00:58:03   just be blank you know stuff like that like the the blank states of these

00:58:06   things could use some help I have mine set with a with a timer at the bottom

00:58:10   so I frequently will start a timer for like brewing tea or cooking something

00:58:14   and the timer when you when nothing is set it has a little timer icon next to

00:58:18   it it has the word set which is dumb just have the timer icon why do I need

00:58:23   the word set I know that when it's empty you tap it and it brings the timer

00:58:26   that's great why you need that so I do like having these things on the watch

00:58:31   watch face. I do recommend if you're getting the watch, if you're setting it up, I do recommend

00:58:36   keeping it simple. And that could just be personal preference, but I find that I'm happier

00:58:44   the less stuff is on my watch face and the fewer glances I'm using. And the advantage

00:58:48   of that, you know, there's information density reasons, there's stress reasons, but just

00:58:52   like there's also battery reasons, I don't have all this stuff updating all the time.

00:58:55   And it's also easier to navigate when you only have a few things that you really care

00:58:59   but it's easier to see, it's easier to scroll,

00:59:01   there's fewer things to scroll through.

00:59:03   So I recommend keeping your watch set up very simple.

00:59:05   But people will figure that out for themselves, I think.

00:59:09   - All right.

00:59:10   How about repairability?

00:59:11   Because iFixit has had a tear down,

00:59:14   and it was, I felt like it was fairly surprising,

00:59:18   even knowing full well that this thing was not large

00:59:21   and was going to be jam-packed,

00:59:23   I was stunned by how much stuff

00:59:27   is crammed into that little tiny case,

00:59:29   especially since my recollection of trying on the watches is that they were not terribly

00:59:34   heavy. I mean the link bracelet weighed quite a bit, but the watches themselves were not

00:59:41   terribly heavy and gosh there's a lot of stuff in there. So do we think that the S1 is going

00:59:48   to be upgradeable in the future?

00:59:50   Well this teardown, I think you remember the iFixit teardowns is everything is zoomed way

00:59:55   in.

00:59:56   So no matter how large the device is, they fill the frame with the thing.

01:00:00   So it seems like, you know, because I finally, by the way, I finally saw an Apple Watch in

01:00:04   person actually on somebody instead of in a store.

01:00:08   Still haven't gone to a try on.

01:00:09   But anyway, they're much smaller than I thought they would be.

01:00:12   Like all of Apple's pictures, they looked huge.

01:00:14   Many people have bought the 42 and thought they've mistakenly got the smaller size, the

01:00:19   38 or whatever it is.

01:00:22   Because they're smaller than you think they are.

01:00:23   And so in these iFixit pictures, you're like, "Wow, look at all that stuff in there."

01:00:27   The key thing is not, "Look at all that stuff," the key thing is, "Look how small that stuff

01:00:30   is."

01:00:31   Because if they zoomed out to a reasonable length, you'd realize how microscopic those

01:00:34   things are.

01:00:35   Because they're zoomed way in to make the entire watch fill the frame.

01:00:38   So that's the thing that strikes me, that it seems like a lot of stuff is like, "Oh,

01:00:43   it's like a little TV dinner."

01:00:44   You have component A, component B, and component C, but those components are like the size

01:00:47   of pinheads.

01:00:48   They're microscopic.

01:00:50   So it's an amazing feat of miniaturization, but yeah, looking at the quote I pulled out

01:00:56   into the show notes here, looking at the S1 module, the little sort of, "Oh, the S1, if

01:01:00   they keep showing it as a little component, it could be replaceable."

01:01:02   Here's what iFixx had to say when they got the S1 module out of the watch.

01:01:06   "Pulling this mess out is a destructive procedure, but after ripping out some soldered connectors,

01:01:10   we got our first real look at the S1.

01:01:12   Despite rumors and hopes of an upgradable product, the difficulty of removing the S1

01:01:16   alone casts serious doubt in the idea of simply swapping out the internals."

01:01:19   iFixit is usually pretty good about like sort of doing surgery on Apple devices and getting

01:01:24   them apart and getting the pieces out.

01:01:25   And if they're saying basically to get the S1 out, you have to just rip off a bunch of

01:01:30   things like it is a destructive process.

01:01:32   I suppose you could just rip out all the guts, like just, you know, shuck it like an oyster

01:01:36   and pull everything out of it and just put entirely new guts inside it.

01:01:40   But even that, it looked like the type of thing that you would want to be done with

01:01:44   precision tools in a factory by people who, you know, people or machines that do the same

01:01:48   operations hundreds of times a day, rather than just having someone in the back room

01:01:53   scrape this stuff out. Because what would it even look like to reseat all these things

01:01:57   and reglue and seal all these things and reconnect all these connectors? Even if Apple wanted

01:02:03   to have internal swaps, it does not look good. I feel like if they offered that service,

01:02:08   I would be like, "Thanks, but no thanks." I really don't want the guts of this thing

01:02:12   pulled out and tried to be replaced because it would just never be the same again.

01:02:15   - Yeah, I mean, it's like if you get a laptop repaired

01:02:18   and it's not quite repaired right,

01:02:20   and it's like some seam doesn't quite line up right.

01:02:23   - What are you trying to say, Marco?

01:02:25   - Yeah, good luck with that.

01:02:26   But still like. (laughing)

01:02:27   - That was, the pre-unibody days, that was a real problem,

01:02:30   'cause I have opened up pre-unibody laptops

01:02:32   and they never go back together the right way.

01:02:34   But the unibody ones are more sturdy in that regard,

01:02:36   but this definitely looks like one of those things

01:02:38   where it's like, it's a one-way operation.

01:02:41   Assemble, seal, it's your thing.

01:02:44   Maybe they could do the battery,

01:02:45   that's kind of on the outside but I even get worried about like because you see

01:02:48   how they take the screen part off like it's not there's no visible screws that

01:02:52   hold the screen on it just I guess it's just glued and they just kind of like

01:02:55   heat up the glue and pry the thing off and you're like really I mean it's the

01:02:59   same way that the windshield is stuck onto your car just you know much much

01:03:04   much smaller I suppose so I guess it's I don't know I just feel nervous about

01:03:08   things that are connected with adhesive I trust the factory sort of the factory

01:03:13   seal of putting, you know, even for windshields, like the windshield was put on my car in a

01:03:18   factory under controlled conditions, under hopefully ideal conditions, and yes, if you

01:03:23   get a crack in your windshield you can hire someone to come and take it off and put it

01:03:27   on a new one and they'll do it, you know, while you're at work inside the parking garage.

01:03:30   That's not ideal conditions.

01:03:32   Are they heating up the adhesive to the right temperature?

01:03:34   Is everything all, it just bothers me as an anal-redentive person.

01:03:38   But anyway, looking at this iFixit thing I'm thinking this was not designed to have the

01:03:43   the internal swapped and even if it could be done I would not want it done to my watch.

01:03:47   Yeah, I mean if you look at like basically all of the components are on top of the S1.

01:03:52   Like the S1 is at the bottom and in order to take it out you have to basically take

01:03:58   everything out of the watch. It's like a Porsche Boxster. Okay. You know, you got to

01:04:04   pull the whole engine out of the car. That isn't true is it? No it is not but I'm

01:04:08   waiting for the Porsche people to come and yell at me but no for mid-engine cars it is

01:04:12   just much harder to get at the parts of the engineering and to get out to the labor costs

01:04:16   for doing stuff like that as much as I would not based on based on how the watch is laid

01:04:22   out internally and all these different parts you have to disconnect and move and go around

01:04:28   in order to pull the S1 out. I'm with you John. I think I think this is clearly designed

01:04:32   to be a one way operation. I don't think anything is going to be offered for upgrading these

01:04:38   to the S2 or whatever. Even for the editions, I don't see it happening. And I think Apple's

01:04:45   solution is going to be just like their solution for everything else they make, which is, this

01:04:49   is today's model, and then eventually we're going to have a new model. And if you want,

01:04:53   you can recycle your old one.

01:04:54   Yeah, but we were talking about the chances of Apple offering upgradeability, specifically

01:04:58   with respect to the edition, because it's so darn expensive, and who's going to spend

01:05:01   all that money only to spend all that money again in three years, right? So that would

01:05:05   be the highest chance of having upgradeability. We talked about all the reasons they would

01:05:08   because it's just not the Apple way,

01:05:10   and if you can afford one 10 to $20,000 watch,

01:05:14   you can certainly afford two, right?

01:05:16   You know what I mean?

01:05:17   Like all these reasons,

01:05:17   one of the reasons we didn't discuss

01:05:19   was that these things are gonna be so darn small,

01:05:21   and Apple's gonna wanna wedge everything

01:05:23   that it possibly can.

01:05:25   And if it came down to a decision between,

01:05:28   we could make it a little bit smaller,

01:05:30   but it would make it much, much harder

01:05:33   or impossible to do a component swap.

01:05:35   Apple would always say, "Make it a little bit smaller."

01:05:37   like they're never optimizing for like,

01:05:40   just look at, go through the iFixit guide.

01:05:42   So the past five or 10 years or however long iFixit's

01:05:44   been around, probably not 10 years,

01:05:46   look at how much sort of empty space and extra stuff

01:05:50   that used to be inside Apple's laptops, for instance,

01:05:52   and compare it to the MacBook One tear down,

01:05:54   where I don't know if they have one of those yet,

01:05:55   I think they do, but they do.

01:05:57   There's like nothing in there.

01:05:58   There is no, like the number of pieces,

01:06:01   like they just want to get everything else out,

01:06:02   any kind of sort of framing mechanism or like cage or guard

01:06:07   or whatever it says, they just want to get everything

01:06:09   out of there and just squish it down till all there is

01:06:11   is like aluminum battery motherboard.

01:06:12   Like every, anything that is not essential has been removed

01:06:15   including key travel on the keyboard.

01:06:17   So that attitude of like, if it comes down to it

01:06:21   can we wedge more stuff in?

01:06:22   Yes, by sacrificing the ease of repair, totally do it.

01:06:26   And so that is, that's kind of what we were talking about

01:06:29   with like Apple's not going to do this

01:06:30   but this is another aspect of why this is not the Apple way.

01:06:33   Like I said, they still could on the addition say

01:06:35   give us your edition, we'll just yank out all the insides and put in all new guts.

01:06:38   They could do that like in the same way that, you know,

01:06:40   Casey's MacBook Air could be sent away to what do they call the Depot repair

01:06:44   center or like the magical fairy land where the Grinch takes your Christmas tree

01:06:47   to repair it. Like you can send your computer there too.

01:06:49   And then like something will happen. It always feels better to like, Oh,

01:06:54   we're going to send it out for repair.

01:06:55   And you can always imagine that that repair facility is much nicer and cleaner

01:06:59   and staffed by precision nano machines and robots and like, but it's not,

01:07:03   It's just different people in a different room.

01:07:06   Anyway, trying to think about it.

01:07:07   (laughing)

01:07:09   I still say Apple could, within the realm of reason,

01:07:13   say, by all means, bring us back your $17,000 watch.

01:07:17   We'll take the insides out and put in the insides

01:07:19   of the Apple Watch 2, which happens to have

01:07:20   the same external case.

01:07:21   They could, they probably won't.

01:07:23   - Well, but even Apple, when they do these depot repairs

01:07:26   and everything, a lot of the times,

01:07:28   the stuff that iFixit will do to try to remove a battery

01:07:31   that's glued to the case, for instance,

01:07:33   Apple doesn't do that.

01:07:34   They just replace the whole, that whole case section.

01:07:38   Like they don't, Apple does not take extreme measures

01:07:42   in their repair departments to disassemble things like that

01:07:46   that are just like super glued in there

01:07:48   and you're never gonna get it out.

01:07:49   They don't do that either.

01:07:50   They're just replacing more of it and charging you for it.

01:07:52   - Well, especially batteries, because if you bend it

01:07:54   or poke it or like puncture it,

01:07:56   or like the batteries become big flaming things very quickly.

01:07:59   So batteries they're very careful with.

01:08:02   But for other things, they do have,

01:08:04   I seem to recall someone telling me,

01:08:06   for taking the iMac screens off

01:08:09   and not getting dust behind them,

01:08:11   they have rigs to take the screen off easily and repeatedly

01:08:14   and blow out all the dust and do all these other things.

01:08:18   Things that may be beyond what a regular person can do,

01:08:23   the repair depots or the back of the Apple Store

01:08:25   may be better equipped to do it

01:08:26   because they have special purpose tools just for this thing.

01:08:28   But yeah, it reaches a limit where they say,

01:08:30   Well, it's actually we're since we're the repair depot. We don't need to carefully remove item a from item B

01:08:35   We just take the entire like top case thing and we get you a new top case or a bottom case or whatever the heck

01:08:40   You know what? I mean, and that's that is much more straightforward

01:08:42   That's why I think for the addition watches that they were gonna do it

01:08:44   They would take everything out of the edition except for the gold empty case and take entirely new insides with entirely new screen

01:08:50   Entirely new sapphire crystal everything is gonna slap it in there. It's just like one sealed edge

01:08:55   edge. I think that is still possible given the iFixit teardown, but not likely for all

01:09:00   the other reasons we talked about in the past about Apple just wants you to buy a new one,

01:09:04   so you will.

01:09:05   Right, I mean, I think at best we would see something that is basically a recycling program,

01:09:11   maybe with a better name, maybe with better rates for maybe just the editions. So maybe

01:09:15   they'd have some kind of trade-in thing, but I don't think you're going to get back the

01:09:19   same watch you trade in with new stuff inside. I think you might get back a brand new watch

01:09:24   and you might get some token rebate for the old one

01:09:26   that you traded in.

01:09:28   And even that, if they never did that,

01:09:30   I wouldn't be surprised, and it wouldn't be a big deal.

01:09:32   But if they're gonna do anything like that

01:09:34   for the addition customers,

01:09:35   I think that's the form it would take.

01:09:36   It wouldn't be you get back the same case

01:09:39   with new stuff inside.

01:09:41   - All right.

01:09:42   - Our finalist bunch of this week is Harry's.

01:09:45   Go to harrys.com and enter promo code ATP

01:09:49   to save $5 off your first purchase.

01:09:52   Harry's offers high quality razors and blades for a fraction of the price of the big razors

01:09:57   brands.

01:09:58   It was started by two guys who wanted a better product without paying an arm and a leg.

01:10:02   Harry's makes their own blades from their own factory, which is actually an old blade

01:10:06   factory in Germany that they liked so much that they actually just bought the whole thing.

01:10:10   These are high quality, high performing German blades crafted by shaving experts.

01:10:15   They give you a better shave that respects your face and your wallet.

01:10:20   Harry's offers factory direct pricing at a fraction of the big brand price.

01:10:24   They're usually about half the price.

01:10:26   Plus you don't have to wait around for some employee to come unlock the anti shoplifting

01:10:31   case at drugstores to buy them.

01:10:33   So none of that stuff.

01:10:35   You order it online, they ship it directly to your door and it's free shipping all the

01:10:39   time.

01:10:40   Free shipping directly to your door.

01:10:41   Now Harry's, I've tried this stuff.

01:10:43   It is really quite good.

01:10:44   I'm very impressed at what they manage for the price.

01:10:47   I would say, I don't think they have yet something that matches the Pro Glide strip on the Fusion,

01:10:57   but with the exception of that, I would say their blades are pretty much identical to

01:11:01   the quality of Fusion blades.

01:11:03   And I have very sensitive skin, so I would probably be able to tell you a difference

01:11:07   if there was one, and I really could not tell any difference at all with the quality of

01:11:11   the shave or the comfort.

01:11:13   It really is quite good.

01:11:15   they run about half the price of the Fusion Blade, so it's a no brainer. And their handles

01:11:19   are really, really nice, their designs are really nice, it's almost like the new Mad

01:11:24   Men kind of style of classic and classy looking things that kind of are inspired by old designs

01:11:32   but still look new and still look modern. Really great stuff, you know, nice heavy handles,

01:11:38   tasteful designs, it doesn't look like the weird Android commercial robot ultra-masky

01:11:45   transformers looking designs that the other razor things all look like. It's

01:11:50   just tastefully designed stuff. Great stuff from Harry's. Now you can get a

01:11:53   starter set from them. Just try it out. It's an amazing deal. You get 15 bucks,

01:11:57   you get a razor moisturizing shave cream or gel, your choice, and three razor

01:12:02   blade cartridges. When you need more blades they're just two dollars each or

01:12:06   less. An eight pack is 15 bucks, a 16 pack is 25 bucks. For instance, for comparison

01:12:14   here. Fusions are 12 for about 40 bucks. So 12 Harry's Blades are 20 bucks. So really

01:12:21   it is about half the price. You get the convenience and ease of ordering online, high quality

01:12:25   blades, great handle, great shaving cream, and excellent customer service if you ever

01:12:30   need it. All this at half the price of the big brands. So get started today with that

01:12:34   starter kit. Handle three blades and shaving cream for just 15 bucks shipped right to your

01:12:39   door and if you use promo code ATP at Harry's.com use promo code ATP to get

01:12:44   five dollars off your first purchase. So thanks a lot to Harry's. Go to Harry's.com

01:12:49   promo code ATP for five bucks off. Thanks a lot. So Microsoft had their first day

01:12:56   of, or I think it was the first day of their build conference, which is kind of

01:13:00   the Microsoft WWDC. Coincidentally it takes place in Moscone West. There were a

01:13:05   couple of big announcements I was interested in. The first one was Visual Studio Code.

01:13:13   Now despite the fact that this has Visual Studio in the name, in a lot of ways it is

01:13:18   not the traditional Visual Studio. But what's interesting about it is, one way or another,

01:13:22   regardless of the name, it is cross-platform. It will run on OS X, it will run on Linux,

01:13:27   and it will run on Windows. And Visual Studio Code, at a glance, looks like Adam, the editor

01:13:34   that we talked about quite a long time ago that was written by GitHub.

01:13:39   It looks like, um, it looks like Atom, but with a little bit of Microsoft flair on it.

01:13:46   And that's not necessarily a bad thing.

01:13:49   In fact, one of the really appealing things to it is it has really good IntelliSense.

01:13:53   And in case you don't know what IntelliSense is, it's the code completion and the little

01:13:58   tool tip that pops up that helps you complete code.

01:14:02   And as someone who's worked on Xcode a fair bit and makes his living working in Visual

01:14:07   Studio, I can tell you that IntelliSense is light years better than anything Xcode has

01:14:14   to offer, and I haven't used Eclipse for years and years and years, thank goodness, but I

01:14:19   don't remember Eclipse being any better.

01:14:20   Well anyway, so at a glance it looks just like Atom.

01:14:23   Well, come to find out, it actually is Atom in some way, shape, or form, and I haven't

01:14:29   really dug into the mechanism by which all this is held together, but there is at least

01:14:35   a part of Atom included in the package that is Visual Studio Code.

01:14:40   I did try Visual Studio Code.

01:14:43   I tried it shortly after it came out, and one of the things it promised is really, really

01:14:48   good Node support, which, as you guys well know, I've been really interested in lately,

01:14:53   and that's what my blog is running on, is Node.

01:14:55   And I went to download it, I downloaded it no problem, I ran it no problem, I opened

01:15:02   up my project, my camel project, and one of the first things I did was try to run.

01:15:12   And I tried to run it and I got a message saying, "Well, you can't because you don't

01:15:18   have Mono installed."

01:15:19   >> Okay, well in case you're not familiar,

01:15:22   Mono is the cross platform C# compiler that will compile to OS 10,

01:15:28   it'll compile to Linux, etc, etc.

01:15:30   And all the libraries associated with it.

01:15:33   So it's crossplatform.net.

01:15:35   >> Does it just make you tired for a month?

01:15:36   >> And it also makes you tired for a month,

01:15:38   and you get it by kissing people.

01:15:39   >> Right.

01:15:40   >> So it says, "Oh,

01:15:42   well, you don't have Mono installed."

01:15:44   So, okay, fine. So I brew install Mono,

01:15:47   And I forget what version it is that I got, but I tried to run again, or debug again,

01:15:54   and it said, "No, no, no, really, you don't have Mono installed."

01:15:57   So I was tweeting in general, and also tweeting at code, which is the Twitter handle that Microsoft somehow got.

01:16:05   I don't know how they landed that, but I tweeted at code and said, you know, I was saying, "Oh, this is weird. It's not working right. It's not working right."

01:16:11   Well, within an hour or two, the guy that was in the introductory video, I think it's

01:16:16   Chris Dias, I probably pronounced that wrong and I'm sorry, but the guy who was in the

01:16:21   intro video was tweeting at me saying, "Hey, hey, have you tried this? Have you tried that?"

01:16:25   What? Like, can you imagine walking out of an Apple WWDC presentation, hell, the keynote,

01:16:34   and trying something, and then Federici is tweeting at you saying, "Oh, did you try this

01:16:38   this thing or the other thing.

01:16:39   What?

01:16:40   That's amazing.

01:16:42   - You know what that means?

01:16:43   That means Microsoft is hungry.

01:16:44   - It does.

01:16:45   No, you're absolutely right.

01:16:46   - Microsoft is showing hustle.

01:16:48   Take a lesson, Apple.

01:16:49   - Oh, it's so true.

01:16:50   So anyway, so come to find out that Chris

01:16:53   had a couple of really good pointers,

01:16:54   and it turns out that I had two copies of Mono

01:16:58   on my computer.

01:16:59   I'm not sure when or how I installed the first one,

01:17:02   but-- - Package management.

01:17:04   (laughing)

01:17:05   - One way or another, I had.

01:17:07   And so I uninstalled the brew version,

01:17:10   or the homebrew version, I uninstalled the other one,

01:17:12   reinstalled the homebrew version,

01:17:14   and sure enough, debugging worked immediately.

01:17:16   I had put a post on my website a while back

01:17:18   about Node Debug, which is a kind of sort of hacky,

01:17:23   but still really nice debugger that works

01:17:25   with the Chrome web inspection tools.

01:17:28   I had tried that in the past and I'd really liked it.

01:17:30   This is so much better though, because this is,

01:17:33   Even though it is in no way actually Visual Studio,

01:17:37   except in name, it felt kind of similar to Visual Studio.

01:17:42   Like I hit F5, which by the way,

01:17:44   I have mapped to the keyboard brightness.

01:17:46   I don't have it mapped to be X, or excuse me, F5.

01:17:49   And sure enough, it figured it out and said, okay,

01:17:52   I'm gonna go debug.

01:17:53   And it worked really well.

01:17:55   And I've only played with it for a few minutes,

01:17:57   but I've been really, really impressed.

01:17:59   It has Git integration right there.

01:18:01   It has Markdown Preview and somebody tweeted at Gruber earlier today, I want to say it

01:18:06   was Rich Segal, but somebody tweeted at Gruber saying, "Hey, did you ever think 10 years

01:18:10   ago that Microsoft would ship a free editor, a free kind of sort of IDE for the Mac that

01:18:16   does Markdown Preview?"

01:18:17   And Gruber was like, "Of course not, no way."

01:18:19   And it's true.

01:18:20   I mean, it's really impressive.

01:18:22   It's a really impressive app.

01:18:24   A little bit of Ajita getting it going, but in the grand scheme of things, I really like

01:18:28   it.

01:18:29   I know everyone loves their own text editor.

01:18:31   I'm not saying it's better than any other.

01:18:33   This is not a vinyl is better kind of situation.

01:18:36   I'm just saying it's worth checking out.

01:18:38   And Marco, I tweeted at you earlier tonight,

01:18:40   although I don't know if you saw it,

01:18:42   that they even have PHP syntax highlighting and all that,

01:18:45   which may or may not be very exciting.

01:18:47   But I was impressed that it was that supportive

01:18:51   of all flavors of programming languages

01:18:54   rather than strictly speaking the Microsoft stuff.

01:18:58   - Yeah, I mean, this is clearly

01:19:00   part of the new Microsoft and realizing like you know the fights that they that

01:19:04   they can and should be fighting the fights that they have already lost and

01:19:08   and where they can where they can still have great value where they can hustle

01:19:12   and and this seems like a pretty good move overall I mean my main question

01:19:17   here is will this will this stay a thing you know will this be followed through

01:19:22   on and will it stay a thing because that's where Microsoft has recently not

01:19:27   done that well and and not even that recently like you know in the last 10

01:19:30   years has had a hard time like following through on stuff and well they thought

01:19:35   they'd follow through of the things they did succeeded in gain traction you know

01:19:39   I mean that's been the problem I think it's like if one of those things caught

01:19:42   fire you sure bet they'd fall through they would love to follow through but

01:19:45   they're being kind of ruthless and like if it doesn't catch on okay let's try

01:19:50   something else like all these things are not moves that are done from someone in

01:19:54   a position of strength it's kind of like Apple they're not as bad as Apple was in

01:19:57   in the 90s, obviously, but when you're in a corner

01:20:02   and when you're behind, you can't afford

01:20:05   to do the old Microsoft way of like,

01:20:07   not invented here syndrome,

01:20:08   we're gonna do everything ourselves.

01:20:09   So why are they using Electron,

01:20:11   which is the framework that GitHub spun out

01:20:14   of their Atom project to say Atom is built on Electron.

01:20:17   It's all, anyway, we'll put a link to it in the show notes.

01:20:19   Why are we using this web-based framework

01:20:21   that uses Google's web browser engine under the covers

01:20:24   to make quote unquote native applications

01:20:27   using a web rendering engine technology,

01:20:30   like why don't we write our own editor?

01:20:31   We have, don't we have our own editor?

01:20:32   We wrote Visual Studio, like, 'cause you can't,

01:20:34   you don't have time, you need,

01:20:35   if you wanna do this thing, we have an idea,

01:20:37   we have the strategy, we're gonna talk about it in a bit,

01:20:39   that's gonna let us do all this cross-platform stuff,

01:20:41   and we wanna have an editor,

01:20:42   we wanna let people develop on the Mac,

01:20:44   can we make a new Mac editor from scratch?

01:20:46   No, we don't have time, like, how fast can we get to market?

01:20:50   And the old Microsoft was too proud,

01:20:53   or too stubborn to do what they're doing now,

01:20:57   which is like, well, we can get to market pretty quickly

01:20:59   if we use this open source engine

01:21:00   that GitHub used to make it editor,

01:21:02   that gives us a huge headstart.

01:21:04   And maybe it's not as nice as the real Visual Studio editor,

01:21:08   but it's not like we're gonna port that to the Mac.

01:21:09   And even if we did, it would be kind of cruddy.

01:21:11   So let's do what it takes to get off the ground.

01:21:13   And just like Apple was like,

01:21:15   can we make a modern operating system?

01:21:17   We tried like three times and it didn't work.

01:21:19   Can we just buy somebody?

01:21:20   And that's what they did.

01:21:21   Can we use the Windows NT kernel?

01:21:23   Can we buy B?

01:21:23   Can we buy next?

01:21:24   They made the right move there.

01:21:25   but this is not something that you do

01:21:30   when you're in the driver's seat of anything.

01:21:32   This is something you do when you're behind

01:21:34   and it's smart and it's expedient,

01:21:35   but it's also, I'm not gonna say it smells of panic,

01:21:39   but it's kind of sad to see.

01:21:44   Like one of the things that made Microsoft great

01:21:45   was like they thought we can do everything ourselves

01:21:47   and they really could do everything themselves.

01:21:50   Usually a reasonably good job of it.

01:21:53   The new Microsoft doesn't have time to do everything itself,

01:21:56   which is probably a good thing,

01:21:58   but they didn't make Atom, GitHub made Atom.

01:22:03   They didn't make Electrum, GitHub did.

01:22:04   They didn't make the Chrome engine, Google did,

01:22:07   and Apple made WebKit, and the value they're adding,

01:22:11   you're right Marco, they're trying to add value

01:22:13   where they can, but they're standing on the shoulders

01:22:15   of all the other companies that used to,

01:22:18   they used to be the giants,

01:22:19   and now they're standing on the shoulders

01:22:20   of all these other companies,

01:22:21   some of which didn't even exist

01:22:22   when Microsoft ruled the roost.

01:22:24   - Yeah, it's very true.

01:22:26   But yeah, this Visual Studio Code thing,

01:22:28   I really dig it in the very, very brief

01:22:30   amount of time I've spent with it.

01:22:32   And it is far and away the best node debugger I've tried,

01:22:35   but I'm a very amateur node programmer,

01:22:38   so there may be some other package or product out there

01:22:40   I'm just not aware of, but--

01:22:42   - Is there a comma in that name?

01:22:44   - No comma debugger.

01:22:45   (laughing)

01:22:48   - I don't even know where to go from here.

01:22:51   So that was Visual Studio Code.

01:22:53   You should definitely try it out.

01:22:54   It's free download.

01:22:55   It's worth at least checking out.

01:22:57   The other really interesting thing that they talked about, which unfortunately we haven't

01:23:02   heard a lot of technical details about yet, is two different projects.

01:23:06   One of them is called Islandwood and the other Astoria.

01:23:09   Let me cover those in reverse.

01:23:11   Project Astoria is to allow Android apps to run on, I believe it was Windows Phone only.

01:23:20   And this is vaguely similar to what BlackBerry did way back when.

01:23:24   And basically any APK, so any Android app that's been built, that doesn't use any of

01:23:33   the proprietary Google stuff.

01:23:34   So my understanding of Android is that there's the Android open source project, and then

01:23:41   as a kind of, is it superset or subset of that?

01:23:44   I don't know.

01:23:45   addition to that is the Google Mobile Services, which has

01:23:50   things like Google Maps and things of that nature.

01:23:53   If your app strictly stays within the APIs that are part

01:23:57   of the open source Android, then you can take your APK,

01:24:01   drop it in Windows, and it will run

01:24:04   supposedly no problem.

01:24:06   The thing that I haven't been made clear, though, is how is

01:24:11   this working from a technical perspective?

01:24:13   And I don't know if this is like a wine kind of situation.

01:24:15   What was a background name for wine?

01:24:18   Like something is not Windows or something like that.

01:24:20   I don't know.

01:24:21   But anyway, the wine is the app that allowed you

01:24:26   to run Win32 apps on Linux.

01:24:29   And did that ever work on OS X?

01:24:31   Was that a thing?

01:24:32   - Wine is not an emulator?

01:24:33   That's gonna be my guess.

01:24:34   - That's what I was thinking of.

01:24:35   Thank you.

01:24:37   That's what I was thinking of.

01:24:38   Anyway, so I don't know if this is like a wine sort of thing

01:24:41   or not, I'm not sure what the mechanism is

01:24:43   for this working under the hood,

01:24:44   but supposedly you'll be able to run certain APKs in Windows

01:24:49   without any modification whatsoever.

01:24:53   So any thoughts on that before I move on to Islandwood?

01:24:55   - Again, this is another one of those days

01:24:57   where an announcement is made on the day that we record

01:25:00   so none of us have really had time to read about this stuff

01:25:02   or watch the thing.

01:25:03   But my understanding from watching tweets from people

01:25:05   I think know what they're talking about

01:25:07   is that the iOS compatibility

01:25:11   is you take your source code and you compile it.

01:25:14   - Oh, well, we're not there yet, man.

01:25:15   We're only talking about Android.

01:25:17   - I know, but I'm comparing and contrasting.

01:25:19   You take your source code and you compile it,

01:25:21   and the Android one is you take your Android app

01:25:24   and you run it.

01:25:26   And that's different.

01:25:26   Like, even if you have the source to your Android app,

01:25:28   I don't think that's how it works.

01:25:29   I don't think they say,

01:25:30   "Oh, take the source to your Android app and compile it."

01:25:32   So that's a big difference

01:25:34   in how they're gonna implement things.

01:25:36   If you're not recompiling it,

01:25:37   that means binary compatibility,

01:25:39   But of course, the Android apps are written on the Dalvik VM,

01:25:42   or written against the Dalvik VM, which is their don't call

01:25:45   it Java, Java VM.

01:25:47   So it's bytecode involved, so why would they

01:25:49   need to recompile it?

01:25:50   If they've got bytecode, they can run it.

01:25:52   The JVM spec and probably the Dalvik spec

01:25:54   are something that they can reverse engineer or read

01:25:56   documentation for.

01:25:58   So it's much more straightforward.

01:26:00   And it's not like Wine, because it's not

01:26:01   native code to begin with.

01:26:03   It's the Java bytecode stuff.

01:26:07   Interesting point.

01:26:08   I see what you're saying.

01:26:08   So that's, I think that's how they're getting away with,

01:26:11   you don't have to recompile, it's not source compatibility.

01:26:13   And it's not like a wine where they're like stubbing out

01:26:15   a bunch of things.

01:26:16   It's like, they're just implementing a Java virtual machine

01:26:20   and the APIs that it links against, I guess.

01:26:23   But you don't have to, you just take your binary and run it.

01:26:25   This is again, from not having watched the presentation

01:26:27   and just seeing people's tweets, this is my understanding.

01:26:30   And yes, if I let the cat out of the bag

01:26:31   about the iOS thing, which everybody knows by now,

01:26:33   they're doing that entirely differently,

01:26:35   that you have to have your source code

01:26:37   and then you compile it against a bunch of their libraries,

01:26:39   which are not Apple's libraries,

01:26:41   but hopefully have all the same APIs and are compatible

01:26:43   and good luck with that.

01:26:45   - Right, so that's the Islandwood thing.

01:26:47   And just like John said, you can take an Xcode project

01:26:51   and suck it into Visual Studio

01:26:52   and then recompile it against a quote middleware layer quote

01:26:57   and apparently magic happens

01:27:00   and then you've got a Windows app.

01:27:02   And not only is it a Windows app,

01:27:06   But it's a universal Windows app, so presumably it can run on the desktop, it can run on a

01:27:11   tablet, it can run on the phone, it can run just about anywhere that's Windows.

01:27:16   What I'm really interested to see is what exactly is doing all of this.

01:27:21   Oh, and it's worth pointing out this is Objective-C only as well.

01:27:25   What is this middleware layer?

01:27:26   How is this held together?

01:27:27   What is really going on here?

01:27:29   But I did read in Ars Technica, and we'll link this in the show notes, or I think I've

01:27:33   read it on Ars Technica, but somewhere I have read that not only does this island would

01:27:40   support just general kind of sort of cross compilation, but it even supports UIKit and

01:27:46   even more importantly than that, core animation.

01:27:50   And if you've never written iOS code before, core animation is phenomenal.

01:27:55   And it's an amazingly easy way to say, "Yeah, make that thing just kind of dance over to

01:27:59   that side of the screen."

01:28:01   And it takes very little code to make really, really fluid, really impressive animations

01:28:05   happen.

01:28:06   And that's in part why so many iOS apps look so darn good, is because for a developer,

01:28:11   even one who has no visual chops like myself, you can make really decent animations really,

01:28:16   really easily.

01:28:17   And the fact that core animation got ported, or whatever, is pretty impressive to me.

01:28:22   So I've been told on Twitter that the session where they talk about the mechanisms behind

01:28:27   all of this will be happening tomorrow afternoon, which is Thursday afternoon as we record this.

01:28:32   I don't know if I'll have the time to watch it anytime soon because life has been a little

01:28:35   busy lately, but I hope at some point that I'll be able to check it out and report back on how

01:28:41   this is all held together because this certainly sounds extremely impressive, although we'll see

01:28:46   if it's actually really particularly functional over the next few weeks, I suppose. That being

01:28:53   said, Candy Crush Saga, the bastion of excellent programming, has already been running on this

01:28:58   project Islandwood.

01:29:00   It sounds like they kind of pulled a GNU step. Do you guys remember GNU step?

01:29:03   I've certainly heard the name, but...

01:29:04   It might still be out there. So if you have... Apple doesn't give you the source code, like

01:29:09   all of Cocoa and all of UIKit. Like some of the Darwin source code includes some, you

01:29:13   know, I think it's some core foundation stuff is in there, whatever. But anyway, most of

01:29:16   the stuff is not open source. But they do publish, you know, the documentation and you

01:29:19   get the header files for all the objective stuff.

01:29:23   The APIs are defined, and what they do is roughly defined.

01:29:26   So you could just look at the APIs and the headers and the documentation and use an actual

01:29:31   Mac to figure out how things work, and you could make a worker-like library for your

01:29:35   platform.

01:29:36   And that's kind of what GnuStep is.

01:29:37   It was trying to take the OpenStep stuff, which also wasn't open source, I don't know

01:29:41   if it ever was, and say, "We would like to be able to write programs with lots of square

01:29:46   brackets on Linux too, but they're not going to give us their code, so can we just sort

01:29:51   of reverse engineering and make a set of libraries with exactly the same functions, take exactly

01:29:54   the same arguments that behave in exactly the same way, but ours will be all open source.

01:29:59   And that effort, you know, with like any sort of open source volunteer-led effort, it's

01:30:03   very difficult to keep pace with Apple and all their engineers adding changes to Cocoa

01:30:08   and everything, so I don't know how far GnuStep has kept up, but this effort seems like that,

01:30:14   funded by Microsoft and a lot more people.

01:30:16   If you're gonna compile my UI kit application source code

01:30:20   without having Apple's code for the UI kit frameworks,

01:30:24   you will have to have frameworks with the same names,

01:30:26   with the same functions in them to link against,

01:30:28   otherwise nothing is going to work.

01:30:30   And where are you gonna get those from?

01:30:31   You're gonna have to write them yourselves

01:30:32   and make sure they behave in the same way as Apple's do.

01:30:35   And that is a humongous job.

01:30:37   And that is so unlike Microsoft,

01:30:38   because now you're sort of signing up

01:30:40   to try to keep pace with Apple's rapid development

01:30:42   of their libraries.

01:30:44   and you're telling people to, you know,

01:30:47   not even port their applications,

01:30:48   just bring your source code over

01:30:49   and it will compile on our platform

01:30:51   as a way to get more applications for your platform.

01:30:54   But like, you don't even have your own platform

01:30:55   at that point.

01:30:56   All you have is another place where, you know,

01:30:58   people are writing it's Apple's platform

01:30:59   and they can also, by the way, reuse that work

01:31:01   to sort of go on your platform as an afterthought,

01:31:03   but you don't even own your platform at that point.

01:31:05   Same thing with the Android apps.

01:31:07   Like, why does Microsoft even bother having an API if,

01:31:12   you know, I understand what they're trying to do.

01:31:13   like, oh, we need more apps on our platform.

01:31:15   We can get more apps by saying, hey, it's really easy

01:31:17   for you to take your existing Android or iOS app

01:31:19   and bring it to our platform.

01:31:20   But then do you even have a platform at that point?

01:31:22   Who in the world is ever gonna write against your API?

01:31:25   Are you really gonna put a lot of effort

01:31:26   into improving your API?

01:31:27   You're gonna be spending all your time

01:31:28   chasing these other APIs that you don't even control

01:31:31   that can be changed in ways that intentionally

01:31:33   or accidentally break your compatibility.

01:31:35   Both of these companies could decide to sue you,

01:31:37   probably not Google, 'cause Oracle's already suing them

01:31:39   over the Java stuff, but Apple could get cranky

01:31:41   and be like, whoa, whoa, whoa.

01:31:43   Like the Cocoa API is copyrighted and you can't blah blah blah,

01:31:45   but that's the whole Oracle Java suit, right?

01:31:47   So depending on that, it shakes out,

01:31:49   Microsoft might not even be on strong legal footing.

01:31:52   This, I'm interested in the technical details as well,

01:31:55   but strategy wise, like I see the benefits.

01:32:00   We're desperate, we need to get more apps on our platform.

01:32:03   This is might be a way to do it.

01:32:04   And usually like file format wise,

01:32:06   this is an approach that has worked for Microsoft

01:32:08   in the past of like, we can read Word Perfect Format.

01:32:10   I don't know if that's a good example

01:32:11   'cause I don't remember if they did that,

01:32:12   But like the way to get, if you're at market

01:32:15   and there's an entrenched leader,

01:32:16   you need your application to,

01:32:17   you need to make it so that using your application

01:32:20   is as easy as possible for the people

01:32:22   who use your competing application.

01:32:23   Like, oh, it's not a big deal.

01:32:24   We can read your old files, we'll import them.

01:32:26   We can save in your old format.

01:32:27   Like everything is great.

01:32:29   Like sort of just be a smooth upgrade path

01:32:32   because you're not in a position to say,

01:32:34   drop everything and come to our platform.

01:32:36   They tried that, it didn't work out.

01:32:38   Now they're trying to say, it's not a big deal.

01:32:39   Like I know you got an iOS app,

01:32:41   but you might want to make a Windows version of it too.

01:32:43   It's really easy, you can just bring your source code over,

01:32:45   recompile it, a couple little tweaks here and there.

01:32:47   Hey, why not, right guys?

01:32:49   That is not a strong pitch,

01:32:50   but that's the thin edge of the wedge if you're lucky.

01:32:53   But if you're not lucky,

01:32:54   all you're doing is just further dooming your own platform

01:32:57   to never be a real platform that you control.

01:32:59   - Well, and I don't think this really solves the problem

01:33:03   of why there aren't more apps on Windows Phone

01:33:06   and Windows, all the more modern Windows platforms.

01:33:10   The problem is really not that we were waiting

01:33:15   for being able to share the same code,

01:33:16   because you said that,

01:33:17   I just said there's tons of problems with that,

01:33:18   tons of problems.

01:33:19   As a developer, I don't think I would trust that

01:33:22   to be a stable, long-term way to do it, first of all.

01:33:25   Like, if I come out there and if I,

01:33:28   let's say I move all my code into this thing,

01:33:31   and I can compile it, and it actually does work,

01:33:33   and I can chip over cats for Windows Phone

01:33:34   using this crazy setup.

01:33:36   I'm gonna then tell people this,

01:33:39   I'm gonna announce this and then I'm gonna get customers

01:33:42   who are on Windows phones who are paying for the app

01:33:46   that's all, it's all based on this crazy setup

01:33:50   continuing to work into the future.

01:33:52   And that's something that I would not wanna rest

01:33:55   my customer trust on and my reputation on.

01:33:58   And it would definitely become yet another thing

01:34:03   I would have to maintain separately,

01:34:04   yet another thing that would have separate support issues,

01:34:07   separate bugs, it would have all the downsides of a new platform in addition to just the

01:34:14   massive liability of this compatibility layer that it would all depend on. And I wouldn't

01:34:20   trust that. And the reason why there aren't more developers making apps for the Windows

01:34:26   phone system and all the, you know, the artists probably known as Metro and all this stuff,

01:34:31   the reason why there aren't more of these is not because we were waiting for our code

01:34:35   to cross compile. It's because, A, none of us use these platforms, basically, you know,

01:34:42   there are some, but it's effectively zero relative to the market, so none of the developers

01:34:46   who are making the kinds of apps that people want to be in these platforms are using them

01:34:50   themselves and there just aren't that many people in the marketplace buying these platforms,

01:34:56   especially the kind of people who would be willing and able to install the kind of apps

01:35:03   that we make. And that's a big limitation right there. That rules out most enterprise

01:35:09   customers, that rules out a whole lot of Windows usage. So people who would actually buy these

01:35:15   apps, the kind of apps that iOS programmers make, and Android programmers make, people

01:35:20   who would actually buy them and use them, whether there's any market there, that's a

01:35:23   very big question. And they just don't have the unit sales to support that in the mobile

01:35:29   platform and on the PC platform it's usually not used that way. So there really is no reason

01:35:37   for developers to spend a whole lot of time on Windows Phone right now. And that might

01:35:42   change in the future. But right now, it's not justified. And this isn't going to change

01:35:49   that. Anything they do on the code side is not going to change that. They need to change

01:35:52   the market. Like the market of their devices and of their platforms. That's what has to

01:35:57   change. I don't know how you go about changing that, but there's no amount of like bribing

01:36:03   developers or making things easy for developers that's going to really solve that problem

01:36:08   on a big scale.

01:36:09   Well, you can see that the company that makes Candy Crush was like King something, I forgot

01:36:13   what their name is, is not bothered by the sort of, you know, reputation and support

01:36:18   concerns that you are because they're like, "It's shovelware. That's a term in the game

01:36:22   industry for a reason." "All right, we can make a Windows version real easily? Why not?

01:36:26   sure we'll do that. Like, whatever, especially with games where it's not much native UI anyway.

01:36:32   They either have the manpower to chase the weird bugs that happen on that, or they just

01:36:36   see it as a small amount of incremental income for not much extra money, or, you know, like,

01:36:42   this is going to get more applications on their platform, but I don't know if it's a

01:36:45   net win for them. And as for like, addressing it like, "Well, if you sold more phones, you

01:36:49   have more software, and if you had more software, you might sell more phones," like, they're

01:36:52   trying it from every possible angle. Again, not that they're in the same position as Apple

01:36:55   was in the 90s, but I get a similar vibe.

01:36:57   Like back when Apple was desperate,

01:36:59   much more desperate than Microsoft,

01:37:01   hope will probably ever be,

01:37:02   they were willing to try all those ideas

01:37:05   that people had been telling them for years

01:37:06   they should do.

01:37:07   Like, you know, try licensing your OS,

01:37:09   try making a TV, try making x86 compatible things,

01:37:13   try selling your computers in Sears.

01:37:15   You know, like they were doing everything

01:37:17   that anyone had ever told them that you might want to try.

01:37:20   And all of them, you know,

01:37:22   like there was reasons they hadn't done them in the past.

01:37:25   right the clones one is the worst one it's like this thing that everyone would

01:37:27   tell them you should license your operating system usually and it's like

01:37:30   with Microsoft it's like you shouldn't be so stubborn and try to invent

01:37:33   everything yourself use things that other people have allow iOS apps to

01:37:37   running your things around Android apps to run on Windows make Windows free make

01:37:41   Windows open source like just all the list of like as we go on the list of the

01:37:45   things that people have said Microsoft should try as it slowly tries them you

01:37:49   realize they were always bad ideas or it's too late or both. So I don't know.

01:37:56   I kind of like the new Microsoft and they're doing exciting things but

01:38:01   I feel like they're losing control of their platform. I feel like they had a

01:38:06   good platform and a potential growth path and just through a series of bad

01:38:09   mistakes they're no longer masters of their own destiny in so many ways

01:38:14   is that it's depressing.

01:38:17   - But I agree with that,

01:38:19   but man is it interesting to watch, isn't it?

01:38:21   I mean, it's just so fun to watch what they're doing.

01:38:25   Even if they're making more missteps,

01:38:27   which I'm not saying,

01:38:28   I don't know if they're making missteps,

01:38:29   but they are definitely desperate, like you said,

01:38:32   but gosh, is it fun to watch?

01:38:33   - Well, and they do fail fast for the most part.

01:38:35   Like they're trying lots of things,

01:38:37   and if they don't stick, they move on.

01:38:38   And in one respect that makes you not trustworthy

01:38:41   as a company, 'cause people don't wanna buy your product,

01:38:43   are like, do I really want to buy that? Because if it doesn't succeed really well, it's going

01:38:47   to be a dead end. On the other hand, they do need to find something that works. And

01:38:51   so it's better than picking a strategy that's going to save the company and sticking with

01:38:55   it for five years, stubbornly refusing to see that your strategy is not working. Like

01:39:00   they are, you know, like what was that? The Microsoft wristband thing? What is that called?

01:39:05   Microsoft band, I believe. Yeah. Like, so that was, I thought that was a reasonable

01:39:09   product. But did it sound big enough numbers that they're going to make a Microsoft band

01:39:12   and two and three and four.

01:39:14   If you bought a Microsoft band,

01:39:15   are you gonna be kind of sore that like,

01:39:17   you know, I kind of like my Microsoft band

01:39:19   and they're just not making one 'cause it wasn't a hit

01:39:21   or 'cause the Apple Watch came out and it couldn't compete.

01:39:23   Like, I don't know what the balance is there.

01:39:26   Should they abandon Microsoft band

01:39:27   because it didn't work out or should they say,

01:39:29   you know what, it was a pretty decent version one product.

01:39:31   It is differentiated from the Apple Watch in interesting way.

01:39:34   Let's make a version two and a version three

01:39:35   and a version four.

01:39:36   I don't quite know what the right thing to do there.

01:39:39   I think Microsoft has done much stronger footing

01:39:41   with its Azure stuff, because that is,

01:39:44   they are the master of their own destiny there.

01:39:46   That platform is different than Amazon Web Services

01:39:49   and EC2 and whatever the hell Google is doing.

01:39:51   Like it is a contemporary, it is in the running

01:39:55   with all the other things that are out there.

01:39:56   It has a lot of good reports.

01:39:58   It's not the great things about that

01:39:59   or all the good things about the new Microsoft.

01:40:00   Don't tie it to Windows only.

01:40:02   Don't make it a Trojan horse to try to get our technology

01:40:04   into other people's things.

01:40:05   Be open, use JSON, accept other languages,

01:40:08   be really good at what you do.

01:40:09   That I think still is their flagship,

01:40:12   that and obviously Xbox.

01:40:13   Other two flagship, non-embarrassing, no excuses,

01:40:17   no stupid Microsoft BS for the most part,

01:40:19   except a little bit on the Xbox One products.

01:40:21   But Windows at this point,

01:40:23   and everything involved with Windows,

01:40:25   and Windows on phones, and Windows on computers,

01:40:27   and Windows on tablets, it's just a mess.

01:40:30   - Yep, I don't know, it's interesting to watch.

01:40:33   And I am very surprised by some of the moves they're making.

01:40:37   But we'll see.

01:40:38   And I'm definitely going to try to watch the session tomorrow afternoon and see what the

01:40:42   technical explanation is behind all this.

01:40:46   I don't think it's really going to solve anything for them for all the reasons that both of

01:40:49   you guys expressed.

01:40:50   On top of that, even if I could cross-compile, say, FastText for the sake of argument, I

01:40:56   could cross-compile FastText, it's going to be on an operating system that has a wildly

01:41:01   different paradigm than iOS.

01:41:03   And so it's going to feel out of place.

01:41:05   This has a ton of issues associated with it.

01:41:08   What am I testing it on?

01:41:09   Am I gonna go buy Illumia just to test for fast text?

01:41:13   So many problems here,

01:41:14   but I think they are trying to cover all the bases,

01:41:17   and they wanna be able to say,

01:41:18   "Well, you know what?

01:41:20   "This Windows phone thing didn't really work out,

01:41:22   "but we did everything we could.

01:41:23   "We even cross-compiled Objective-C.

01:41:26   "That's how serious we were about it,

01:41:28   "and you know what?

01:41:29   "It just didn't work."

01:41:30   - Yeah, how did that team,

01:41:31   how must that team have felt when Swift was announced?

01:41:34   (laughing)

01:41:35   doing all this work to have this Objective-C cross compiler

01:41:38   and all this stuff and then they announced like,

01:41:40   seriously, I mean, I don't know,

01:41:41   maybe they didn't even start this project until after,

01:41:43   but I have to assume that it's so much work

01:41:44   that it was happening then and like,

01:41:46   that's what everyone said, it's like,

01:41:48   well, you know, you're kind of skating to where the puck was

01:41:51   on the whole Objective-C cross compiling thing there,

01:41:53   at least it seems that way.

01:41:55   - Yep.

01:41:57   - All right, thanks a lot to our three sponsors this week,

01:41:59   Backblaze, Casper, and Harry's,

01:42:02   and we will see you next week.

01:42:04   Now the show is over, they didn't even mean to begin

01:42:11   'Cause it was accidental, oh it was accidental

01:42:16   John didn't do any research, Marco and Casey wouldn't let him

01:42:21   'Cause it was accidental, oh it was accidental

01:42:27   And you can find the show notes at ATP.fm

01:42:32   And if you're into Twitter, you can follow them

01:42:37   @caseyliss

01:42:41   So that's Casey Liss, M-A-R-C-O-A-R-M

01:42:46   Auntie Marco Arment, S-I-R-A-C

01:42:51   USA, Syracuse, it's accidental

01:42:56   They didn't mean to, accidental

01:43:01   Is there anything else we have to talk about?

01:43:08   I can tell you more about the watch. I mean, I have, so on this week's Connected,

01:43:13   I haven't listened yet. Okay, well, I'm going to spoil one part of it. Is this a nose

01:43:18   spoiler? It is a nose spoiler. So Mike was saying that because the watch is really not

01:43:26   a two is really not a one-handed device like you know the hand that that it's

01:43:30   being worn on can't interact with the screen unless you have some crazy hand

01:43:34   that I've never seen before but for the most part most human hands won't be able

01:43:38   to interact with it from the hand that's wearing the watch so you have to use

01:43:41   your other hand to actually touch the screen or the buttons so really it's a

01:43:46   two-handed device it's not like a phone you can use entirely within one hand

01:43:51   because the same hand holding it can be using it unless you get a six plus

01:43:55   Watches are not that way.

01:43:56   So Mike was saying that he has already on many occasions

01:44:01   used his nose to touch the screen on the watch,

01:44:03   even going as far as to hit buttons with his nose

01:44:07   to reply to text messages, which is awesome.

01:44:10   And they all made fun of him, unconnected, for saying this.

01:44:13   However, I'm here to say that I have done that too.

01:44:16   I have not gone as far as pushing buttons,

01:44:18   but I have used my nose to wake up the screen

01:44:21   to show me the time if the wrist thing is failing

01:44:24   and I need to see the time for something.

01:44:26   - May I offer my gross suggestion?

01:44:28   - Absolutely.

01:44:29   - My gross suggestion and also my suggestion

01:44:31   as someone with a very large nose

01:44:33   is that your tongue is probably

01:44:34   a more precise pointing device.

01:44:36   - That is a gross suggestion.

01:44:38   - But I don't wanna get my screen all wet or lick it.

01:44:40   - That's true, I said it was gross.

01:44:42   I'm just throwing that out there.

01:44:43   If accuracy is your concern and germs are not

01:44:45   or you want to strengthen your immune system

01:44:47   by taking in lots of foreign germs, go for the tongue.

01:44:51   - As someone who is blessed with a particularly

01:44:55   large schnoz from--

01:44:56   - What are you talking about?

01:44:58   You do not have a big nose.

01:44:59   - I do have a big nose.

01:45:00   I'm half Italian and half Jewish.

01:45:02   How do I not have a big nose?

01:45:03   - You don't.

01:45:04   I have a big nose.

01:45:05   You have an average nose.

01:45:07   - Oh, whatever.

01:45:08   Well, anyways.

01:45:09   - Accidental nose podcast.

01:45:11   We can all compare our nose sizes.

01:45:13   - Marco's nose is the only boring small one.

01:45:17   - Anyway, the point being-- - Thanks?

01:45:18   (laughing)

01:45:21   There's so many jokes I can make here,

01:45:22   but I'm gonna let it go.

01:45:23   Anyway, so I use my schnoz, especially in bed,

01:45:27   to hit the back button. (laughing)

01:45:30   Oh God, to hit the back button on my iPhone.

01:45:34   I don't know why I do that all the time, but I do.

01:45:37   And I don't have a watch to try it on.

01:45:39   - Wait, wait, wait, wait, why?

01:45:41   To hit the back button, like in a web browser?

01:45:44   What are you?

01:45:44   - No, no, no, no, no, no, no.

01:45:45   Like the upper left-hand corner of the screen.

01:45:48   Because my hand isn't big enough

01:45:49   to reach the upper left-hand corner of the screen.

01:45:51   Why don't you just do a hand shimmy?

01:45:52   Why are you using your nose?

01:45:54   - You're over there using your tongue.

01:45:56   - I'm not, no, I am not using my tongue.

01:45:58   I have, I can use, I can hit everything

01:46:00   with one hand on my phone.

01:46:01   I don't have a watch.

01:46:01   I'm just saying for the people who find themselves

01:46:03   like carrying a bag of groceries and having the watch

01:46:05   and they need to hit a specific button,

01:46:07   for Mike specifically,

01:46:08   I don't feel like I could hit it with my nose.

01:46:10   I feel like my nose touches the entire screen

01:46:12   or none of the screen.

01:46:14   That's not a precision pointing device.

01:46:16   - Well, either way, it's so much quicker

01:46:19   than the hand shimmy to just smack the damn phone

01:46:23   with my nose and so much easier.

01:46:27   - Have you injured yourself doing this?

01:46:28   'Cause that would be a tough injury to explain.

01:46:30   - No, but the amount of times I've dropped my phone

01:46:33   onto my face unrelated to using my nose

01:46:36   as a pointing device is immeasurable.

01:46:39   I've done that so many times, I can't even tell you.

01:46:42   - Are you like laying on your back

01:46:43   and then have it propped on your chest?

01:46:46   - No, typically I'm holding it up

01:46:47   and maybe it's like winter time

01:46:50   and I only have one hand out of the covers

01:46:52   because I'm cold.

01:46:53   And so rather than doing the shimmy

01:46:56   or pulling my extremely comfortably warm hand

01:47:01   out of the covers, I'll just smack with the nose.

01:47:05   - Does your wife see you do this?

01:47:06   - Oh yeah.

01:47:07   Oh, she's stuck.

01:47:08   We're married, she's stuck.

01:47:09   There's nothing she can do now.

01:47:10   - Just saying, if I saw somebody do that,

01:47:12   I wouldn't take a second look.

01:47:13   I'd be like, what is going on?

01:47:15   And I think if I saw someone do the watch with their nose,

01:47:18   especially again, if they were like struggling

01:47:20   with a bag of groceries, I would like understand.

01:47:22   But if an able-bodied person with two free hands

01:47:26   is hitting things on the phone with their nose,

01:47:28   I'd be like, that just doesn't read right to me.

01:47:33   I feel like there are better options.

01:47:34   - I hope that someday I get to see somebody else

01:47:38   like out in the world in real life

01:47:40   hit their smartwatch with their nose.

01:47:43   'Cause I'm trying to feel like,

01:47:44   I mean, I just said that I've done this,

01:47:47   but I don't think I've done it in public.

01:47:49   I think I've done it in my house.

01:47:51   I don't think I would do it in public.

01:47:53   I think I would be more embarrassed by that

01:47:56   than I would be by picking my nose in public.

01:47:58   - Well, it's no worse, and it's probably actually better

01:48:01   than the thing that I've done myself.

01:48:03   When coming into the house, carrying a bunch of things,

01:48:06   putting something in your mouth so you can get,

01:48:08   like putting, not your phone,

01:48:11   but like holding something in your mouth

01:48:12   so you can get a free hand to do something, you know?

01:48:15   Like when you're carrying a lot of things,

01:48:17   using your mouth as a third hand to hold something.

01:48:21   And usually what you're holding is just as gross or grosser

01:48:23   than your watch might be, you know what I mean?

01:48:26   - I don't do that, that's gross.

01:48:28   - Well, you don't carry a lot of things, I guess.

01:48:30   (laughing)

01:48:31   - No, I carry them in my hands.

01:48:33   - Yeah, but when you have a lot of stuff and you know,

01:48:35   or even if it's like someone calls you on the phone

01:48:37   and you want to answer the phone,

01:48:38   but you have your keys and you have something else,

01:48:40   you put your keys in your mouth,

01:48:41   at the phone and say, "Oh." You know, you see it in movies all the time, and I know

01:48:44   I've done it myself. And you do the sort of lip curl under things so you're not actually

01:48:47   like touching your mouth mouth to the thing you're holding.

01:48:50   It's just as bad.

01:48:51   Yeah, probably. Anyway, I think the whole germaphobe thing of like I don't want to put

01:48:56   anything in my mouth that might have germs on it, like if you do that, that will not

01:48:59   lead to a strong immune system. You should be taking dirt filled with animal poop and

01:49:03   shoving it into your mouth when you're an infant and blah blah blah.

01:49:06   Oh, God.

01:49:07   Wow.

01:49:08   Oh, man.

01:49:09   Oh man.

01:49:10   And rest assured that your toddler either is or will be, in both of your cases, doing

01:49:15   that anyway.

01:49:16   So they are strengthening their immune system.

01:49:18   Lay off the antibiotic soap.

01:49:21   [ Silence ]