ATP

00:00:00 ◼   ► Time was when all AOL users were behind one big proxy.

00:00:04 ◼   ► We have important follow up this week.

00:00:08 ◼   ► First of all, if you got two t-shirts, that is fine.

00:00:12 ◼   ► That was supposed to happen. What happened was the first t-shirt

00:00:16 ◼   ► had some printing problems, mostly on the back, where

00:00:20 ◼   ► the font wasn't rendered, so it like

00:00:24 ◼   ► fell back to a default font and it printed our monospace code

00:00:28 ◼   ► in a fixed width font.

00:00:29 ◼   ► - Variable width font you mean.

00:00:31 ◼   ► - Yes, sorry, thank you.

00:00:32 ◼   ► I've missed you, John.

00:00:34 ◼   ► And then--

00:00:35 ◼   ► (laughing)

00:00:36 ◼   ► - I was just picturing all the feedback

00:00:38 ◼   ► if we just left that slide.

00:00:39 ◼   ► - Thanks.

00:00:40 ◼   ► So anyway, yeah, so the second one is

00:00:43 ◼   ► it has the fixed font issue,

00:00:45 ◼   ► 'cause it also messed up all the spacing

00:00:47 ◼   ► and everything, it was really weird looking before.

00:00:49 ◼   ► So Teespring was kind enough to reprint all the shirts

00:00:52 ◼   ► at no cost and send them to everybody.

00:00:55 ◼   ► That was supposed to happen, so it was no mistake

00:00:57 ◼   ► if you got a second one that's in fact it might be a mistake if you didn't and

00:01:01 ◼   ► You don't have to return the other ones that you got or anything

00:01:04 ◼   ► Just you know, keep them with teespring's compliments and we thank teespring for fixing the error so quickly and fully

00:01:11 ◼   ► And the question is which shirt now is more valuable

00:01:14 ◼   ► As we mentioned in the past shows about that the shirt being printed if the printing on the back messes up

00:01:19 ◼   ► It's like the upside-down airplane stamps and now you have two shirts

00:01:22 ◼   ► you have one where the printing on the back is totally messed up and it's like

00:01:27 ◼   ► In a variable with font but the metrics of the letter spacing are like the metrics of a mod the mono space font

00:01:32 ◼   ► Yeah, crazy amounts of space. It just looks terrible

00:01:35 ◼   ► And then there's the one printed the way we intended which one of those shirts is better now because you could be like I've got

00:01:41 ◼   ► One of the ATP shirts with the printing here, but everybody's got one with with the printing error at this point

00:01:45 ◼   ► Anyone who's got any teachers, right?

00:01:47 ◼   ► It's the same amount of both and it will always be the same amount of both because it's not like you can go order the

00:01:52 ◼   ► shirt now and get the fixed one only. It's the same quantity of both shirts that will

00:01:57 ◼   ► probably ever exist.

00:01:58 ◼   ► Yeah, so next year we'll probably have a different shirt and we'll see how this turns

00:02:02 ◼   ► out. But we've been fielding a lot of tweets and emails from people asking about the double

00:02:06 ◼   ► shirts. Just keep them. And a lot of emails and tweets from people who said they got two

00:02:11 ◼   ► shirts but did not notice any difference between them. And that was surprising to me for listeners

00:02:16 ◼   ► of the show.

00:02:17 ◼   ► Don't tell Teespring.

00:02:18 ◼   ► (laughing)

00:02:20 ◼   ► - Well, but there's definitely a difference.

00:02:21 ◼   ► I feel like once we tell them,

00:02:22 ◼   ► oh, there was a problem with the printing on the back,

00:02:24 ◼   ► and they lay them both down and look at the two backs,

00:02:26 ◼   ► there's a big difference between them.

00:02:28 ◼   ► - Yeah, also the ATP badge on the front,

00:02:32 ◼   ► on the new shirt is a little bit lighter in the shades

00:02:34 ◼   ► because when you give somebody a color to print on a shirt,

00:02:38 ◼   ► it has to go through color conversion

00:02:40 ◼   ► and become a print color,

00:02:41 ◼   ► and that entire world just sucks in every possible way.

00:02:45 ◼   ► The entire world of color converting for print

00:02:47 ◼   ► and trying to get colors to look right in print.

00:02:49 ◼   ► It's just a terrible, terrible existence.

00:02:52 ◼   ► And here's to Teespring for living in that world

00:02:55 ◼   ► so that we all don't have to.

00:02:57 ◼   ► - Yep, now that was very nice of them to write that wrong.

00:03:00 ◼   ► And regardless of who was at fault,

00:03:02 ◼   ► it sounds like it might have been a little bit

00:03:03 ◼   ► of column A, a little bit of column B, but--

00:03:05 ◼   ► - Oh no, Teespring was entirely at fault.

00:03:06 ◼   ► That's not saying it was at fault.

00:03:08 ◼   ► Because, I mean, so here's the thing.

00:03:10 ◼   ► Like, I feel bad for them.

00:03:11 ◼   ► It's good, Teespring has always been good to me

00:03:13 ◼   ► and they've taken action to correct problems,

00:03:17 ◼   ► but it's better if you don't have the problem to begin with.

00:03:19 ◼   ► And I'm speaking mostly from Teespring's perspective,

00:03:21 ◼   ► because from the customer's perspective,

00:03:22 ◼   ► everybody got two shirts for the price of one.

00:03:25 ◼   ► So it's great for customers, but for Teespring,

00:03:27 ◼   ► like Marco said, it's like, you know,

00:03:28 ◼   ► he uploaded the thing, it looked correct in the preview,

00:03:31 ◼   ► and they printed a huge number of shirts.

00:03:33 ◼   ► Like, no human looked at them to compare them

00:03:35 ◼   ► to the preview on their website.

00:03:37 ◼   ► Like, you figured you'd print one, look at it and say,

00:03:39 ◼   ► "Yep, okay, print a thousand or so of those," right?

00:03:42 ◼   ► So I feel like they need to amend their process,

00:03:45 ◼   ► because they're a great company that they're eating

00:03:47 ◼   ► that cost and doing it, but on the other hand,

00:03:49 ◼   ► it's kind of their fault, and I'd feel like they would have

00:03:52 ◼   ► processes in place to do a human sanity check

00:03:55 ◼   ► of runs over a thousand shirts or something.

00:03:57 ◼   ► - Well yeah, and in their defense,

00:04:00 ◼   ► as soon as they became aware of the problem,

00:04:02 ◼   ► they were in contact with me a lot,

00:04:04 ◼   ► and I had talked to like three different people there,

00:04:06 ◼   ► and everyone was trying to resolve the thing,

00:04:08 ◼   ► they wanted me to send them the original file

00:04:09 ◼   ► that caused the error so they could fix it,

00:04:11 ◼   ► and they told me they're changing all these policies,

00:04:13 ◼   ► So I think it scared them enough and probably caused

00:04:16 ◼   ► them to lose enough money that they made some changes

00:04:20 ◼   ► as a result.

00:04:21 ◼   ► I mean, they bent over backwards for my hypercritical shirts

00:04:23 ◼   ► as well.

00:04:24 ◼   ► They've definitely been--

00:04:25 ◼   ► I feel like they have that part of the business down

00:04:27 ◼   ► where you're responsive to customers

00:04:28 ◼   ► and do the right thing.

00:04:29 ◼   ► They've just got to work on the other part of the business

00:04:31 ◼   ► where you don't make mistakes to begin with.

00:04:33 ◼   ► And I mean, if I wanted to dink Marco for anything,

00:04:35 ◼   ► I assumed you were uploading outlines.

00:04:36 ◼   ► If I had known you were uploading text,

00:04:38 ◼   ► I would have advised you to upload outlines.

00:04:40 ◼   ► But--

00:04:40 ◼   ► In my defense, yeah.

00:04:41 ◼   ► So the issue was they recommend you upload an EPS.

00:04:44 ◼   ► Okay, and I don't know anything about this stuff.

00:04:46 ◼   ► This was one of the first times I'd ever used Illustrator

00:04:48 ◼   ► was to do this.

00:04:49 ◼   ► And I uploaded the EPS and I checked

00:04:52 ◼   ► that it had the embedded font.

00:04:54 ◼   ► And I didn't convert to outlines.

00:04:56 ◼   ► I didn't know that was a thing that you could or should do.

00:04:58 ◼   ► So I didn't convert to outlines.

00:04:59 ◼   ► I just uploaded the EPS with the embedded copy of the font.

00:05:03 ◼   ► And every designer, once we had the problem

00:05:06 ◼   ► and I mentioned this, every designer was like,

00:05:07 ◼   ► "Oh no, you never do that

00:05:09 ◼   ► "because printers so often screw up."

00:05:10 ◼   ► "Oh, I didn't know that.

00:05:11 ◼   ► "I didn't have the experience to know that."

00:05:12 ◼   ► And so I thought, I put the embedded font, it's fine, right?

00:05:16 ◼   ► And then the bigger problem, I think,

00:05:18 ◼   ► was that in their image preview on the site,

00:05:22 ◼   ► it rendered it correctly because the thing

00:05:25 ◼   ► that generates the image preview has a different rasterizer

00:05:29 ◼   ► than the thing that actually prints onto the shirts.

00:05:31 ◼   ► And so that's kind of a problem.

00:05:33 ◼   ► - Yeah, and that's why it's ultimately Teespring's fault.

00:05:36 ◼   ► Marco is the inexperienced of not converting to Outline

00:05:38 ◼   ► because he hadn't been using Illustrator since Illustrator 88

00:05:42 ◼   ► on his black and white Mac and comparing things to outlines.

00:05:45 ◼   ► But the preview shows one thing, and you

00:05:48 ◼   ► get something else printed.

00:05:50 ◼   ► That's the whole basis of the site.

00:05:51 ◼   ► It's like you click these buttons, you see a picture,

00:05:53 ◼   ► you say, yes, I want that picture to become a real thing.

00:05:55 ◼   ► You click some more buttons, and then a real shirt

00:05:58 ◼   ► shows up at your house.

00:05:59 ◼   ► Right.

00:05:59 ◼   ► And in their defense, one of the things they asked me for

00:06:02 ◼   ► was a copy of the original file, because they

00:06:04 ◼   ► said their rasterizer should have used the embedded font.

00:06:07 ◼   ► that was not intended behavior that it didn't,

00:06:09 ◼   ► and they didn't want to require everyone to use outlines

00:06:11 ◼   ► 'cause Teespring's used by a lot of amateurs.

00:06:14 ◼   ► It's not always professional designers using it.

00:06:16 ◼   ► It's oftentimes like some group making shirts

00:06:20 ◼   ► to raise money for their youth group or something.

00:06:22 ◼   ► And so they want it to work with people

00:06:26 ◼   ► who are not Illustrator experts.

00:06:28 ◼   ► So they, anyway, they resolved it well,

00:06:30 ◼   ► so I'm happy with them.

00:06:31 ◼   ► Anyway, let's move on.

00:06:33 ◼   ► We have a lot of feedback.

00:06:34 ◼   ► Our feedback is like 15 pages long in our document.

00:06:37 ◼   ► What is all this?

00:06:38 ◼   ► I know.

00:06:39 ◼   ► I think it's mostly John because I certainly didn't do it, but another really quick one.

00:06:43 ◼   ► We were all obviously in San Francisco last week for WWDC and I just wanted to quickly

00:06:49 ◼   ► say thank you to everyone who said hi to us, to everyone that purchased and then wore a

00:06:56 ◼   ► shirt, to anyone who went to the talk show and to Gruber for having us all on the talk

00:07:01 ◼   ► show.

00:07:02 ◼   ► I had a really fantastic week.

00:07:03 ◼   ► I think I speak for the two of you guys as well in saying so and so again anyone who

00:07:08 ◼   ► Spent any amount of time with us even if it was just to say hey

00:07:11 ◼   ► Thank thank you for that and I have some more follow-up on the talk show

00:07:16 ◼   ► Oh, and it goes in addition to saying yeah, I agree. That was awesome. Thank you

00:07:20 ◼   ► You know the the the the ditto of what Casey you said?

00:07:22 ◼   ► I was completely wrong about how cloud kit limits work and I I

00:07:28 ◼   ► Definitely said the wrong thing on the talk show and I think I even said the wrong thing on our show last week

00:07:33 ◼   ► where the CloudKit limits are raised per user by something like 1 meg of database use, 100

00:07:41 ◼   ► megs of assets use, but each user is not limited to that 1 meg and that 100 megs.

00:07:47 ◼   ► That simply, every user adds that amount to your total pool, and the resources are pooled

00:07:52 ◼   ► between everybody.

00:07:53 ◼   ► So that does dramatically change things for how you can use CloudKit and whether you can

00:07:59 ◼   ► rely on it being sufficient for you or not.

00:08:02 ◼   ► It still leaves a few questions, like we still don't really know what happens when you hit

00:08:08 ◼   ► that limit, and anything involving like, you know, what happens if one user is a complete

00:08:17 ◼   ► outlier and uses like a gig of the database somehow because they have some script going

00:08:22 ◼   ► that runs awry or something, you know, like how do you solve issues like that with it?

00:08:26 ◼   ► So there's still some issues, but it isn't nearly as scary as I assumed it was because

00:08:31 ◼   ► the resources are pooled by that amount, not limited per user.

00:08:35 ◼   ► I'm not sure that clarification is important, but I'm not sure it changes it entirely.

00:08:41 ◼   ► We were talking about the example of "Could you do Instagram on this?"

00:08:46 ◼   ► Even though users will have varying amounts, if it's only 100 megs of blob storage per

00:08:51 ◼   ► user, any user who uses Instagram will blow through that in a year if they're regular

00:08:56 ◼   ► users.

00:08:57 ◼   ► It's like your average user will blow through that.

00:08:59 ◼   ► It opens it up a little bit to more variability.

00:09:03 ◼   ► If you think most of your users are gonna use like one meg

00:09:05 ◼   ► and a couple of people are gonna use like 500 megs,

00:09:07 ◼   ► then yeah, having it add 100 megs to the pool

00:09:11 ◼   ► for each user for blob storage is good.

00:09:14 ◼   ► But anything that's involving serious amounts of data,

00:09:16 ◼   ► you immediately get to that question,

00:09:18 ◼   ► all right, well, if I start my own little Instagram,

00:09:20 ◼   ► after a year, everybody's gonna have

00:09:22 ◼   ► more than 100 megs of photos, right?

00:09:23 ◼   ► So then what happens?

00:09:25 ◼   ► - Right, but the instance I gave on the talk show,

00:09:28 ◼   ► which was, could they use this for Vesper?

00:09:30 ◼   ► You know, Vesper is a note-taking app

00:09:31 ◼   ► that supports image blob attachments.

00:09:34 ◼   ► But most people don't use a lot of images,

00:09:36 ◼   ► most people just use it for text,

00:09:37 ◼   ► and most people would have a hard time

00:09:39 ◼   ► making a megabyte of text notes,

00:09:41 ◼   ► unless they used it very, very heavily.

00:09:42 ◼   ► But you know, so in Vesper's case,

00:09:44 ◼   ► it might be a bad idea to say,

00:09:47 ◼   ► you can only have up to a megabyte of text notes,

00:09:50 ◼   ► but I bet their average is so far below a megabyte

00:09:54 ◼   ► that they would be okay to use it if they wanted to.

00:09:57 ◼   ► I would pull stats from all active users

00:09:59 ◼   ► before I committed to that, you know what I mean?

00:10:01 ◼   ► Like, see what the average, find out,

00:10:03 ◼   ► you can find out what the average is,

00:10:05 ◼   ► like make sure, you know, whatever your user data

00:10:07 ◼   ► collection agreement allows for this or whatever,

00:10:09 ◼   ► and then just see what the average size is,

00:10:10 ◼   ► because who knows what people are doing with Vesper,

00:10:13 ◼   ► you know, like maybe people are using it

00:10:14 ◼   ► as a photo collecting app for all we know.

00:10:17 ◼   ► - One of the discussions they were having

00:10:19 ◼   ► before they launched the sync service was,

00:10:22 ◼   ► you know, like they don't actually know,

00:10:24 ◼   ► 'cause you know, before there was a sync service

00:10:25 ◼   ► a few weeks ago, they didn't actually know

00:10:27 ◼   ► what the average usage was of these things.

00:10:29 ◼   ► How much people actually, how many pictures,

00:10:31 ◼   ► 'cause you know, hosting pictures is obviously

00:10:33 ◼   ► a lot more data than hosting text,

00:10:35 ◼   ► and they're like, well how many pictures

00:10:36 ◼   ► do people take with a Vesper?

00:10:37 ◼   ► Are there a lot of image notes?

00:10:39 ◼   ► Are the images very large?

00:10:40 ◼   ► Are there outliers that have just tons and tons and tons

00:10:43 ◼   ► of image notes?

00:10:44 ◼   ► And you don't know that.

00:10:46 ◼   ► I'm facing this with Overcast now.

00:10:47 ◼   ► I'm actually getting reasonably close to launch now,

00:10:50 ◼   ► and I have no idea what to expect on the server side,

00:10:53 ◼   ► and therefore the cost side.

00:10:55 ◼   ► It's very hard to predict these things.

00:10:57 ◼   ► That's one of the reasons why CloudKit is very attractive

00:11:00 ◼   ► to a lot of people because if you have this rockin' lunch

00:11:05 ◼   ► where you get way more people than you thought you would,

00:11:09 ◼   ► your costs are not going to skyrocket,

00:11:12 ◼   ► well, unless you hit one of those master limits

00:11:16 ◼   ► that's really, really high,

00:11:17 ◼   ► but I think that would be quite difficult.

00:11:21 ◼   ► the unpredictability of your initial cost

00:11:24 ◼   ► and initial server needs is kind of removed

00:11:27 ◼   ► as a stress and pain and possible business model pain point.

00:11:31 ◼   ► So it is a very attractive option

00:11:34 ◼   ► and I think a lot of people are gonna use it.

00:11:37 ◼   ► - I hope somebody does so that we can see

00:11:39 ◼   ► if things have really improved

00:11:41 ◼   ► because I can think back to when you mentioned

00:11:43 ◼   ► if you have a really big launch,

00:11:44 ◼   ► you'll be happy you did this.

00:11:45 ◼   ► - Letterpress?

00:11:46 ◼   ► - Yeah, letterpress.

00:11:48 ◼   ► What's the first really popular app

00:11:50 ◼   ► to use this Apple-provided cloud service.

00:11:52 ◼   ► And in the case of Game Center, it was letterpress,

00:11:55 ◼   ► it was very popular, and Game Center fell over

00:11:58 ◼   ► and cried like a little baby.

00:12:00 ◼   ► Well, apparently, Apple's new Photos app and the Photos Sync

00:12:04 ◼   ► service is supposedly all built on CloudKit.

00:12:07 ◼   ► It's the same thing.

00:12:09 ◼   ► And who knows, they probably have some kind of reserve

00:12:11 ◼   ► capacity maybe, but I don't know.

00:12:13 ◼   ► Supposedly, that is what they are doing.

00:12:16 ◼   ► So in other news, it is currently 9/15 and I was trying to use my Showbot and Marco tried

00:12:26 ◼   ► to tell me…

00:12:27 ◼   ► It broke already.

00:12:28 ◼   ► Hold on.

00:12:29 ◼   ► Marco tried to tell me when I was writing the Showbot a couple of weeks ago that people

00:12:35 ◼   ► – these are my words, not Marco's – that people are jerks and they're going to do

00:12:39 ◼   ► whatever they can to mess with the Showbot.

00:12:41 ◼   ► And I said, "No, no, it'll be fine.

00:12:44 ◼   ► Our listeners are great.

00:12:46 ◼   ► They're awesome.

00:12:47 ◼   ► And all of a sudden while you were talking, I look at the Showbot, which is all run kind

00:12:52 ◼   ► of real-time using WebSockets, and next thing I know I see that the "Does Casey Showbot

00:12:59 ◼   ► Actually Work?"

00:13:00 ◼   ► title just skyrocketing in votes.

00:13:03 ◼   ► And so someone apparently has decided to make Web API requests or WebSocket requests to

00:13:12 ◼   ► add a vote to that particular item and then eventually crash the showbot.

00:13:17 ◼   ► They're not being jerks, they're being good testers.

00:13:19 ◼   ► Yeah, maybe. But anyway, so the point is that it lasted 15 minutes and that is how long

00:13:26 ◼   ► the chat room can handle a new toy to play with and then subsequently break.

00:13:33 ◼   ► You're gonna be a good father.

00:13:34 ◼   ► Have I mentioned that your showbot has never worked for me?

00:13:38 ◼   ► I mean it's never worked.

00:13:39 ◼   ► It is. Remember, I tried to use it once from work, and I said, "Oh, maybe it's not working

00:13:42 ◼   ► because of a firewall, because of some weird WebSocket thing," and I've tried it from home,

00:13:46 ◼   ► and all I ever see is the word "connecting." That's all I've ever seen of your Showbot.

00:13:49 ◼   ► Well, if you want, you can see that now.

00:13:50 ◼   ► Oh, shut up. I just restarted.

00:13:53 ◼   ► I know. And I'm looking at it now, and it says "connecting." So, I'm, you know, "Marko's

00:13:59 ◼   ► crappy PHP thing 1, Casey's Node Showbot 0."

00:14:03 ◼   ► Well, that's what I get for trying to do something fancy.

00:14:05 ◼   ► his is AJAX-y. It's even better, it's WebSockets. He leapt right past AJAX.

00:14:10 ◼   ► He went too far. Oh, wait, something happened. Now it says titles. No, something's happening.

00:14:14 ◼   ► So that was like a seven-minute load time.

00:14:17 ◼   ► If you remember, Casey, what I also told you is, you know, it would be nice if it had some

00:14:21 ◼   ► kind of persistent storage. That way, if it crashes, you don't lose all the titles that

00:14:25 ◼   ► have been suggested up until that point.

00:14:27 ◼   ► I know, I know. The problem is…

00:14:28 ◼   ► So it would be nice if it could write it out to a file every few seconds.

00:14:31 ◼   ► Well, but that's the thing. It's all kidding aside, I did look into this, and the problem

00:14:34 ◼   ► is because this is hosted on Heroku, because I generally like Heroku, the file system is

00:14:40 ◼   ► ephemeral. And so even if I dump something to file, then if I do anything, that file

00:14:46 ◼   ► will go away. Like if I restart the dyno like I just did, the file will go away.

00:14:50 ◼   ► Well, I should just make a web request to someone who's running a real web server.

00:14:55 ◼   ► I was waiting for it.

00:14:58 ◼   ► You know, I hate to tell you Casey, but my servers can save files. It's this crazy

00:15:02 ◼   ► I know it's bleeding edge brand new technology,

00:15:05 ◼   ► but my servers are actually able to save data to disk.

00:15:08 ◼   ► - Well, and that's the thing, is what I need to do

00:15:09 ◼   ► is, as people are recommending in the chat,

00:15:11 ◼   ► I need to use Postgres or some other SQL

00:15:13 ◼   ► or something like that.

00:15:15 ◼   ► - You don't need to use Postgres for show titles,

00:15:17 ◼   ► for crying out loud.

00:15:18 ◼   ► Make a file with numbers in it.

00:15:20 ◼   ► - That's why I didn't!

00:15:21 ◼   ► That's why I didn't.

00:15:22 ◼   ► But apparently Heroku's file system is all ephemeral,

00:15:25 ◼   ► or maybe it isn't, and I just don't realize it.

00:15:27 ◼   ► But the only reason I was supposed to,

00:15:28 ◼   ► or that was trying to bring this up,

00:15:30 ◼   ► was to point out that it took virtually no time for the chat room to just completely tear apart my

00:15:36 ◼   ► showbot. Now they're putting in script tags, putting in long titles.

00:15:41 ◼   ► Welcome to the web, Casey.

00:15:42 ◼   ► Have you ever used a computer?

00:15:44 ◼   ► Oh my god, this is why you a**holes don't have nice things. This is ridiculous.

00:15:48 ◼   ► I quit. I quit the show and I quit you all. Well, not you two, everyone else.

00:15:53 ◼   ► I love that you were so naive that you didn't think this would happen.

00:15:56 ◼   ► You're such a nice person, Casey. You really are.

00:15:59 ◼   ► I had faith in humanity and God how wrong I was.

00:16:02 ◼   ► I'm sorry.

00:16:03 ◼   ► Anyway, we can carry on.

00:16:05 ◼   ► I'm just going to ignore my own creation, whatever, you big jerks.

00:16:10 ◼   ► The three of us are cool.

00:16:11 ◼   ► It's everyone else I don't like.

00:16:14 ◼   ► Anyway, moving on.

00:16:16 ◼   ► All right, so we have some follow-up.

00:16:18 ◼   ► Next bit of follow-up.

00:16:19 ◼   ► This is after WWDC, a bunch of people tweeted this at me and I thought it was clever.

00:16:23 ◼   ► We didn't get to it in the actual WWDC show.

00:16:27 ◼   ► And I haven't actually looked into this because I have not seen any of the sessions related

00:16:30 ◼   ► to this because they're mostly iOS related and everything, but in the keynote they mentioned

00:16:35 ◼   ► bundles.

00:16:36 ◼   ► You can sell a bunch of apps on the App Store together for one price.

00:16:39 ◼   ► And apparently, kind of like in iTunes where there's the "complete this album" button where

00:16:43 ◼   ► you buy the album minus the cost of the songs you already bought in the album, there's the

00:16:50 ◼   ► "complete this bundle" thing on the App Store—I'm assuming there is because these people were

00:16:54 ◼   ► were telling me there was, where if you bought one app in a two app bundle, you could complete

00:16:58 ◼   ► this bundle by simply paying the difference for the second app. And people were proposing

00:17:03 ◼   ► this as a terrible confusing way to do upgrade pricing. Because what you would do is when

00:17:09 ◼   ► you have a new version of your app, you keep the first version of the app on the store,

00:17:12 ◼   ► you introduce the new version at whatever price you want to sell to new users. And then

00:17:16 ◼   ► you make a bundle that includes the old version and the new version for less than the sum

00:17:20 ◼   ► of their prices, which allows people who already own the old version to complete this bundle,

00:17:25 ◼   ► effectively giving them an upgrade price. Which, again, like I said, I think that is

00:17:28 ◼   ► terrible and confusing and not obvious, but it is a clever hack of the rules as presented

00:17:34 ◼   ► to me by these people who sent me this information. Do either of you know if that stuff is accurate

00:17:38 ◼   ► about "Complete this Bundle"?

00:17:39 ◼   ► I do not.

00:17:41 ◼   ► Are there any bundles actually in the store yet that we could even see? I don't think

00:17:43 ◼   ► there are. I don't think you can make them yet.

00:17:45 ◼   ► If this is all true and it actually works like this, people will do this for their upgrades

00:17:50 ◼   ► for upgrade pricing and it will be confusing to users.

00:17:53 ◼   ► And so that's a shame.

00:17:55 ◼   ► Imagine people on their websites trying to tell people like, "If you want upgrade pricing,

00:18:00 ◼   ► go get the bundle, but do complete this bundle."

00:18:02 ◼   ► And ugh, I don't know.

00:18:05 ◼   ► This is like, if this loophole exists, Apple needs to close it, not because it's terrible

00:18:09 ◼   ► to allow upgrade pricing, but because it's so confusing and you know developers will

00:18:12 ◼   ► jump on it in a second, everyone will have these stupid instructions on their website

00:18:16 ◼   ► telling people how to get upgrade pricing by buying a bundle, which is not obvious.

00:18:19 ◼   ► Yeah, well also, I mean, the biggest thing that I think would blow a hole in this is,

00:18:24 ◼   ► do you have to have all the apps that are in the bundle still in the app store by themselves?

00:18:29 ◼   ► Because if you do, that means you have to have your old version still for sale while

00:18:33 ◼   ► your new version is for sale, and that's terrible.

00:18:35 ◼   ► Well, so the idea, I mean, again, the terrible proposed idea is you keep the old one in the

00:18:40 ◼   ► store for a limited time because the upgrade pricing would be some window of

00:18:44 ◼   ► like a couple months and you raise the individual price of the like the

00:18:47 ◼   ► standalone old version to something crazy that no one would accidentally buy

00:18:50 ◼   ► of course you know that's like you hope no more actually buy until someone

00:18:54 ◼   ► actually buys your you know 999 dollar app and it's pissed off at you or

00:18:58 ◼   ► whatever but that all this is just like a terrible workaround to get into a

00:19:02 ◼   ► loophole to give your existing customers a cheaper price for your new app but it

00:19:06 ◼   ► has so many bad side effects they just need they need to close this whole and I

00:19:09 ◼   ► I don't know how they will close it,

00:19:10 ◼   ► because if they have a complete this bundle thing,

00:19:13 ◼   ► I don't know how you stop this from happening.

00:19:16 ◼   ► - Yeah, I think it's just one of those things,

00:19:17 ◼   ► like developers are, they're so desperate to,

00:19:20 ◼   ► they, we, are so desperate to get ongoing revenue,

00:19:25 ◼   ► because you know what happens in the app store,

00:19:26 ◼   ► when you have a paid app, this is what always happens,

00:19:28 ◼   ► you launch and you get this nice big spike of sales,

00:19:31 ◼   ► but then it starts dropping,

00:19:32 ◼   ► and sometimes that drop can be pretty steep.

00:19:35 ◼   ► So eventually, you know, you start making not enough money

00:19:38 ◼   ► from your current version and you're like,

00:19:39 ◼   ► okay, well now I wanna work on this new version

00:19:42 ◼   ► or we have been working on this new version,

00:19:43 ◼   ► I gotta release it, gotta make more money from this.

00:19:47 ◼   ► Now, in the olden days, you release a new version

00:19:50 ◼   ► and charge an upgrade price,

00:19:51 ◼   ► because the full price was generally high enough,

00:19:55 ◼   ► you know, if you're charging 100 bucks

00:19:57 ◼   ► for the full price of your app,

00:19:59 ◼   ► and then you release a new version,

00:20:00 ◼   ► your existing customers would like it a lot

00:20:02 ◼   ► if you gave them the new version for maybe 50 bucks

00:20:04 ◼   ► or 40 bucks or whatever.

00:20:05 ◼   ► So the upgrade discount thing happened.

00:20:07 ◼   ► Well now in the iOS app store,

00:20:11 ◼   ► there's pretty much no good way to do it,

00:20:13 ◼   ► or the Mac app store,

00:20:14 ◼   ► there's pretty much no good way to do that.

00:20:15 ◼   ► And developers are always looking for a way

00:20:18 ◼   ► to get around this and to do this,

00:20:20 ◼   ► but the reality is all these things are terrible hacks.

00:20:24 ◼   ► And they will all anger some portion of your customers

00:20:28 ◼   ► to some degree no matter how you do them.

00:20:30 ◼   ► And or cause you lots of support costs.

00:20:34 ◼   ► And so it just, it never, there is no good solution to this.

00:20:38 ◼   ► The only solution to this is what Apple's doing

00:20:42 ◼   ► with their apps in the App Store.

00:20:43 ◼   ► The few Apple apps that they still charge money for

00:20:46 ◼   ► in the App Store, which that number keeps going down,

00:20:48 ◼   ► but the few apps, like the Pro apps, they charge money for,

00:20:52 ◼   ► you know, Logic, Aperture, stuff like that, Final Cut,

00:20:55 ◼   ► they just, they drop all the prices down

00:20:57 ◼   ► to a fraction of what they used to be.

00:20:58 ◼   ► You know, these things used to be like $300, $600, $1,000.

00:21:02 ◼   ► Now, things that were $1,000 are now $300,

00:21:06 ◼   ► things that were $300 are now like 70 bucks,

00:21:09 ◼   ► Logic is only 200.

00:21:11 ◼   ► They dropped all their prices,

00:21:12 ◼   ► and then when a new version comes out,

00:21:14 ◼   ► it's just like Apple hardware.

00:21:16 ◼   ► The new version costs the same thing as the old version.

00:21:18 ◼   ► If you bought the old version, too bad.

00:21:19 ◼   ► You want the new one, pay full price.

00:21:21 ◼   ► And that's a little more palatable now than it used to be

00:21:24 ◼   ► because the prices have gotten lower.

00:21:26 ◼   ► It still kinda sucks, but in a lot of ways,

00:21:29 ◼   ► it's quote, fair.

00:21:31 ◼   ► It's also, from Apple's point of view,

00:21:32 ◼   ► it's also very, very simple.

00:21:34 ◼   ► The new version of the app is a brand new standalone app.

00:21:38 ◼   ► The old version is removed from the store,

00:21:40 ◼   ► so you can only buy the new one.

00:21:41 ◼   ► And upgrades are, you know, well,

00:21:44 ◼   ► you're paying less than you used to pay, be happy.

00:21:47 ◼   ► And so that same model applies to iOS,

00:21:50 ◼   ► but the prices of course are even lower.

00:21:52 ◼   ► Instead of $100, you're paying $1 or $5 at most.

00:21:55 ◼   ► You know, almost no one's paying over $5 on iOS

00:21:58 ◼   ► for anything really.

00:22:00 ◼   ► "If you buy Tweetbot one for three bucks

00:22:04 ◼   ► "and then next week, Tweetbot two comes out

00:22:06 ◼   ► "and it's another three bucks, too bad."

00:22:09 ◼   ► Like that's the attitude that we have to have.

00:22:12 ◼   ► That's their only choice really.

00:22:13 ◼   ► That's what Apple has done with their stuff.

00:22:15 ◼   ► They're leading the way on that

00:22:16 ◼   ► and I think people are pretty much used to it.

00:22:19 ◼   ► I mean people will complain no matter what you do

00:22:20 ◼   ► if you charge money.

00:22:21 ◼   ► There's always gonna be people who complain

00:22:22 ◼   ► but I don't think you'd have fewer people complaining

00:22:26 ◼   ► if you had upgrade pricing.

00:22:28 ◼   ► And so the reality is this is the world we live in now.

00:22:30 ◼   ► This is, you know, Apple has been very clear that

00:22:35 ◼   ► if you want to do some kind of upgrade pricing scheme,

00:22:37 ◼   ► figure out a way to make an in-app purchase.

00:22:38 ◼   ► You know, the in-app purchase is the way that you

00:22:41 ◼   ► add features to an existing app

00:22:42 ◼   ► and try to get more money from them.

00:22:44 ◼   ► That's it.

00:22:46 ◼   ► - So the chat room has come through

00:22:47 ◼   ► and linked us to the documentation for the app bundles

00:22:51 ◼   ► and two relevant rules from this page.

00:22:53 ◼   ► From Apple, app contains in a bundle must be available

00:22:56 ◼   ► for sale on their own as well,

00:22:57 ◼   ► so you do have to keep that other older app for sale in the store.

00:23:01 ◼   ► And app bundles also support Complete My Bundle.

00:23:04 ◼   ► Complete My Bundle provides customers who previously purchased one or more of the apps

00:23:07 ◼   ► included within the bundle an additional discount on the app bundle.

00:23:10 ◼   ► So I don't know if it's a straight subtraction of price or whatever, but...

00:23:13 ◼   ► Perfect.

00:23:14 ◼   ► So you can do this, but it's a terrible idea, too.

00:23:15 ◼   ► Yeah, and I assume people will do it, and I'm not sure how Apple will react to it.

00:23:20 ◼   ► Because if they just sort of shrug and go, "Well, they're within the rules we defined,"

00:23:25 ◼   ► I expect there to be some confusion and sadness.

00:23:27 ◼   ► but it's the App Store.

00:23:29 ◼   ► - If you have both versions of your app in the store

00:23:31 ◼   ► at the same time and they're both paid apps,

00:23:33 ◼   ► you're gonna get so many angry customers,

00:23:35 ◼   ► so many angry support emails, so many one-star reviews.

00:23:38 ◼   ► I think that problem will be self-correcting.

00:23:40 ◼   ► It's a terrible, terrible idea to have both versions

00:23:43 ◼   ► in the store at the same time

00:23:44 ◼   ► because people will accidentally buy the old one

00:23:46 ◼   ► and then be very angry that you spent money

00:23:48 ◼   ► on an obsolete app that doesn't entitle them to much

00:23:51 ◼   ► or any of a discount on the new one in total.

00:23:53 ◼   ► And yeah, it's--

00:23:55 ◼   ► - Although, I mean, there's the whole thing

00:23:56 ◼   ► cranking the price up on the old one and then just refunding anyone who

00:23:59 ◼   ► accidentally buys it and eating 30% of that. But you can't refund them. Apple has to

00:24:03 ◼   ► refund. That's the thing. Like you as the developer, you can't just push a refund

00:24:06 ◼   ► button when they email you and say, "Oh, sorry, I corrected that for you. You can't, you

00:24:08 ◼   ► have no control over that." It's, this is why it's a terrible idea. Developers,

00:24:12 ◼   ► please do not do this. It is, you will regret it. Your customers will be upset.

00:24:16 ◼   ► It is, it is just a very, very bad idea. I wonder if the people who are most

00:24:21 ◼   ► likely to do this are the very few people on the App Store who still have

00:24:25 ◼   ► expensive apps like over 50 bucks and they want to offer upgrade pricing because like

00:24:29 ◼   ► you said upgrade pricing is pointless when you're selling $3 apps right but if it's like

00:24:34 ◼   ► $50 $100 are there any of those left I don't even know used to be that Apple was close

00:24:38 ◼   ► to that range but now not really anymore these days although was aperture like 80 or something

00:24:43 ◼   ► yep anyway Apple stop playing the game the Apple service home I'm playing it so I wonder

00:24:46 ◼   ► if there's someone out there with like some specialty CAD application that wants to offer

00:24:50 ◼   ► upgrade pricing and tries this and then we can wait for their blog post where they cry

00:24:54 ◼   ► about their customers getting angry at them or people being confused.

00:24:58 ◼   ► Yeah, I could see, you know, obviously the higher the priced app, the more sense it might

00:25:03 ◼   ► make to try a scheme like this, but I just think it's a terrible idea from the start.

00:25:08 ◼   ► And speaking of looking at web pages at developer.apple.com and reading about them on the air, there was

00:25:14 ◼   ► a lot of confusion during the week about what the hell was the NDA situation for WWDC, because

00:25:20 ◼   ► all of us were at the beginning of every session they had the little announcer person come on and say

00:25:25 ◼   ► reminder everything under the in the session is NDA and blah blah blah and like you know so

00:25:30 ◼   ► it the boilerplate text and speech that we encountered at wwc was the same as every year

00:25:34 ◼   ► which is basically the keynote is public it's streamed to the world all the sessions are under

00:25:38 ◼   ► NDA and they are but the content of that NDA according to various blog posts included this

00:25:43 ◼   ► new clause that said oh by the way you're allowed to talk about any technical information you see

00:25:47 ◼   ► in these things you just can't like post screenshots copy and paste text from slides

00:25:51 ◼   ► or distribute the software or review products or something to that effect but i'm not a lawyer i

00:25:57 ◼   ► didn't know what that text meant so we've all been very cautious about it but here i think is the

00:26:00 ◼   ► kicker after coming home from wwc i think we all realize that anybody and in any web browser can

00:26:08 ◼   ► go to apple's developer website without having an account of any kind without signing anybody

00:26:12 ◼   ► agreeing to anything and simply play videos from WWDC. You can watch videos of all the sessions of

00:26:18 ◼   ► WWDC this year for free without signing any agreement with Apple. Which means basically

00:26:23 ◼   ► that anything contained in those videos we're allowed to talk about because anybody can see

00:26:27 ◼   ► them because it's essentially public information. I know enough about Apple's NDAs to know that if

00:26:30 ◼   ► it's public information you can talk about it. So that is a relief in one respect and that we don't

00:26:35 ◼   ► have to tiptoe around this stuff because if it was in a WWDC session we can talk about it because you

00:26:39 ◼   ► You can just go watch the video right now and on the other hand what it means is that?

00:26:42 ◼   ► anybody who cares can sit there watch every video from WBC and then we'll have no reason to read my os10 review because

00:26:49 ◼   ► Everything everything in those videos you will know more than you're going to learn from my review because all I do is summarize

00:26:55 ◼   ► WBC but no one wants to watch all those videos

00:26:58 ◼   ► So you're probably read it anyway because as long as my review is it is still way shorter than watching everybody from WBC

00:27:03 ◼   ► Yeah, I think you'll be alright

00:27:05 ◼   ► Anyway, our first sponsor this week is a new sponsor.

00:27:10 ◼   ► It is Automattic.

00:27:12 ◼   ► This is a relatively new product that actually got the domain name, Automattic.com.

00:27:16 ◼   ► Even spelled the normal way, not like the WordPress way.

00:27:18 ◼   ► It's spelled the actual way that you spell the word "Automatic."

00:27:22 ◼   ► Automattic is your smart driving assistant for your smartphone.

00:27:26 ◼   ► So if you go to Automattic.com, you can see this thing.

00:27:29 ◼   ► We talked about this in neutral, in I think the second to last episode, or the third to

00:27:32 ◼   ► last episode of neutral.

00:27:33 ◼   ► We actually talked about this, and we are probably irresponsible enough drivers, well

00:27:38 ◼   ► Casey and I are at least, that this would have probably some bad news for us.

00:27:43 ◼   ► Because here's what it does.

00:27:44 ◼   ► So this plugs in, they have this little, I think it's a Bluetooth dongle, but they

00:27:48 ◼   ► have a little dongle that plugs into the OBD, which one's the wrapper, which one's the

00:27:53 ◼   ► port?

00:27:54 ◼   ► Oh, on board data, OBD-C, right?

00:27:57 ◼   ► OBD-2.

00:27:58 ◼   ► Oh yeah, sorry, yes, yes, yes.

00:28:01 ◼   ► It's that port in your car that it's usually in the driver's foot well and it's what they

00:28:06 ◼   ► plug the diagnostic thing into when you go get service.

00:28:10 ◼   ► They have a little dongle that plugs into that and then that can connect to an app on

00:28:12 ◼   ► your phone that they have.

00:28:15 ◼   ► This diagnostic port in almost any modern car can dump out tons of useful information

00:28:19 ◼   ► from the engine, like what kind of gas mileage you're getting, what the engine is doing,

00:28:23 ◼   ► what kind of condition the engine is in, if there's any problems with the engine.

00:28:27 ◼   ► When the check engine light comes on, what that really means is there is an error.

00:28:32 ◼   ► It's just a Boolean.

00:28:33 ◼   ► Like, there is something in your car reporting an error code.

00:28:37 ◼   ► This thing can actually tell you what those codes mean.

00:28:40 ◼   ► It can decode them from almost all popular car types.

00:28:42 ◼   ► You can go on their website and learn more about which ones are supported and everything,

00:28:45 ◼   ► but it's great.

00:28:46 ◼   ► So then it can even tell you how you've been doing on your gas mileage.

00:28:50 ◼   ► You can see all these pretty graphs and averages, and you can kind of compete with yourself.

00:28:56 ◼   ► You can see how you've been doing,

00:28:57 ◼   ► how much fuel is costing around your area,

00:28:59 ◼   ► and how much your driving style is helping

00:29:01 ◼   ► or hurting your fuel cost goals.

00:29:03 ◼   ► It can also remember where you parked.

00:29:06 ◼   ► And you can locate your car on a nice little map,

00:29:08 ◼   ► and look at their app.

00:29:09 ◼   ► It's all very Apple design-y.

00:29:12 ◼   ► It's a very well-made app.

00:29:14 ◼   ► Right now it is iPhone only.

00:29:15 ◼   ► It is coming soon to Android.

00:29:18 ◼   ► Oh, one more little benefit,

00:29:19 ◼   ► well, big benefit potentially.

00:29:22 ◼   ► If you are in a crash, in a serious crash,

00:29:25 ◼   ► They can actually automatically detect that from the OBD port.

00:29:31 ◼   ► And they can notify local authorities automatically

00:29:35 ◼   ► if you're in a crash, so this thing could really help you out.

00:29:37 ◼   ► Hopefully you won't need that feature, but if you do,

00:29:40 ◼   ► you might be glad you have it.

00:29:41 ◼   ► So anyway, iPhone only coming soon to Android.

00:29:43 ◼   ► This ships in only two business days.

00:29:45 ◼   ► And there's a 45-day return policy.

00:29:47 ◼   ► So just get it.

00:29:48 ◼   ► Try it out.

00:29:48 ◼   ► See if you like it.

00:29:49 ◼   ► I bet you will.

00:29:50 ◼   ► There's free shipping.

00:29:51 ◼   ► Anyway, go to automatic.com/ATP.

00:29:55 ◼   ► Normally, it's $99.95.

00:29:57 ◼   ► And there's, by the way, no subscription fees,

00:29:58 ◼   ► nothing per month for that service.

00:30:00 ◼   ► You just buy the thing and that's it.

00:30:02 ◼   ► No subscription fees.

00:30:03 ◼   ► So $100 normally, you can get 20% off.

00:30:06 ◼   ► Get this thing for just $80 at automatic.com/ATP.

00:30:10 ◼   ► Thanks a lot to Automatic for sponsoring our show.

00:30:13 ◼   ► When we talked about this on "Neutral," it was sort of

00:30:15 ◼   ► in the context of car manufacturers should be ashamed

00:30:19 ◼   ► of themselves, that they're not already doing this.

00:30:21 ◼   ► right, but now here we are a year or two later, whatever it's been and

00:30:25 ◼   ► That's the new status quo with Apple doing carplay

00:30:29 ◼   ► It's like give up on car makers trying to do something Apple's just like look just let us take over your screen

00:30:34 ◼   ► We will project a UI onto it from our device and it's like sort of

00:30:38 ◼   ► externalizing the

00:30:39 ◼   ► Technology part into a part that you upgrade faster than you upgrade your car and never relying on the kind of manufacturers do anything

00:30:45 ◼   ► So it's like, you know with automatic there's this port it has information car manufacturers have been really slow

00:30:51 ◼   ► to put that information to use, especially in non-high-end cars, but they all have this

00:30:57 ◼   ► port, so why don't we sell this thing that people can buy, stick in there and use in

00:31:00 ◼   ► conjunction with their smartphone or whatever to sort of add technology to their otherwise

00:31:06 ◼   ► inexpensive and woefully under-technologicalified, that's not a word, but sounds good, car.

00:31:13 ◼   ► Now it is.

00:31:14 ◼   ► So, one more thing on the WWC NDA stuff. As people have noted, if you go to the video

00:31:20 ◼   ► page which we will put in the show notes for anyone who wants to watch WWDC videos.

00:31:24 ◼   ► The ones that they pretty much never release on video in my recollection are the lunchtime

00:31:29 ◼   ► sessions, which are not technical sessions presented by Apple, but Apple invites in guest

00:31:33 ◼   ► speakers from various places.

00:31:36 ◼   ► A couple years ago Pixar did it, J.J. Abrams was there one year.

00:31:39 ◼   ► This year the lunch session that I went to was presented by someone who works for Lucasfilm

00:31:44 ◼   ► on the Star Wars franchise, and he was talking about the new animated show that he's working

00:31:49 ◼   ► on but kind of did a big tour of the Star Wars universe and his history with it, those

00:31:56 ◼   ► you're not going to be able to find.

00:31:58 ◼   ► And it's not because I don't think there's any NDA type things, it's just that I think

00:32:01 ◼   ► their deal with the people who want to come in and speak is, "We're not going to film

00:32:04 ◼   ► you, we're not going to release your film, just come and talk to the people who are attending

00:32:07 ◼   ► our conference as a nice thing to do."

00:32:09 ◼   ► And that's probably an easier sell than, "Hey, famous person, come here and we'll film you,

00:32:13 ◼   ► and then we'll sell access to your video or release it for free."

00:32:16 ◼   ► So you won't find the Star Wars session, which is the one most nerds are interested in this

00:32:20 ◼   ► year.

00:32:21 ◼   ► I was there, and it was good, and it was fun, and it's a shame you can't see it, but oh

00:32:25 ◼   ► well.

00:32:26 ◼   ► All right, so we have a mountain more follow-up to get through.

00:32:29 ◼   ► This might be a hypercritical-style 80% follow-up episode.

00:32:32 ◼   ► Geez.

00:32:33 ◼   ► Well, I was going to say we could just do this next item and then leave the rev, because

00:32:37 ◼   ► there's one item on cloud and then two items on medical stuff that I think maybe we mentioned

00:32:42 ◼   ► in the previous episode.

00:32:43 ◼   ► I just wanted to read this Apple and cloud feedback because it was the first time that

00:32:47 ◼   ► I had heard this information, although I think one of you two had alluded to it before.

00:32:52 ◼   ► This is from anonymous source, who knows if it's real, whatever, anyway.

00:32:58 ◼   ► Typical caveats about feedback, do not take this as gospel, but here's what this person

00:33:02 ◼   ► said in regards to Apple and the cloud.

00:33:05 ◼   ► They said, "iTunes, iCloud, and the majority of server-based interfacing are handled by

00:33:09 ◼   ► a huge division of Apple that acts almost like its own company due to sheer size.

00:33:13 ◼   ► They employ mostly contractors, including offshore.

00:33:15 ◼   ► From my experience, they were the slowest responding division by far and seemed to be

00:33:19 ◼   ► completely locked in politics."

00:33:21 ◼   ► This is like the scenario that we imagined thinking about Apple's cloud stuff.

00:33:26 ◼   ► It seems like the people who do that aren't as good as the people who do the other stuff.

00:33:30 ◼   ► Wait, hold on.

00:33:31 ◼   ► I didn't say they weren't as good.

00:33:32 ◼   ► I said they seem like they're not getting the resources or the priority that they need

00:33:36 ◼   ► to be good.

00:33:37 ◼   ► Well, so here's the next bit of his.

00:33:39 ◼   ► software developer outside the division who needed to work with them in parallel to build something,

00:33:43 ◼   ► it's a nightmare. Imagine submitting a request to the NSA to give you information and receiving

00:33:47 ◼   ► a piece of paper saying they can't due to technical reasons six months later.

00:33:51 ◼   ► It's not just that they're under-resourced, it's that they seem to be not as responsive.

00:33:58 ◼   ► They're not working on the same team. Within Apple, you imagine, "All right, we've got to

00:34:02 ◼   ► get the new iPhone out, and the OS needs to be done, and whatever frameworks for these new

00:34:06 ◼   ► new features need to be on there,

00:34:08 ◼   ► and the hardware design,

00:34:09 ◼   ► like it's all, everyone's working together

00:34:11 ◼   ► to just get this job done.

00:34:12 ◼   ► And then there's this other entity far away,

00:34:14 ◼   ► disconnected with a huge latency,

00:34:16 ◼   ► where every time you have to deal with them,

00:34:17 ◼   ► it's like a big turnaround time,

00:34:19 ◼   ► and it seems like they're not on the same page as you.

00:34:20 ◼   ► Is it just because they're under resources?

00:34:22 ◼   ► Is it just because they're remote?

00:34:23 ◼   ► I don't know, but this, you could,

00:34:25 ◼   ► if this is true, this would explain

00:34:27 ◼   ► whether Cloud stuff has such problems,

00:34:28 ◼   ► because it's almost like it's not Apple doing it.

00:34:31 ◼   ► It's like this other entity that's not,

00:34:34 ◼   ► that's not running the same race with them.

00:34:35 ◼   ► It's not sort of together on producing this big thing,

00:34:38 ◼   ► but outsourcing, as the Spolsky thinks,

00:34:42 ◼   ► don't outsource your core competency.

00:34:43 ◼   ► If cloud is an Apple's core competency at this point,

00:34:46 ◼   ► then what is?

00:34:47 ◼   ► Like, they need to bring that back in

00:34:49 ◼   ► if it really is outsourced like this,

00:34:50 ◼   ► 'cause if this is true,

00:34:52 ◼   ► it is very depressing and a bad situation.

00:34:54 ◼   ► - It's almost as if they need increased collaboration.

00:34:58 ◼   ► - Now, that was just increased collaboration

00:34:59 ◼   ► amongst the hardware and software guys

00:35:01 ◼   ► to stop them from fighting.

00:35:02 ◼   ► Those are the people who already were working

00:35:03 ◼   ► pretty well together.

00:35:04 ◼   ► this is, who knows what this is.

00:35:05 ◼   ► So again, I don't know if any of this is true.

00:35:08 ◼   ► Anonymous sources, who knows, but if it is true,

00:35:11 ◼   ► it confirms my worst fears and presumably it is changing.

00:35:15 ◼   ► Like the CloudKit stuff would indicate a change

00:35:17 ◼   ► in that direction because it sure seems like, you know,

00:35:20 ◼   ► that Apple is dogfooding its own stuff

00:35:22 ◼   ► and to a degree that they hadn't before

00:35:25 ◼   ► and that CloudKit looks like a much more sane API

00:35:29 ◼   ► than the previous ones, much more like the APIs

00:35:31 ◼   ► that everyone was building for themselves.

00:35:33 ◼   ► So things are looking up in this area.

00:35:35 ◼   ► This is a sort of a look backward

00:35:37 ◼   ► on how things might have been if this is all accurate.

00:35:39 ◼   ► - Yeah, and that does corroborate what I had heard,

00:35:43 ◼   ► although that's more detailed than I had heard,

00:35:45 ◼   ► but also I heard from some, so we had somebody else,

00:35:48 ◼   ► I don't know if it was the same person or not,

00:35:49 ◼   ► or maybe it was a random conversation I had somewhere,

00:35:51 ◼   ► but somebody else said that this actually was

00:35:54 ◼   ► like actively changing and so like Federighi

00:35:58 ◼   ► was now able to like kind of push against that

00:36:01 ◼   ► to some degree and get stuff moving

00:36:04 ◼   ► and that apparently changes were happening.

00:36:05 ◼   ► And that's all that we know really,

00:36:08 ◼   ► that we don't even really know that.

00:36:10 ◼   ► But it basically sounds like it has been bad

00:36:15 ◼   ► and it is still a little bit bad but is making progress.

00:36:19 ◼   ► And I think what Apple is showing, what they're announcing,

00:36:22 ◼   ► and even just how their stuff is performing recently,

00:36:25 ◼   ► I think the results seem to be bearing that out.

00:36:29 ◼   ► That seems like a plausible explanation

00:36:31 ◼   ► of what we're seeing.

00:36:32 ◼   ► Anyway, before we move on to the official topics,

00:36:35 ◼   ► let me get one more sponsorship out of the way,

00:36:36 ◼   ► because we were a little late to the first one.

00:36:39 ◼   ► Our second sponsor this week is lynda.com, L-Y-N-D-A dot com.

00:36:44 ◼   ► lynda.com helps you learn and keep up to date

00:36:47 ◼   ► with your software, pick up your brand new skills,

00:36:49 ◼   ► or explore new hobbies with easy to follow,

00:36:51 ◼   ► excellently produced, professional video tutorials.

00:36:55 ◼   ► Now, you can learn a new programming language,

00:36:57 ◼   ► you can create UI, you can get design tips,

00:37:00 ◼   ► You can get your first code up and running

00:37:02 ◼   ► with Objective-C or probably even Swift pretty soon.

00:37:05 ◼   ► I imagine they're working on that very quickly.

00:37:06 ◼   ► They're usually very quick to put up new stuff

00:37:08 ◼   ► when new stuff comes out.

00:37:09 ◼   ► You can even use lynda.com to learn new applications

00:37:12 ◼   ► for new creative endeavors, like if you wanna learn

00:37:15 ◼   ► Photoshop or Illustrator or Logic or anything like that.

00:37:19 ◼   ► You can learn video editing, post-production,

00:37:21 ◼   ► all this crazy stuff.

00:37:22 ◼   ► You can learn all that stuff from beginner to advanced.

00:37:24 ◼   ► They have over 2,400 courses.

00:37:26 ◼   ► They're taught by industry experts

00:37:28 ◼   ► and they add more courses every week.

00:37:30 ◼   ► You can always go,

00:37:31 ◼   ► as I said, they're always up to date with new stuff.

00:37:34 ◼   ► They work directly with software companies,

00:37:35 ◼   ► so that way, when a new version of Photoshop comes out,

00:37:38 ◼   ► they can have a tutorial on day one.

00:37:40 ◼   ► You can get started right away.

00:37:42 ◼   ► Really fantastic.

00:37:43 ◼   ► Now, lynda.com, this is the best part, I think.

00:37:46 ◼   ► You get all this,

00:37:47 ◼   ► you can have access to their entire library.

00:37:49 ◼   ► It isn't per video.

00:37:51 ◼   ► You pay $25 per month, and that's a flat fee.

00:37:55 ◼   ► You get unlimited access to everything

00:37:57 ◼   ► for 25 bucks a month.

00:37:58 ◼   ► So that kind of removes the pressure in your head.

00:38:01 ◼   ► You're like, well, do I really wanna learn about this thing

00:38:04 ◼   ► for another $3?

00:38:06 ◼   ► You don't have to worry about that.

00:38:07 ◼   ► 25 bucks a month gets you unlimited access

00:38:09 ◼   ► to their entire catalog.

00:38:11 ◼   ► So browse around.

00:38:11 ◼   ► You might find something that you didn't think

00:38:14 ◼   ► you'd want to be interested in,

00:38:15 ◼   ► but watch a few minutes of it and see.

00:38:17 ◼   ► And then you might be like,

00:38:18 ◼   ► oh, I always assumed that would be harder.

00:38:20 ◼   ► Or, oh, that's how you can do this cool thing.

00:38:22 ◼   ► I never knew that.

00:38:23 ◼   ► So anyway, go to lynda.com, L-Y-N-D-A.com.

00:38:27 ◼   ► slash ATP, and you can get a free seven day trial.

00:38:32 ◼   ► Go check it out, see for yourself.

00:38:34 ◼   ► We've all, all three of us have watched stuff on there.

00:38:36 ◼   ► I've learned a lot from their stuff,

00:38:38 ◼   ► especially with some of the podcast production stuff

00:38:39 ◼   ► that I've done, and chances are I'll probably go on there

00:38:42 ◼   ► to learn Swift or JSON or, sorry, Node or all sorts of stuff.

00:38:45 ◼   ► They have all sorts of language courses.

00:38:47 ◼   ► You can even learn PHP if you're the only person

00:38:49 ◼   ► on Earth who doesn't know it yet, John,

00:38:52 ◼   ► and you can learn all sorts of new stuff there.

00:38:55 ◼   ► 3D printing, they even have 3D printing.

00:38:57 ◼   ► I mean, they have this crazy list of stuff.

00:38:59 ◼   ► Go check it out.

00:39:00 ◼   ► These are nicely produced professional video tutorials.

00:39:03 ◼   ► Go to L-Y-N-D-A dot com slash ATP.

00:39:07 ◼   ► That is Linda dot com with a Y.

00:39:09 ◼   ► Linda dot com slash ATP.

00:39:11 ◼   ► Thanks a lot to Linda for sponsoring our show once again.

00:39:14 ◼   ► - Pretty sure I've written more PHP than Casey.

00:39:16 ◼   ► - That's possible but not necessarily true.

00:39:20 ◼   ► When Aaron and I got engaged,

00:39:24 ◼   ► I had an old ThinkPad running some flavor of Ubuntu and I wrote our wedding website

00:39:34 ◼   ► in PHP and this was actually my first exposure to databases because I had always been writing

00:39:39 ◼   ► client-side apps.

00:39:41 ◼   ► And so I wrote a wedding website which allowed you to RSVP online and I was extremely proud

00:39:48 ◼   ► of myself.

00:39:49 ◼   ► That was all PHP.

00:39:50 ◼   ► And I thought that went really well until my friend who had an apostrophe in his surname

00:39:57 ◼   ► went to register and all of a sudden everything took a turn because I didn't properly escape

00:40:02 ◼   ► everything.

00:40:03 ◼   ► But like I said, it was my first experience and that went pretty well.

00:40:06 ◼   ► And we didn't have to use the "not" in order to do that because, I don't know,

00:40:11 ◼   ► everyone used the "not" at that point and I didn't want to.

00:40:14 ◼   ► So.

00:40:15 ◼   ► JAYLEE STIFFER You really wanted to show Erin right up front

00:40:17 ◼   ► what she was getting into.

00:40:18 ◼   ► Basically, yes.

00:40:20 ◼   ► - Yeah.

00:40:20 ◼   ► - And she's still here.

00:40:21 ◼   ► - That's good, yeah.

00:40:23 ◼   ► You found the right one.

00:40:24 ◼   ► - So anyway, what are we talking about tonight?

00:40:27 ◼   ► - Talking about Marco's Mac Pro woes.

00:40:29 ◼   ► - Oh yes, how could I forget?

00:40:31 ◼   ► So--

00:40:32 ◼   ► - That was fun, yeah.

00:40:33 ◼   ► - So what happened?

00:40:34 ◼   ► - I woke up to a computer that was showing

00:40:37 ◼   ► signs of disk failure.

00:40:39 ◼   ► Basically it had totally locked up.

00:40:42 ◼   ► Like you could still move the mouse and everything.

00:40:43 ◼   ► You could still like launch bar.

00:40:45 ◼   ► You could still invoke commands and click on stuff

00:40:48 ◼   ► and move stuff, but anything that would cause a disk access

00:40:51 ◼   ► would just freeze whatever you were using.

00:40:54 ◼   ► And so I tried, like I couldn't even launch console,

00:40:58 ◼   ► like I tried to launch a console app

00:41:00 ◼   ► to see if it was showing me anything useful,

00:41:02 ◼   ► and I couldn't even do that

00:41:02 ◼   ► because that would require reading from the disk.

00:41:05 ◼   ► So I thought, oh crap, my SSD is dead.

00:41:08 ◼   ► And I rebooted and rebooted and rebooted

00:41:13 ◼   ► and could never get past the gray spinner.

00:41:16 ◼   ► in verbose mode it would stop at the,

00:41:19 ◼   ► how do you pronounce the FSCK, Disc Command Checking thing?

00:41:22 ◼   ► - Just spell it out, it's fine.

00:41:23 ◼   ► - Anyway, it wouldn't get,

00:41:25 ◼   ► it was doing that and would never finish it.

00:41:27 ◼   ► I gave it lots of time, I gave it an hour to see,

00:41:30 ◼   ► like is it taking a while to finish?

00:41:32 ◼   ► No, it wasn't, it was just stuck.

00:41:35 ◼   ► And you know, I restored it from a super duper clone,

00:41:39 ◼   ► which is by far the fastest recovery method.

00:41:42 ◼   ► And so I did that and it was fine,

00:41:45 ◼   ► and so I'm fine, I'm back up now and everything's fine.

00:41:48 ◼   ► It seems, oh, and for whatever it's worth,

00:41:52 ◼   ► while it was in the unbootable state,

00:41:55 ◼   ► I booted from the clone and ran Disk Utility

00:41:58 ◼   ► against the damaged disk.

00:42:00 ◼   ► It appeared to have no errors.

00:42:02 ◼   ► I did the whole repair, verify everything.

00:42:04 ◼   ► It reported nothing wrong.

00:42:07 ◼   ► In conclusion, it appears as though

00:42:09 ◼   ► this was almost certainly a file system related problem.

00:42:14 ◼   ► You know, I don't know.

00:42:15 ◼   ► Anyway, this is boring, but I had an issue that caused

00:42:18 ◼   ► enough disk corruption to make the disk unbootable,

00:42:21 ◼   ► and it seems to have been related to the file system

00:42:25 ◼   ► I was using, which was HFS+.

00:42:28 ◼   ► - Ding! - Yep.

00:42:29 ◼   ► That's it.

00:42:31 ◼   ► - We're doing the verbal sound effects now.

00:42:33 ◼   ► - Yeah, well, I don't have one of those crazy sound boards,

00:42:36 ◼   ► like on Bionic, where I can just hit a button.

00:42:38 ◼   ► (laughing)

00:42:39 ◼   ► I love Bionic.

00:42:40 ◼   ► Bionic is--

00:42:42 ◼   ► - Oh, it's so good, but I couldn't even tell you why.

00:42:45 ◼   ► It's just so good.

00:42:46 ◼   ► - It's so bad that it's so good, and they know it.

00:42:49 ◼   ► That's the, like, it's the show that,

00:42:52 ◼   ► it's like, it's like Milton, the guy from Office Space,

00:42:56 ◼   ► like, you know, it's like, you know,

00:42:57 ◼   ► the show got fired six years ago and no one told them,

00:43:01 ◼   ► and they just keep going in every day.

00:43:03 ◼   ► It's just this ongoing piece of performance art.

00:43:06 ◼   ► (laughing)

00:43:07 ◼   ► Between two people who are very funny together

00:43:11 ◼   ► and just have nothing to talk about

00:43:13 ◼   ► because the show is like totally off the rails.

00:43:16 ◼   ► - Oh, it's so wonderful.

00:43:17 ◼   ► - And it is comical, like it is such good humor

00:43:21 ◼   ► and the funniest part is you're sitting there listening

00:43:24 ◼   ► and thinking, why am I still listening to this?

00:43:26 ◼   ► And they're sitting there producing it saying,

00:43:28 ◼   ► why are we still producing this?

00:43:29 ◼   ► And yet it works, I don't know why it works,

00:43:32 ◼   ► but it does work and it's hilarious and I really enjoy it.

00:43:35 ◼   ► - Oh, I agree.

00:43:36 ◼   ► - I hope they didn't pay for that spot

00:43:37 ◼   ► because I'm not sure you're really selling it there.

00:43:39 ◼   ► (laughing)

00:43:40 ◼   ► "Why am I still listening to this?" says Marco Armond.

00:43:43 ◼   ► - No, nobody should be listening to it,

00:43:45 ◼   ► but I listen to it and I love it.

00:43:46 ◼   ► - Oh, I completely agree with everything you just said.

00:43:48 ◼   ► It's so good.

00:43:49 ◼   ► - And I think they would agree with me.

00:43:50 ◼   ► I think they would say don't listen to it.

00:43:52 ◼   ► - Oh, it's so wonderful because it really has got,

00:43:54 ◼   ► it's like the entire show,

00:43:55 ◼   ► like if you think "Back to Work" is a show of in-jokes,

00:43:58 ◼   ► you have no idea because Bionic is one entire show

00:44:02 ◼   ► made of nothing but in-jokes.

00:44:04 ◼   ► It's ridiculous and it's wonderful in every possible way.

00:44:07 ◼   ► - Yeah, it is really quite good.

00:44:10 ◼   ► Anyway, I forgot we were even talking about that.

00:44:15 ◼   ► How did we even get there?

00:44:16 ◼   ► - Ding soundboard, HFS Plus.

00:44:18 ◼   ► - Nice, right, okay.

00:44:19 ◼   ► - There you go.

00:44:20 ◼   ► - Anyway.

00:44:21 ◼   ► - So I put the link of the show notes to this article

00:44:23 ◼   ► that a bazillion people have sent me on Twitter

00:44:25 ◼   ► from this person complaining about HFS Plus

00:44:27 ◼   ► because they had a bunch of photos

00:44:30 ◼   ► and it was something like 15,000 photos

00:44:33 ◼   ► stored over six years and he ran some checksums to compare.

00:44:37 ◼   ► I guess he had valid checksums someplace

00:44:39 ◼   ► compared it to the the file store in hms plus and he lost 28 files over six years and these are images

00:44:45 ◼   ► and some of them some of these like jpegs would half load some of them would load completely again

00:44:49 ◼   ► this is what we're talking about uh bit rot you know that digital storage is not forever uh errors

00:44:56 ◼   ► will be introduced and if they're introduced in your images uh you may lose some or all of

00:45:01 ◼   ► individual pictures like if the error is towards the end maybe you'll be able to see most of the

00:45:05 ◼   ► picture for the errors in the middle you go see the top half of the picture maybe there's some

00:45:08 ◼   ► clever thing they could recover from those. Maybe the whole file is host and you just get a bunch of

00:45:11 ◼   ► static or garbage. But anyway, that's a shame if that's the only copy of the picture you had.

00:45:15 ◼   ► And this article talks about HFS+ and how it's old and how it's crappy and so on and so forth.

00:45:19 ◼   ► And a lot of people sent me this thing going, "See, you're always talking about the HFS+ right?

00:45:22 ◼   ► Look at this poor guy. He's backing you up saying HFS+ is crappy." And I didn't respond to most of

00:45:29 ◼   ► these people and I didn't really tweet out the link to the article because this article is not

00:45:33 ◼   ► really about HFS+ being crappy. I already wrote a bit about HFS+ being crappy in one of my reviews,

00:45:38 ◼   ► that is linked to in this article at the end in an addendum.

00:45:43 ◼   ► HFS+ is crappy, and it does corrupt itself.

00:45:46 ◼   ► And I'm not even sure that's what happened to Marco's thing, because

00:45:49 ◼   ► Disk Utility said everything was okay, so HFS+ probably didn't

00:45:52 ◼   ► corrupt its metadata structure. It probably does think it's keeping track of everything.

00:45:57 ◼   ► But bitrot is something that is sort of a separate issue, where

00:46:02 ◼   ► when data goes bad on disk, the metadata keeping track of where that data is

00:46:07 ◼   ► could be perfectly valid. Like, HFS+ has not corrupted itself, it knows where all of your

00:46:11 ◼   ► data is for every single file, it's just that that data itself is garbage. And HFS+ does

00:46:16 ◼   ► not care what the content of that data is, it just needs to keep track of it and remember

00:46:20 ◼   ► where it is and how much is available free and how much is occupied and where each piece

00:46:25 ◼   ► of individual files are and what order they go in. And HFS+ does fail in many respects

00:46:31 ◼   ► in doing that job, and that is one of my complaints about HFS+. But the second complaint about

00:46:35 ◼   ► and all sort of file systems that predate the ZFS dawning of reliability, of getting

00:46:43 ◼   ► religion about reliability, they don't care what your data is either.

00:46:46 ◼   ► They just cross their fingers and hope that the disk storage system is reliable.

00:46:49 ◼   ► And ZFS was the first popular file system to get serious about saying that we're going

00:46:56 ◼   ► to take responsibility for keeping track of whether the stuff you wrote to disk is what

00:47:02 ◼   ► is still there.

00:47:03 ◼   ► do that with checksums, not just checksums on the metadata,

00:47:05 ◼   ► but also checksums on the data itself,

00:47:07 ◼   ► which is computationally expensive and takes memory

00:47:09 ◼   ► and has all these other limits.

00:47:11 ◼   ► But ZFS would be able to detect errors

00:47:12 ◼   ► in the storage system.

00:47:13 ◼   ► Sometimes there are errors in the hard drive,

00:47:15 ◼   ► sometimes there are errors in the disk controller,

00:47:16 ◼   ► sometimes there are errors in the driver stack

00:47:19 ◼   ► and in the operating system.

00:47:21 ◼   ► No matter where the source of the error is,

00:47:22 ◼   ► ZFS's thing was like end-to-end data protection,

00:47:25 ◼   ► where if we write data and then later on go to read it,

00:47:29 ◼   ► and if what we read is not what we wrote originally,

00:47:31 ◼   ► we will tell you that there's a problem.

00:47:34 ◼   ► That doesn't save your data, it's still garbage,

00:47:36 ◼   ► but then you can, if you have the ability

00:47:37 ◼   ► to detect when something went wrong,

00:47:39 ◼   ► if you're using any form of redundant storage,

00:47:41 ◼   ► which the EFS offered itself,

00:47:42 ◼   ► where it could store your data twice on disk

00:47:44 ◼   ► or three times on disk, you could use a multi-disk scenario,

00:47:47 ◼   ► you could have backups, so on and so forth,

00:47:49 ◼   ► the key is having the file system let you know immediately,

00:47:51 ◼   ► oh, this data is bad,

00:47:54 ◼   ► but I know I have another copy of this data over there,

00:47:56 ◼   ► and I can tell if that data is good,

00:47:58 ◼   ► so I don't just have to guess,

00:47:59 ◼   ► like, oh, this one is bad, that other one must be good.

00:48:01 ◼   ► I can also check that one.

00:48:02 ◼   ► Oh, that data is good.

00:48:02 ◼   ► Let me write it back on top of this one.

00:48:04 ◼   ► Let me, you know, it could, you know,

00:48:06 ◼   ► sort of the self healing file system.

00:48:08 ◼   ► Not knowing whether your data is bad,

00:48:10 ◼   ► like this person manually made a bunch of checksums

00:48:11 ◼   ► or whatever, and then compared them.

00:48:13 ◼   ► If you don't know your data is bad,

00:48:15 ◼   ► you won't find out that your data is hosed

00:48:17 ◼   ► until that corrupt data has like spread

00:48:19 ◼   ► to all of your backups.

00:48:20 ◼   ► It's been pushed to your cloud backup.

00:48:22 ◼   ► It's all, it's, you know,

00:48:23 ◼   ► your last three months worth of backups

00:48:24 ◼   ► have long since been overwritten.

00:48:25 ◼   ► And all you have is the corrupt data

00:48:27 ◼   ► because you don't have your data going back

00:48:29 ◼   ► to the beginning of time.

00:48:30 ◼   ► have is multiple copies of your corrupt data. And you won't know that again until like you're

00:48:34 ◼   ► trying to get out pictures for someone's high school graduation and you want a baby picture

00:48:37 ◼   ► and you realize the baby picture you wanted is corrupt. If you had a file system that

00:48:43 ◼   ► had end-to-end data integrity, the file system could have notified you of that immediately

00:48:48 ◼   ► when it happened and could have potentially repaired it from a good copy. So that's I

00:48:53 ◼   ► think what this article is really complaining about, that all file systems are not like

00:48:58 ◼   ► ZFS and offer data integrity guarantees and not so much that oh yeah HFS is

00:49:03 ◼   ► crappy because it loses track of where your data is which does happen

00:49:06 ◼   ► and people do tweet about that all the time but that's not really what this

00:49:09 ◼   ► article is about. We are sponsored this week by Backblaze. Our friends at

00:49:14 ◼   ► Backblaze. It is unlimited, unthrottled, uncomplicated, available anywhere. You can

00:49:19 ◼   ► try it for free with no credit card required so go to backblaze.com/ATP.

00:49:24 ◼   ► This is awesome online backup.

00:49:27 ◼   ► And believe me, this is a great time to have them

00:49:29 ◼   ► as a sponsor because when my hard drive,

00:49:32 ◼   ► when I thought my drive had died this morning,

00:49:35 ◼   ► and either way, regardless of what I thought,

00:49:37 ◼   ► I had to wipe it and start fresh,

00:49:40 ◼   ► I was never nervous that I was going to lose data

00:49:42 ◼   ► because I know that yes, I have my time machine here.

00:49:46 ◼   ► I also have my super duper clone here,

00:49:48 ◼   ► which is what gave me a very fast recovery today.

00:49:51 ◼   ► But I also knew that no matter what,

00:49:53 ◼   ► Backblaze would have my data also.

00:49:55 ◼   ► And so there's always this other place.

00:49:57 ◼   ► There's always this backup for you.

00:49:59 ◼   ► And actually, I met the two founders,

00:50:01 ◼   ► I think they were the two founders.

00:50:02 ◼   ► I met two high up people at Backblaze

00:50:04 ◼   ► for tea in San Francisco,

00:50:06 ◼   ► and they actually bought me an $11 tea,

00:50:08 ◼   ► which is very generous of them,

00:50:09 ◼   ► 'cause nothing at Semivar tea lounge or whatever is cheap.

00:50:12 ◼   ► It was quite good though.

00:50:14 ◼   ► Also quite large.

00:50:15 ◼   ► I don't know how anybody can drink a tea

00:50:16 ◼   ► that size of what they serve there.

00:50:17 ◼   ► Like, it was like a 20 ounce pitcher of tea

00:50:20 ◼   ► for one person anyway.

00:50:22 ◼   ► So these guys are really sharp.

00:50:24 ◼   ► They know their stuff.

00:50:25 ◼   ► And they're just nice people.

00:50:28 ◼   ► You can tell when you use the product.

00:50:29 ◼   ► You can tell when you look at the website.

00:50:30 ◼   ► You can tell when you read their blog

00:50:32 ◼   ► where they explain technical details of things like,

00:50:34 ◼   ► here's how we built these crazy storage pods

00:50:36 ◼   ► that have all these hard drives in them.

00:50:37 ◼   ► Here's the plans.

00:50:38 ◼   ► Here's what we do.

00:50:39 ◼   ► Here's how we get drives that are cheap

00:50:42 ◼   ► by driving all around the country

00:50:43 ◼   ► and doing crazy stuff with rebates.

00:50:44 ◼   ► Like, they're just good people and they know their stuff.

00:50:48 ◼   ► It was founded by ex-Apple engineers,

00:50:49 ◼   ► so their software on the Mac is great.

00:50:51 ◼   ► They have all this new stuff.

00:50:52 ◼   ► They have email alert notifications.

00:50:54 ◼   ► You can get like a report every couple of weeks saying,

00:50:58 ◼   ► hey, this is what we have.

00:50:59 ◼   ► Just so you know, we have this much data.

00:51:00 ◼   ► It was last backed up at this time.

00:51:02 ◼   ► They'll send you an alert

00:51:03 ◼   ► if they haven't heard from your computer in a while.

00:51:04 ◼   ► So that way, if you like pause the backup

00:51:07 ◼   ► and you're recording a podcast from it

00:51:08 ◼   ► and then you forget to unpause it,

00:51:10 ◼   ► they will send you an email alert in a couple of days

00:51:12 ◼   ► saying, hey, by the way, we haven't heard from you.

00:51:13 ◼   ► This is unusual.

00:51:15 ◼   ► You should probably know about this and fix it.

00:51:17 ◼   ► It is fantastic.

00:51:18 ◼   ► They have a 15 day free trial with no credit card required.

00:51:22 ◼   ► You just enter an email and password and that's it,

00:51:24 ◼   ► you're off.

00:51:25 ◼   ► And this is really, it's $5 a month and that's it.

00:51:28 ◼   ► There's no add-ons, there's no gimmicks,

00:51:30 ◼   ► there's no tacked on charges or fees or surcharges.

00:51:33 ◼   ► $5 a month per computer for unlimited space.

00:51:36 ◼   ► And we, I mean, jeez, my wife has about

00:51:39 ◼   ► three terabytes on hers.

00:51:41 ◼   ► I have about a terabyte and a half on mine.

00:51:44 ◼   ► My mom has a whole 38 gigs on hers.

00:51:47 ◼   ► Obviously a big range here, but it doesn't matter.

00:51:49 ◼   ► Same price, five bucks a month, easy.

00:51:51 ◼   ► They don't have to charge you more if you go

00:51:52 ◼   ► to the three gig, or three terabyte limit like my wife,

00:51:55 ◼   ► it's fine.

00:51:56 ◼   ► It is the simplest online backup,

00:51:57 ◼   ► and in my experience, it's also the fastest.

00:52:00 ◼   ► Just so you know, if you have a nice fast upstream,

00:52:02 ◼   ► I've had very good experience with getting stuff

00:52:03 ◼   ► to Backblaze very quickly.

00:52:05 ◼   ► So, go to backblaze.com/atp.

00:52:09 ◼   ► If you mistype it as Blackblaze,

00:52:11 ◼   ► they actually own that domain name too.

00:52:13 ◼   ► Don't look up what it used to be.

00:52:14 ◼   ► Just go to backblaze.com/atp.

00:52:17 ◼   ► Thank you very much to Backblaze for sponsoring our show

00:52:19 ◼   ► once again and for buying me tea last week.

00:52:22 ◼   ► - Okay, so let's talk about metal

00:52:24 ◼   ► and I do not mean the flavor of music.

00:52:27 ◼   ► So I am not at all qualified to talk about

00:52:30 ◼   ► any 3D programming, anything,

00:52:31 ◼   ► because I've never really done it.

00:52:32 ◼   ► And Marco, I don't believe you have either.

00:52:35 ◼   ► - That is not true.

00:52:36 ◼   ► - Oh, really?

00:52:37 ◼   ► - I am slightly,

00:52:39 ◼   ► slightly a hair above completely unqualified.

00:52:43 ◼   ► - Excellent.

00:52:43 ◼   ► because I, in college, I tried writing a 3D version

00:52:48 ◼   ► of Scorched Earth, and I tried it in like three

00:52:51 ◼   ► different languages using at least two different

00:52:54 ◼   ► graphics APIs, using DirectX first, then OpenGL,

00:52:57 ◼   ► and then something else using OpenGL.

00:52:59 ◼   ► It was basically, it was my way of fooling around

00:53:02 ◼   ► with various 3D stuff that I was learning in college,

00:53:04 ◼   ► and then I kinda used my most recent version of that

00:53:07 ◼   ► as a demo to get myself my first job,

00:53:10 ◼   ► and then I never looked at 3D programming again.

00:53:11 ◼   ► So I know enough, I know very basics of 3D programming

00:53:16 ◼   ► as it stood in 2003 and 2004.

00:53:19 ◼   ► So it's probably completely useless by today's standards.

00:53:22 ◼   ► And I would never consider myself

00:53:24 ◼   ► a knowledgeable programmer at all.

00:53:26 ◼   ► However, I at least know like the kinds of things

00:53:29 ◼   ► that low-level graphic APIs do.

00:53:31 ◼   ► - Right, so John, why don't you tell us about M

00:53:35 ◼   ► - I don't know why you think I know any more about,

00:53:37 ◼   ► I mean, I've done about the same amount of 3D programming

00:53:39 ◼   ► Marco like demo apps on the SGIs in college, you know, where you just get something up

00:53:44 ◼   ► on the screen and maybe move it around and you're like, "Well, that was really hard.

00:53:46 ◼   ► I'm not doing this anymore," because it's super hard to get anything impressive up on

00:53:51 ◼   ► the screen, especially back then. But I mean, that—

00:53:53 ◼   ► I mean, really, if you want to know about Metal as an API, just start listening to debug,

00:53:57 ◼   ► and eventually Guy English will say enough things, because he actually is qualified to

00:54:01 ◼   ► know, and, like, because he's actually a real game programmer. So listen to what he says

00:54:06 ◼   ► about Metal, and you'll learn more about it from him than from us.

00:54:11 ◼   ► But the executive summary is what's important. I mean, we're going to talk about it from

00:54:14 ◼   ► a business perspective, but it's like, so OpenGL and OpenGL ES is what Apple was using

00:54:19 ◼   ► previously. These are open standards that are in theory not controlled by any one company.

00:54:23 ◼   ► There's a consortium, blah, blah, blah. These are very old standards that have evolved

00:54:26 ◼   ► over the years. When these standards were created, the current crop of 3D hardware,

00:54:32 ◼   ► current GPUs did not exist in this form. And so the API, even though it has evolved over

00:54:37 ◼   ► the years, it's had to maintain backward compatibility, it is not well suited to current hardware.

00:54:45 ◼   ► Not just in the particular details, because a lot of that is handled by a driver, but

00:54:48 ◼   ► just in terms of the programming model. Like, what do you do? You issue a series of function

00:54:52 ◼   ► calls that set things up and then cause them to be drawn. And the sequence of events that

00:54:58 ◼   ► presented through an OpenGL API have almost no bearing on what's actually happening with

00:55:02 ◼   ► the hardware in terms of when does something happen, when do commands get sent to the GPU,

00:55:07 ◼   ► when do the things that are going to be drawn get sent to the GPU's memory, how do things get

00:55:12 ◼   ► written back, how does that affect this? A lot of that happens in the driver, but that's just

00:55:15 ◼   ► another level of indirection, so you're there doing things step by step in your program,

00:55:20 ◼   ► and there's this whole other program going on behind the scenes to manage that stuff to say,

00:55:24 ◼   ► "All right, when should I ship this stuff off to the GPU so it's available in time?

00:55:28 ◼   ► oh wait, the program just read something back from the GPU.

00:55:30 ◼   ► Actually, we have to stall and read that back from the GPU.

00:55:33 ◼   ► So previously, the GPU was two frames ahead of the CPU,

00:55:35 ◼   ► but now the CPU wants to read that data back,

00:55:37 ◼   ► so we have to wait for it to catch up and resynchronize.

00:55:39 ◼   ► There's latency between the GPU and the CPU,

00:55:42 ◼   ► but if you don't accept that latency,

00:55:44 ◼   ► you might not be completely utilizing the GPU.

00:55:46 ◼   ► And all sorts of details that come from the mismatch

00:55:49 ◼   ► between this relatively high-level API

00:55:52 ◼   ► that does not acknowledge the existence of the GPU

00:55:55 ◼   ► in its current form,

00:55:56 ◼   ► and what's actually going on behind the scenes.

00:55:57 ◼   ► So that requires game programmers to do all sorts of voodoo stuff where they're like,

00:56:02 ◼   ► "Well, I know if I do these sequence of events and do this little thing here, it will force

00:56:06 ◼   ► these things to all be uploaded to the GPU.

00:56:08 ◼   ► And I know if I do this, my shaders will be compiled and we'll be ready for me."

00:56:13 ◼   ► That's another thing with OpenGL.

00:56:14 ◼   ► There's OpenGL shader language, and that's sort of a high-level language, but it has

00:56:18 ◼   ► to be compiled for the individual GPU.

00:56:20 ◼   ► And whether that happens at the time your program runs or at the time it draws a frame

00:56:25 ◼   ► or you want your shadies to be precompiled,

00:56:27 ◼   ► and DirectX has this method where they precompile them down to sort of a,

00:56:31 ◼   ► not bytecode, but more compact form,

00:56:33 ◼   ► and then Apple's drivers had to use LLVM for that back in the day.

00:56:36 ◼   ► Anyway, it's an extremely complicated world,

00:56:39 ◼   ► and what developers who do 3D programming wanted to do is say,

00:56:43 ◼   ► "This is way too complicated.

00:56:44 ◼   ► I can't keep track of all the different machines

00:56:46 ◼   ► and the Rube Goldberg device that causes graphics to go up on the screen."

00:56:50 ◼   ► If you just gave me a programming model

00:56:53 ◼   ► that was a closer match to the way things actually happen, it would be a lot easier

00:56:57 ◼   ► to get something that performs well on a wide range of hardware. Metal is an interesting

00:57:03 ◼   ► entry into that field because, first of all, Metal right now only works on iOS and only works on the

00:57:08 ◼   ► A7. So certainly it's not helping people deploy their, you know, write code that's going to work

00:57:15 ◼   ► on a wide variety of platforms or a wide variety of hardware. Going forward, I'm sure they will

00:57:19 ◼   ► expand support for it, I'm sure it supports the A8 now and all that other good stuff,

00:57:23 ◼   ► and who knows what it'll support in the future,

00:57:25 ◼   ► but for now it's very narrowly constrained.

00:57:27 ◼   ► But the programming model of Metal

00:57:29 ◼   ► does fit closer to the way things work.

00:57:31 ◼   ► It shows you, here's a command buffer,

00:57:34 ◼   ► batch your commands up, batch your things up,

00:57:37 ◼   ► ship them off to the GPU in the correct order.

00:57:40 ◼   ► Like we're exposing all these things

00:57:41 ◼   ► that were previously happening in the driver

00:57:43 ◼   ► that were completely opaque to you,

00:57:44 ◼   ► we expose them as objects,

00:57:45 ◼   ► and you have a programming language

00:57:47 ◼   ► that lets you construct them, package them up,

00:57:49 ◼   ► send them off, you know,

00:57:51 ◼   ► and it's much cleaner than OpenGL,

00:57:53 ◼   ► and it's a great fit for the way modern GPUs work today.

00:57:57 ◼   ► And the reason Apple thinks this is something

00:58:00 ◼   ► that's going to help people is not because they think

00:58:02 ◼   ► people are going to write their programs in Metal

00:58:04 ◼   ► and target only A7 devices or greater,

00:58:06 ◼   ► but because the people who make the game engines,

00:58:09 ◼   ► which they had all on the stage during the keynote,

00:58:11 ◼   ► that's why they had Epic up there with their Unreal engine,

00:58:13 ◼   ► and they showed Unity and all the other game engines

00:58:16 ◼   ► out there whose names I can't remember,

00:58:18 ◼   ► like the big four engine makers for 3D engines.

00:58:21 ◼   ► That's what most game makers use

00:58:23 ◼   ► because writing a 3D engine is super hard.

00:58:25 ◼   ► Well, if the engine middleware makers

00:58:30 ◼   ► make sure their engines are able to run with metal

00:58:32 ◼   ► when they're running on the A7, then anyone

00:58:35 ◼   ► who develops a game on top of that engine gets the advantage of,

00:58:38 ◼   ► oh, if you play my game on an A7, I'll get better performance.

00:58:42 ◼   ► So it's an interesting move by Apple

00:58:44 ◼   ► to sort of try to get away from OpenGL

00:58:47 ◼   ► and take control of their own destiny with a more modern API.

00:58:50 ◼   ► And it's also interesting that they didn't choose

00:58:52 ◼   ► do what Microsoft did, which is Microsoft wanted to compete with OpenGL years and years

00:58:55 ◼   ► ago and they made DirectX, which learned from OpenGL's mistakes and was more modern, but

00:59:00 ◼   ► is not that new.

00:59:01 ◼   ► I mean, Metal is much newer than OpenGL.

00:59:03 ◼   ► And of course AMD has Mantle, which is a similar type of thing to Metal.

00:59:07 ◼   ► As far as I'm aware, I have not read a lot about Mantle, but sort of closer to the Metal,

00:59:12 ◼   ► which is I'm assuming where Apple got its name.

00:59:14 ◼   ► Mantle isn't closer to the Metal API.

00:59:17 ◼   ► It's one of the few areas in computing where the APIs are actually getting not so much

00:59:22 ◼   ► lower level, but exposing more of the foibles of the hardware rather than going the other

00:59:28 ◼   ► direction rather than abstracting it all.

00:59:30 ◼   ► Because OpenGL is more distant from the hardware than either Mantle or Metal or probably DirectX

00:59:37 ◼   ► for that matter.

00:59:38 ◼   ► So that's the situation with Metal, that's what Apple is doing, and when Apple announced

00:59:41 ◼   ► Metal in the keynote I tweeted something to the effect of, "So is Apple going to make

00:59:46 ◼   ► a game console or what? Because they just made their own sort of low-level game console-y

00:59:52 ◼   ► type 3D API. They've got a little box that you connect to a TV. They've got controller

00:59:58 ◼   ► support APIs built into iOS. Everyone keeps expecting them to offer some way to put apps

01:00:03 ◼   ► on the Apple TV. What's the deal here? Are they going to make one? Are they not going

01:00:08 ◼   ► to make one? Are they just going to dance around this until they have all the pieces

01:00:10 ◼   ► but then just refuse to do anything? Kind of like they did with eBooks for years and

01:00:13 ◼   ► years where they had all the pieces in place to be the world's dominant ebook maker and

01:00:17 ◼   ► then just let Amazon do it because they couldn't be bothered. And this comes up because there

01:00:22 ◼   ► was a strategic repost today that says Apple TV might disrupt Microsoft and Sony that talks

01:00:29 ◼   ► about the same issue. Have you guys read this?

01:00:31 ◼   ► Yeah, I think it's interesting. I mean, it's definitely true. One of Ben's main arguments

01:00:37 ◼   ► there is that the official high profile game consoles out there are these very high end,

01:00:46 ◼   ► hot, expensive devices and that it is necessarily keeping them high end because there are these

01:00:53 ◼   ► $400 boxes that it kind of locks out casual gamers who it's just not worth them spending

01:01:01 ◼   ► $400 on this box. Meanwhile, so many people are now getting things like the Apple TV or

01:01:07 ◼   ► Roku or the new Amazon fiery box, whatever it is, that, well, Merlin says it's good.

01:01:14 ◼   ► Anyway, so many people are getting these little cheap boxes that can stream video or watch

01:01:19 ◼   ► Netflix or whatever and that model, like that market, is probably very likely to disrupt

01:01:27 ◼   ► the gaming market at some point soon. That's kind of the gist, right?

01:01:31 ◼   ► I mean, basically.

01:01:32 ◼   ► I mean, I think if you look at Metal, I mean, Metal is really interesting just because,

01:01:37 ◼   ► you know, "Oh, look, it's a new level API that is a lot faster." But it's worth

01:01:41 ◼   ► asking why they made Metal. You know, why did Apple put so much effort into something

01:01:48 ◼   ► that is entirely about games? Because Apple historically has, hell, we were talking about

01:01:54 ◼   ► this a couple of weeks ago, how Apple just doesn't really seem to care about games.

01:01:58 ◼   ► Obviously they do. Obviously we were at least partially wrong on that.

01:02:01 ◼   ► I don't think it's entirely about games, though.

01:02:04 ◼   ► Well--

01:02:05 ◼   ► Because, I mean, think about--

01:02:07 ◼   ► the games is one big aspect of it, definitely.

01:02:10 ◼   ► But their whole UI uses OpenGL.

01:02:12 ◼   ► And viewed through the lens of iOS,

01:02:17 ◼   ► Metal starts to look more like a battery saving feature, where

01:02:20 ◼   ► they can more efficiently use the GPU with more precision,

01:02:22 ◼   ► exactly doling out the commands to it in the right order,

01:02:26 ◼   ► and having 10 times faster draw call performance,

01:02:30 ◼   ► and all these things.

01:02:31 ◼   ► That makes their UI snappier and has a potential battery savings

01:02:36 ◼   ► because you're spending less time flogging the GPU

01:02:38 ◼   ► to get your stuff done.

01:02:39 ◼   ► And their whole UI is basically-- it's not a 3D game,

01:02:42 ◼   ► but it's all OpenGL on iOS and on the Mac, for that matter,

01:02:46 ◼   ► these days.

01:02:47 ◼   ► So there is a platform benefit to it,

01:02:50 ◼   ► even if no game maker ever used it.

01:02:52 ◼   ► I don't think they would have gone through the effort

01:02:53 ◼   ► if no game maker used it.

01:02:55 ◼   ► And clearly, game makers like Metal

01:02:58 ◼   ► because all the big engines are saying,

01:03:00 ◼   ► yes we're going to support it. And I don't think Apple had to twist their arms because

01:03:03 ◼   ► this is what game makers want. They want their games to go faster. They don't like fighting

01:03:06 ◼   ► with OpenGL ES, trying to figure out what weird incantations they have to do to get

01:03:10 ◼   ► good performance. And game makers want good battery life too. They don't want people

01:03:14 ◼   ► to play a game and have it destroy your phone's battery.

01:03:16 ◼   ► That's fair. And I also think it's very obvious that there are a number of major strategic

01:03:22 ◼   ► benefits here. Obviously a lot of people are saying, "Oh well, this will encourage

01:03:26 ◼   ► people to write games only for iOS. And I think the number of people who will make that

01:03:32 ◼   ► decision based on Metal is probably very, very small because most people aren't writing

01:03:36 ◼   ► this low-level code. I mean, as you see, the people who make the engines, they're writing

01:03:41 ◼   ► code at this level, and they're possibly investing into making Metal ports for their

01:03:46 ◼   ► engines that will, you know, they're not going to stop making OpenGL versions. You

01:03:50 ◼   ► know, they're still going to, like, the big engines out there that everyone licenses

01:03:53 ◼   ► or uses, they're still going to have OpenGL versions so they can keep running on Android

01:03:56 ◼   ► and everything else. They're not going to just stop doing that. But it will reinforce

01:04:02 ◼   ► the pretty frequent idea that games, while they might run on Android stuff, they usually

01:04:10 ◼   ► run better on iOS. And it's usually because most Android stuff out there has a pretty

01:04:14 ◼   ► wide range of GPUs and GPU power. And some of the cheap Android stuff, like some of the

01:04:19 ◼   ► cheap Android tablets, historically have had very weak GPU power compared to comparable

01:04:24 ◼   ► iOS devices, partially at a cost,

01:04:26 ◼   ► partially out of other concerns, who knows,

01:04:28 ◼   ► but for whatever reason,

01:04:29 ◼   ► Android GPUs have historically sucked.

01:04:32 ◼   ► And yes, I know there are some devices that have good ones,

01:04:34 ◼   ► but I think the average is pretty bad

01:04:36 ◼   ► for what's actually sold.

01:04:37 ◼   ► ♪ It's not a toxic health stew ♪

01:04:40 ◼   ► ♪ Toxic health stew ♪

01:04:41 ◼   ► And I think this is a way for Apple to keep that lead going

01:04:45 ◼   ► for a while longer, you know,

01:04:46 ◼   ► like as the Android stuff gets better hardware

01:04:48 ◼   ► and starts catching up,

01:04:50 ◼   ► I think this is a way to just be like,

01:04:52 ◼   ► all right, let's keep that lead going.

01:04:53 ◼   ► but ultimately I think this does,

01:04:56 ◼   ► they were very specific that this is for the A7.

01:04:59 ◼   ► - Well, and for the A8, you assume,

01:05:01 ◼   ► I mean, like obviously.

01:05:03 ◼   ► But yeah, like, I mean, it's a total Apple thing to do.

01:05:06 ◼   ► What other company could even, you know,

01:05:09 ◼   ► could feasibly do this?

01:05:10 ◼   ► Because Apple has such a limited line

01:05:12 ◼   ► and is so relentless about moving things forward

01:05:16 ◼   ► and doesn't really care that this API

01:05:18 ◼   ► doesn't apply to their old things,

01:05:19 ◼   ► it's going to apply to all of their iOS devices

01:05:21 ◼   ► from this time forward, assuming the iPod Touch gets the A7 this year.

01:05:25 ◼   ► Oh yeah, the iPod Touch exists. Yeah, no, I mean, I think the fact that this is A7 only,

01:05:32 ◼   ► or A7 and above only, probably signifies that if Apple's going to enter the game market,

01:05:38 ◼   ► the way they're going to do it is by waiting until there's an A7 in the Apple TV, and then

01:05:42 ◼   ► they're like, "All right, hey, now we have this thing, start making games for it." And

01:05:47 ◼   ► I think we're waiting until that becomes economical.

01:05:50 ◼   ► And I think that's why we don't have an Apple TV SDK yet.

01:05:53 ◼   ► I think that's why the Apple TV hardware has done

01:05:56 ◼   ► very little in the last few years,

01:05:57 ◼   ► like has moved forward very little.

01:05:58 ◼   ► I think they're just waiting for economics

01:06:00 ◼   ► until they can combine this thing.

01:06:03 ◼   ► Keep the same price point, 100 bucks,

01:06:04 ◼   ► maybe even drop it by 20, 30 bucks,

01:06:06 ◼   ► but probably keep it at 100 bucks

01:06:09 ◼   ► and eventually put an A7 in there.

01:06:11 ◼   ► And that, heck, that might be this fall.

01:06:14 ◼   ► It would be a little aggressive,

01:06:15 ◼   ► but they could probably do it if they wanted to.

01:06:17 ◼   ► I think there's a very good chance of that happening

01:06:19 ◼   ► and of that happening soon.

01:06:21 ◼   ► That might even be the big fall thing is,

01:06:24 ◼   ► hey, you guys are all talking about wearables

01:06:26 ◼   ► that nobody wants.

01:06:27 ◼   ► We kind of made a game system that's gonna kick butt

01:06:29 ◼   ► in the market because we're gonna sell a ton of them anyway

01:06:32 ◼   ► and then you can start making games for it

01:06:33 ◼   ► using all of our existing infrastructure, using Metal,

01:06:36 ◼   ► and here's a new controller

01:06:37 ◼   ► so the controlling doesn't suck and that's it.

01:06:39 ◼   ► I mean, that would be a pretty amazing fall.

01:06:42 ◼   ► - Like I said during the V

01:06:42 ◼   ► Like I said during the W3C keynote,

01:06:44 ◼   ► Apple has all the pieces to make a run

01:06:48 ◼   ► of this exact strategy.

01:06:49 ◼   ► An inexpensive box with pretty good 3D performance.

01:06:53 ◼   ► They've got their own chip, they've got their own GPU,

01:06:57 ◼   ► they've got their own OS,

01:06:59 ◼   ► they've got now a low level console style API

01:07:01 ◼   ► for doing 3D and taking advantage of that hardware.

01:07:05 ◼   ► They've got the API for controller support.

01:07:08 ◼   ► All the pieces are there.

01:07:10 ◼   ► Unfortunately, all those pieces, the technology pieces,

01:07:13 ◼   ► are not the hard part of being successful

01:07:15 ◼   ► in the game console business.

01:07:17 ◼   ► Just ask Microsoft.

01:07:18 ◼   ► Getting the technology right is good,

01:07:21 ◼   ► but I don't think it's even necessary.

01:07:23 ◼   ► It's not like a necessary but sufficient thing.

01:07:25 ◼   ► I don't think it's even necessary to get the tech right.

01:07:27 ◼   ► The Wii's technology was disgusting,

01:07:29 ◼   ► and they were successful in the market, right?

01:07:31 ◼   ► So the hard part of being successful in the game market

01:07:34 ◼   ► is how do you get people who spend money on games

01:07:39 ◼   ► to buy your thing?

01:07:40 ◼   ► And the answer to that would be like,

01:07:41 ◼   ► oh, they'll buy it because it's like,

01:07:43 ◼   ► you know, it's a great TV puck to watch Netflix on

01:07:45 ◼   ► and whatever, and it's cheap,

01:07:47 ◼   ► so that'll get people to buy things.

01:07:48 ◼   ► And then how do you get game developers

01:07:51 ◼   ► to develop good games for your platform?

01:07:53 ◼   ► And just ask the Ouya people how easy that is to do.

01:07:56 ◼   ► It is really, really hard,

01:07:58 ◼   ► and probably costs potentially billions of dollars

01:08:01 ◼   ► if you want to make a run at this market

01:08:04 ◼   ► and get all those people who are currently buying

01:08:06 ◼   ► $60 games for their PS4s.

01:08:08 ◼   ► and like we've talked about before,

01:08:09 ◼   ► the next generation of consoles was announced,

01:08:11 ◼   ► the question was, is there still a place in the market

01:08:13 ◼   ► for big expensive devices that play mostly or only games?

01:08:17 ◼   ► The answer is yes, people love the PS4,

01:08:20 ◼   ► it is selling pretty well,

01:08:21 ◼   ► certainly well enough to make Sony happy.

01:08:23 ◼   ► Microsoft's console is selling pretty well as well,

01:08:25 ◼   ► no one wants the Wii U because it's crappy.

01:08:27 ◼   ► Maybe their, you know, E3 was this past week,

01:08:31 ◼   ► so maybe Nintendo's fortunes will rise next year,

01:08:34 ◼   ► we don't know.

01:08:35 ◼   ► But this generation of consoles, I would say,

01:08:37 ◼   ► is a success. And when this generation ends, the next generation will ask the same question again.

01:08:42 ◼   ► Does anyone want these stupid $400 boxes? Do people want to pay $60 per game? That question

01:08:46 ◼   ► will be asked again. But for this generation, it's true. If no one wants those boxes, and there's not

01:08:53 ◼   ► enough people out there to sustain that, the potential for the entire gaming market to switch

01:08:58 ◼   ► to $99 pucks, I think most gamers would consider that kind of a bad situation. And the question is,

01:09:04 ◼   ► Sorry, well if there's so few of those people it doesn't matter what they think they'll be sad, but oh well who cares

01:09:09 ◼   ► It's like the command line people being sad when the GUI came and took over everything and they just went and played with Linux and no

01:09:14 ◼   ► One cared about them anymore

01:09:16 ◼   ► I don't think that scenario is likely though even though these so-called core gamers

01:09:21 ◼   ► And I hate that term because it makes no sense, but the so-called core gamers even though they are not the majority

01:09:25 ◼   ► They sure as hell spend a lot of money

01:09:28 ◼   ► When when when a game as someone just been the chat room in the game like Grand Theft Auto 5 comes out and makes a hoe

01:09:33 ◼   ► billion dollars, casual gamers are not buying that, right? Hardcore gamers are

01:09:38 ◼   ► paying $60 a pot for these games, they're doing it willingly, they're happy with

01:09:42 ◼   ► the results, and they pump a huge amount of money into the ecosystem. And it's not

01:09:46 ◼   ► like just a few whales like on Candy Crush where most of the money is made

01:09:48 ◼   ► off of these poor addicted small group of people when everyone else just kind

01:09:52 ◼   ► of uses it for a little while or whatever. Everybody pays $60 to get

01:09:56 ◼   ► grant that thought of. And those people are still willing to do it. And that I

01:10:01 ◼   ► I think is the business that Apple would have a seriously hard time getting into, because

01:10:05 ◼   ► as we've discussed previously, it doesn't seem like Apple's that into games. If you're

01:10:09 ◼   ► not that into games, there's no way you're going to do what it takes to woo the good

01:10:13 ◼   ► developers. Plus, they would have to have some IP of their own, like Microsoft said

01:10:16 ◼   ► Halo and Sony's got its own first-party games, Nintendo's got Mario and Donkey Kong and all

01:10:22 ◼   ► that stuff. I don't think Apple can be a player in that market.

01:10:25 ◼   ► Well, I don't think Apple wants to be a player in that market.

01:10:28 ◼   ► Well, that was the question like and then I said to Ben Thompson is like

01:10:31 ◼   ► That's the things that they have are the technical ingredients to be a player in that market

01:10:36 ◼   ► But it sure seems like Apple doesn't want that market

01:10:38 ◼   ► Why would you want to be part of what by all accounts is a declining market?

01:10:42 ◼   ► Despite the fact that is clearly feasible now and makes tons of money

01:10:46 ◼   ► Why would Apple want in on that market?

01:10:48 ◼   ► Is that a market that's taking off like a rocket ship to the Sun and is gonna be super awesome for years to come?

01:10:54 ◼   ► Doesn't seem like it right now. So why would Apple be saying we really need to get in on this market

01:10:59 ◼   ► We really need to make a device that's competitive with the ps5 and the Xbox whatever the hell they're gonna call it

01:11:04 ◼   ► I don't see Apple thinking that that is a good thing to do regardless of whether they could do it

01:11:09 ◼   ► Yeah, but that wasn't that wasn't Ben's point though was it I thought Ben's point was more to grab the portion of the market

01:11:16 ◼   ► That's more like Marco and I where we don't really take games too seriously

01:11:20 ◼   ► But we do enjoy it and hey man if we've already got this Apple TV box sitting

01:11:26 ◼   ► Oh, well, I mean presumably a new one. You know I just bought this Apple TV

01:11:29 ◼   ► Oh, you know I guess it plays games

01:11:31 ◼   ► Maybe I should give that a shot and see if that's any good and yes, they may have this white this world-class

01:11:37 ◼   ► Okay, that might be a stretch, but they may have this really really great hardware and really really great tool chain

01:11:43 ◼   ► But that's kind of ancillary to the thought that here's a low-cost box

01:11:48 ◼   ► that makes for really easy consumption.

01:11:52 ◼   ► And I think that was more Ben's point.

01:11:55 ◼   ► - Yeah, but Apple already has that market.

01:11:56 ◼   ► It's called the iPhone.

01:11:58 ◼   ► - You know, if you look at what the existing game consoles

01:12:00 ◼   ► do, and you look at the direction they are going,

01:12:03 ◼   ► they have all decided, even the previous generation,

01:12:06 ◼   ► I mean, not the PS2, but the PS3, the Xbox 360,

01:12:11 ◼   ► and the Wii, the first Wii, they all started adding

01:12:15 ◼   ► all these media center features.

01:12:16 ◼   ► Once streaming Netflix became a thing

01:12:18 ◼   ► and all these streaming video sites,

01:12:21 ◼   ► that all happened in the last generation of consoles.

01:12:24 ◼   ► And they all added these features

01:12:25 ◼   ► and they all took advantage of that.

01:12:26 ◼   ► And they all, with their current generation,

01:12:28 ◼   ► the new generation, with the exception of the Wii U,

01:12:31 ◼   ► which didn't really do this kind of stuff,

01:12:33 ◼   ► but the other ones moved much further

01:12:36 ◼   ► towards that media center role.

01:12:39 ◼   ► 'Cause you know, people who buy these things,

01:12:41 ◼   ► if you have very little interest in playing games,

01:12:44 ◼   ► you're probably not buying an Xbox One, the new Xbox One stupid name, or the PS4. You're

01:12:51 ◼   ► probably not going to think that's worth it, which is what Ben's article says. This

01:12:56 ◼   ► must be really frustrating because Ben's in the chat room listening to us mangle his

01:12:58 ◼   ► article but, sorry Ben, but you know if you aren't that into games you're not buying

01:13:05 ◼   ► those things period. Like you're not going to be getting those unless maybe your kids

01:13:09 ◼   ► want one so you get it for your kids but you're probably, as the parent, you're probably

01:13:12 ◼   ► not gonna use that yourself very much if you can help it. There's this whole other market

01:13:18 ◼   ► of people like me and Casey, people who primarily want it for its media functions and if it

01:13:25 ◼   ► can also play games, great. But we weren't going out there and spending 60 bucks to buy

01:13:29 ◼   ► GTA 6. Like that was not us. That was possibly never us. It's definitely not us now and

01:13:37 ◼   ► there's a lot of us like this. And I think the game console makers realize that they're

01:13:42 ◼   ► market is declining because they're trying to be this media boxes also.

01:13:46 ◼   ► They're trying to push in this direction.

01:13:48 ◼   ► Apple already has a very successful little media box people buy and put on their TVs

01:13:53 ◼   ► and Apple is going to approach it from the other way.

01:13:56 ◼   ► The same way the Amazon Firebox is doing it.

01:14:01 ◼   ► Here's a mid-powered cheap box that you can plug into your TV.

01:14:05 ◼   ► Here's a game controller for it and an app store where you can spend $3 to buy a game

01:14:09 ◼   ► for it if you want to.

01:14:10 ◼   ► and it's not gonna be a AAA awesome world-class game

01:14:15 ◼   ► with a $50 million budget, it's gonna be Flappy Bird 3.

01:14:18 ◼   ► And you're gonna like it and that's it.

01:14:20 ◼   ► And it's a totally different market.

01:14:23 ◼   ► - Apple TV will add games the same way

01:14:25 ◼   ► that the other consoles added TV features.

01:14:27 ◼   ► They'll add it because it's right there,

01:14:29 ◼   ► you might as well do it, you have the hardware available,

01:14:32 ◼   ► but people are not buying a PlayStation

01:14:34 ◼   ► so they can watch TV shows and people are not gonna buy

01:14:37 ◼   ► the Apple thing to play games unless Apple gets serious

01:14:39 ◼   ► about games, because what Apple will be competing with is the $99 box from PlayStation, which

01:14:45 ◼   ► they've had out for a while in Europe and they just renamed and are introducing the

01:14:49 ◼   ► US as PlayStation TV.

01:14:52 ◼   ► Why would you buy the $99 box from PlayStation that streams Netflix and does whatever the

01:14:56 ◼   ► hell they're going to do, you know, like does all the same things that these little TV-connected

01:14:59 ◼   ► pucks do?

01:15:00 ◼   ► Why would you buy the PlayStation 1 and not the Amazon one?

01:15:02 ◼   ► Why would you buy the Amazon one and not the Apple one?

01:15:04 ◼   ► Well, if you care anything about games, of course you're going to buy the PlayStation

01:15:07 ◼   ► because they have that thing where you can play old PS3 games on it with a streaming service,

01:15:11 ◼   ► it plays Vita games natively, it's always going to have the better games. But if you don't care

01:15:16 ◼   ► about games, you'll buy the Amazon Fire TV and play Angry Birds, or you'll buy the Apple one

01:15:20 ◼   ► and play whatever iOS games are ported to it. That bifurcation in the market already exists,

01:15:25 ◼   ► and everyone thought the iOS app store was going to totally disrupt the gaming space,

01:15:29 ◼   ► and in many respects it did in terms of what kinds of games most people play,

01:15:34 ◼   ► But it didn't disrupt it to the degree that the home console business became unviable.

01:15:39 ◼   ► And so if that split just continues on the television, I think that you'll still end up

01:15:44 ◼   ► with the same split scenario where you have the core gamer market and the casual gamer market.

01:15:48 ◼   ► And I don't think anyone will buy the Apple Puck to play games on if the PlayStation Puck exists,

01:15:54 ◼   ► or the Microsoft Puck, or whatever. It's going to be the same thing in the Puck Wars on the TV,

01:15:58 ◼   ► where it's just which one you want depends on what you care about more.

01:16:02 ◼   ► the way you win in the game space is not having the tech and not having the APIs,

01:16:05 ◼   ► not having the price point, you gotta have the games.

01:16:08 ◼   ► And Apple has certain kinds of games that they have locked up.

01:16:11 ◼   ► Like, you know, the iOS-style games, they have them first, they have them best,

01:16:15 ◼   ► they run the best there, but they don't have the kinds of games that people tend to sit in front

01:16:19 ◼   ► of a TV and play. And I'm not sure that the casual type of games that Apple is dominant on,

01:16:24 ◼   ► if ported to the television, if that's the type of thing that people will sit in front of the TV

01:16:28 ◼   ► and play on a couch. I think once they sit on a couch and are going to play games in that context,

01:16:32 ◼   ► they would prefer to get the PlayStation puck for a similar price and play those games.

01:16:38 ◼   ► You're coming at this all wrong because you're coming at this from the perspective of a gamer,

01:16:42 ◼   ► and I don't think that's what Ben is trying to portray. He's trying to say it's a more

01:16:46 ◼   ► opportunistic attention grab based on the fact that Marco and I both have these brand new Apple

01:16:52 ◼   ► TVs, and you know what? I wouldn't mind playing an okay game every once in a while. I don't give

01:16:57 ◼   ► crap about Grand Theft Auto or Call of Duty or any or what have you I just want

01:17:01 ◼   ► to occasionally sit down and blow some crap up and for that this phantom Apple

01:17:06 ◼   ► TV is perfect. We're gonna make our buying decisions not not based on which

01:17:12 ◼   ► of these pucks has the games we want we're gonna make it based on other

01:17:15 ◼   ► factors like if we if we use if we have Amazon Prime and this fire TV is really

01:17:21 ◼   ► cheap and comes with this this video streaming service that we're already

01:17:23 ◼   ► paying for with Prime is like well might as well get that if we if we if we buy a

01:17:27 ◼   ► a lot of stuff on iTunes, if we buy a lot of movies and TV shows on iTunes, we're

01:17:31 ◼   ► going to buy the Apple TV puck because the PlayStation puck can't play those videos.

01:17:34 ◼   ► And so it's like, if we're going to buy a puck anyway, you know, if we only want

01:17:39 ◼   ► to have one or if we only can afford one, then the other factors that go into these

01:17:44 ◼   ► various media ecosystems that these things are made to play are going to play a big role

01:17:49 ◼   ► in that decision. And so you're going to have Amazon customers buying the Amazon one

01:17:52 ◼   ► and iTunes customers buying the Apple one.

01:17:54 ◼   ► Well, that's what I said before, that if you're choosing based on games, you're going to choose the Sony one,

01:17:59 ◼   ► but if you're not choosing based on games, you'll choose the Amazon one or the Apple one.

01:18:02 ◼   ► But the whole, the premise of this article is how Apple might disrupt Microsoft and Sony.

01:18:06 ◼   ► And it's not disrupting Microsoft and Sony if people are not changing their purchase decisions.

01:18:09 ◼   ► If you were going to always buy an Apple TV and now games are just a bonus,

01:18:12 ◼   ► you are not, you know, disrupting Microsoft or Sony unless you were previously going to buy a PS5 or an Xbox, whatever,

01:18:19 ◼   ► and you decide, you know what, instead, I'm going to buy the Apple thing because that's good enough,

01:18:22 ◼   ► or I'm going to buy the Sony puck because that's good enough. To disrupt the market you have to basically take your low-end thing and

01:18:28 ◼   ► buy that instead of the high-end thing

01:18:31 ◼   ► and you have to eat your way up the chain. And I'm saying eating your way up the chain at a certain point you cannot

01:18:36 ◼   ► eat up any higher because you just don't have the games. And it's not saying Apple could not do this,

01:18:41 ◼   ► I'm saying the hard part is not the technology,

01:18:44 ◼   ► the hard part is getting the games to consume up in the higher end. And again people thought this was gonna happen in the

01:18:49 ◼   ► the mobile game space like oh you think these mobile games are crappy and casual but soon they're gonna just destroy the entire market and

01:18:55 ◼   ► No one's gonna be able to like low-end disruption you you poo-poo these games

01:18:58 ◼   ► You think Angry Birds is stupid and no one should play that and Candy Crush is dumb

01:19:01 ◼   ► But you just wait in five years there won't even be a high-end market because low-end disruption will destroy them

01:19:05 ◼   ► That has not yet happened. It doesn't seem like it's happening

01:19:08 ◼   ► I think you cannot you cannot disrupt the big players in the game market without having really good games

01:19:15 ◼   ► And I don't think it doesn't mean they have to be as high-tech or whatever

01:19:18 ◼   ► you just have to have the really good games.

01:19:19 ◼   ► And thus far, Apple has not shown that it's willing

01:19:22 ◼   ► to do what it takes to get the really good games

01:19:24 ◼   ► to stop people from saying, you know what,

01:19:26 ◼   ► I was gonna buy a PS5, but I don't need it anymore.

01:19:28 ◼   ► All my gaming needs are now met by this Apple device.

01:19:31 ◼   ► - I get your point.

01:19:32 ◼   ► Your point is correct that Apple will not take over

01:19:35 ◼   ► the high end, but I think Apple or boxes like this,

01:19:40 ◼   ► if Amazon continues their effort,

01:19:42 ◼   ► if it actually succeeds with games,

01:19:44 ◼   ► I think the big risk here is not that these media companies

01:19:47 ◼   ► you're gonna take over the high end of gaming,

01:19:49 ◼   ► it's that they're going to be good enough

01:19:51 ◼   ► and with such a massive price advantage,

01:19:54 ◼   ► I mean you see this all the time,

01:19:55 ◼   ► this is a classic disruption.

01:19:56 ◼   ► - Well, I was just saying,

01:19:57 ◼   ► there would be no more high end,

01:19:58 ◼   ► that's low end disruption.

01:19:59 ◼   ► There is no, the high end is not viable,

01:20:01 ◼   ► then no one can be a high end.

01:20:02 ◼   ► They don't have to take over the high end,

01:20:03 ◼   ► then no one can be the high end 'cause it's not viable.

01:20:04 ◼   ► But that's what I'm saying.

01:20:06 ◼   ► If this was gonna happen,

01:20:07 ◼   ► I would think we'd be seeing it happen already.

01:20:09 ◼   ► Like, that there is no more room for high end.

01:20:13 ◼   ► The $60 games are not a thing.

01:20:15 ◼   ► Apple doesn't have to get those customers,

01:20:17 ◼   ► just have to starve the companies that make them to death to say you can keep

01:20:20 ◼   ► trying to do that but we're gonna make your business unviable and then we're

01:20:22 ◼   ► all that's left in the end and I don't think that can happen because I think

01:20:26 ◼   ► enough people want really good games the kind of which Apple has thus far shown

01:20:31 ◼   ► it is not willing or able to get made for its platform maybe anyway the one

01:20:39 ◼   ► thing Apple has going for it is that in the realm of TV connected pucks Apple

01:20:45 ◼   ► Apple has a good store, Amazon has a pretty good store.

01:20:49 ◼   ► The game console makers are not that great about having nice, convenient app stores that

01:20:54 ◼   ► regular people can use that don't suck, that don't have your credit cards stolen from them,

01:20:57 ◼   ► speaking of Sony.

01:20:59 ◼   ► Microsoft may be doing a little bit better.

01:21:02 ◼   ► The one advantage that Apple has is they have experience in selling you $3 things electronically,

01:21:09 ◼   ► and people are familiar with their stores, and if they can leverage their iOS advantage

01:21:13 ◼   ► that you can say, oh, you buy it here and you can play it on your TV as well and all

01:21:16 ◼   ► these other things like that.

01:21:17 ◼   ► That is an advantage they have in the gaming space for casual gaming, but the way I see

01:21:21 ◼   ► it playing out, unless something changes seriously, is that your TV will become just like every

01:21:27 ◼   ► other part of the current market where there's a split between the people who want the big

01:21:33 ◼   ► expensive thing and buy $60 games and the people who want everything else.

01:21:36 ◼   ► And the ratios, I imagine, would be similar there.

01:21:39 ◼   ► And I do think it's a declining market.

01:21:40 ◼   ► I think the people who want $60 games is declining, it's just not declining so precipitously that

01:21:45 ◼   ► I'm willing to say that Apple or Microsoft, or Sony for that matter, are poised to disrupt

01:21:51 ◼   ► the high-end gaming market by making it unviable.

01:21:55 ◼   ► Sony clearly is hedging its bets with its little TV puck thing, saying, "Well, if it's

01:21:58 ◼   ► going to happen, we should do it to ourselves, and they'll sell this puck and they'll see

01:22:01 ◼   ► how it sells, and if it starts to catch on, I'm sure Sony has no problem redirecting their

01:22:05 ◼   ► attention in the next generation to puck-like devices."

01:22:08 ◼   ► I don't know if Microsoft's planning to have a puck, but it wouldn't surprise me.

01:22:11 ◼   ► So everyone is kind of hedging their bets, but I think the success of the PS4 and Xbox

01:22:15 ◼   ► One are making people say, "Alright, we've got one more generation breathing room.

01:22:18 ◼   ► Let's start planning our next gen and see how things shake out."

01:22:22 ◼   ► So thanks a lot to our three sponsors this week, Automatic, Lynda.com, and Backblaze.

01:22:30 ◼   ► And we will see you next week.

01:22:32 ◼   ► Now the show is over, they didn't even mean to begin

01:22:39 ◼   ► 'Cause it was accidental, oh it was accidental

01:22:45 ◼   ► John didn't do any research, Marco and Casey wouldn't let him

01:22:50 ◼   ► 'Cause it was accidental, oh it was accidental

01:22:56 ◼   ► And you can find the show notes at ATP.fm

01:23:01 ◼   ► And if you're into Twitter, you can follow them @C-A-S-E-Y-L-I-S-S

01:23:09 ◼   ► So that's Kasey Liss, M-A-R-C-O-A-R-M

01:23:14 ◼   ► Auntie Marco Arment, S-I-R-A-C-U-S-A-C-R-A-C-Uza

01:23:21 ◼   ► It's accidental (It's accidental)

01:23:24 ◼   ► They didn't mean to (accidental)

01:23:29 ◼   ► Tech, podcast, so long

01:23:34 ◼   ► We have a special guest tonight. It is my wonderful wife, Tiff.

01:23:38 ◼   ► And because Tiff actually did my homework for me.

01:23:42 ◼   ► Oh, that's right!

01:23:44 ◼   ► Tiff has played Journey. After years of listening to John tell us about Journey and why we should all be playing Journey,

01:23:53 ◼   ► Tiff actually played it because John discovered while talking to Tiff at WVDC this week that she is a gamer.

01:24:00 ◼   ► that I have apparently hidden this fact from all of you guys until this point,

01:24:04 ◼   ► but she is a gamer sometimes and she's actually better than me at almost

01:24:11 ◼   ► every game that we've both played. And anyway, so she did my homework for me and

01:24:15 ◼   ► here she is. Hey, what's going on? Lay it on me, Tiff. All right, well, wait, now

01:24:22 ◼   ► before you do anything, if I hypothetically did eventually want to

01:24:25 ◼   ► play this game, is this gonna spoil everything? Yeah, you should probably

01:24:28 ◼   ► leave. And neither should Mark over that. Did he watch you play? He watched me play enough that he

01:24:32 ◼   ► knows what's going on because he's never going to play. I know him. He's so terrible. Yeah, I'm

01:24:36 ◼   ► never going to play either. I don't have a PlayStation, so I'm never going to play.

01:24:39 ◼   ► Casey, you should not listen. I'm telling you you should go away and come back later.

01:24:43 ◼   ► Yeah, where's that big spoiler bell? Do you really want me to not listen?

01:24:47 ◼   ► Because the likelihood of me playing is- The spoiler bell. Jason Snell is crying

01:24:49 ◼   ► somewhere. It's a spoiler horn, isn't it? Do you really want me to go away?

01:24:55 ◼   ► Yes, you should. You played Monument Valley. You could potentially play this. I think you

01:24:59 ◼   ► should go away and come back. Well, just to be clear, I cannot play this game until somebody...

01:25:05 ◼   ► You'll be able to play it on your $99 PlayStation puck in about a year or so.

01:25:09 ◼   ► All right. I'm going to take my headphones off and just watch the chat ruin everything for me.

01:25:13 ◼   ► It's just going to be you and me, John. Yay!

01:25:15 ◼   ► That's all we need. We don't need those jokers.

01:25:17 ◼   ► Wow! Zing! All right, I'm taking my headphones off.

01:25:21 ◼   ► Bye. Is he gone?

01:25:24 ◼   ► Yes, we're safe.

01:25:25 ◼   ► Okay.

01:25:26 ◼   ► So, where should I start?

01:25:28 ◼   ► I guess at the beginning.

01:25:29 ◼   ► I've done all my homework.

01:25:30 ◼   ► I played the game now four times, and I listened to The Incomparable, and I read your article.

01:25:37 ◼   ► So, homework done.

01:25:39 ◼   ► So you tweeted after you played with an emoji.

01:25:41 ◼   ► I'm not emoji-convercent enough to know what you meant by that emoji, but it was like the

01:25:48 ◼   ► biting down teeth kind of like...

01:25:50 ◼   ► What it made it seem like is that you did not like this game when you first played it.

01:25:53 ◼   ► Well, the first time I played it without anything, just sitting down playing it, knowing nothing.

01:25:59 ◼   ► I didn't like it.

01:26:00 ◼   ► All right. So, so explain that.

01:26:01 ◼   ► Well, I thought, first of all, so spoiler bell, I thought like the people who I was

01:26:06 ◼   ► with were AI. So I was like, Oh, this is just stupid. This person just walking around with

01:26:11 ◼   ► me. I'm just doubled. And then I felt like it was a game that like you got siphoned through.

01:26:16 ◼   ► See that is super interesting because that's why I don't want people to know. Like I told

01:26:21 ◼   ► Marco to make sure that the PlayStation was connected to the network but not to tell you that it was a network game because I

01:26:25 ◼   ► Don't want people to know that the people are real people and it's interesting to see

01:26:28 ◼   ► Some people play the game and immediately realize those are other people some people play the entire game and just assume their AI the entire

01:26:35 ◼   ► Game right and that's what I assumed and then because there's a really no like challenge

01:26:39 ◼   ► You just kind of like ooze through this world and you do some things and you know certain things will attack you

01:26:45 ◼   ► But there are no real quads like no consequences for it. So I

01:26:50 ◼   ► Just like this is just boring. See here's what I was hoping. I was hoping that as a gamer you would

01:26:57 ◼   ► Realize when the people are interacting with you that they couldn't possibly be AI because they exhibit behaviors in no way

01:27:03 ◼   ► I could possibly exhibit well

01:27:04 ◼   ► Maybe like the people that I played with were just kind of like helping me like jump and walking around with me

01:27:09 ◼   ► They weren't doing anything extraordinary

01:27:11 ◼   ► but the

01:27:12 ◼   ► after I listened to the incomparable and then realized that they were real people and then I started getting into like everything that you

01:27:18 ◼   ► guys were talking about on that episode about why it was a beautiful game and why it was

01:27:22 ◼   ► good and why it was fun and the whole the article that you wrote about people being

01:27:26 ◼   ► nice to each other and all that that made me want to play the game again. So when I

01:27:30 ◼   ► played it the second time, I really felt like I got it, you know, because then I was also

01:27:35 ◼   ► playing with someone who was like showing me where things were so I can get my scarf

01:27:40 ◼   ► really long. So that was fun. So I was like following him around. And so I felt like that

01:27:45 ◼   ► was I guess more of the right way to play the game.

01:27:49 ◼   ► Yeah, you only get one chance to play it for the first time, which is kind of a bummer.

01:27:52 ◼   ► And like I said, I think I said on the incomparable, I feel lucky that my first playthrough was

01:27:56 ◼   ► so perfect in the way that it was. The other thing I was hoping you as a gamer who like

01:28:00 ◼   ► plays things on hard mode was that you wouldn't, that you wouldn't like realize that there

01:28:06 ◼   ► are no consequences for failure because you wouldn't fail like that you wouldn't be attacked

01:28:10 ◼   ► by those things and have your scarf being eaten and realize, oh, no matter how many

01:28:14 ◼   ► times attack I never died as my scarf get shorter until a certain point I don't actually

01:28:18 ◼   ► die as I said in the incomparable the great thing about the game is it doesn't tell you

01:28:23 ◼   ► what the consequences are so you don't you don't have a health meter you don't know what

01:28:26 ◼   ► it takes for you to die or if you can die it also so if you're in a particular mindset

01:28:30 ◼   ► you might be like I don't know what's going to happen here I have nothing to fall back

01:28:33 ◼   ► on I can't look at my number of hearts up on the screen and know that I'm okay if I

01:28:36 ◼   ► get hit once for all I know it's single hit death and I don't want to die because I'm

01:28:40 ◼   ► in the middle of this experience with this other person, or even depending on what kind

01:28:44 ◼   ► of gamer you are, even if you assume it's an AI, I'm the type of person who could find

01:28:48 ◼   ► themselves getting attached to an AI. Like if there was just a really good AI in a game

01:28:52 ◼   ► and I felt like protective of it or attached to it as a companion, I would be sad if it

01:28:57 ◼   ► died or like in my journey experience when I was playing with that person, even though

01:29:01 ◼   ► I realized it was a person, even if it was AI, I would feel like they let them down by

01:29:04 ◼   ► like when you're hiding in those little, in the little pieces in the snowy part and someone

01:29:08 ◼   ► pokes out and like, oh, now you're both screwed because you made a mistake. And I just felt

01:29:12 ◼   ► terrible that I did that to that person. It was like attention for the rest of the game

01:29:16 ◼   ► that I had let this person down, you know? Yeah. I mean, I totally get that. And I especially

01:29:20 ◼   ► felt that the second time through, like I was really upset when I lost my, my companion

01:29:25 ◼   ► the second time I played it. Cause he was helping me do all this stuff and we got separated

01:29:29 ◼   ► avarice, you know, one part with the dragon and I was really upset and I was like waiting

01:29:33 ◼   ► for him and like singing my little song in the dark, like at the end, like waiting and

01:29:37 ◼   ► he wasn't showing up. And then I ended up in the next level with some little tiny short

01:29:43 ◼   ► scarf noob jumping up and down. And I'm like, "Oh, that was me yesterday." And he was singing

01:29:47 ◼   ► and yelling and jumping. And I'm like, "I can't help you. I don't know what I'm doing

01:29:50 ◼   ► either." So I just left him in the dust and I kept on going. And then my long scarf friend

01:29:54 ◼   ► showed up again. And so I was like, "We rejoice and sang little songs together." And so that

01:29:58 ◼   ► was really great. But that was, again, my second experience when I knew it was a real

01:30:03 ◼   ► And then but the first time I played through right at the end, the when you're going up

01:30:09 ◼   ► the tower, the snow tower, my like little buddy who I was with, who I thought was an

01:30:13 ◼   ► AI like froze to death, he just like sat down and was like shivering and not moving. And

01:30:18 ◼   ► I was like singing Adam and nothing was happening. So I just went on alone. And when I was the

01:30:23 ◼   ► first time I thought like, Okay, so I had a companion through the middle of the game,

01:30:27 ◼   ► I didn't have one in the beginning. And then he kind of like froze to death at the end.

01:30:31 ◼   ► So I had to like do the end alone.

01:30:33 ◼   ► I'm like, oh, okay, I guess this is just the time that I don't have a double of me that

01:30:37 ◼   ► I go up with.

01:30:38 ◼   ► So it was definitely the first time around wasn't as fun as the second time.

01:30:44 ◼   ► What I told my wife when she was asking that I said that I thought you didn't like the

01:30:47 ◼   ► game and she was asking why I said my theory was that since you are a gamer, that I didn't

01:30:53 ◼   ► know that you had the did that you thought it was an AI the whole time.

01:30:56 ◼   ► But this fits is that if you feel like you're playing a game, the whole time you're playing

01:31:01 ◼   ► this game, and this would fit for somebody who plays games on hard because you want a

01:31:04 ◼   ► challenge.

01:31:05 ◼   ► If you were always in the headspace of "I am playing a game," it's kind of like if you

01:31:09 ◼   ► watch a movie and you're always in the headspace of "I am watching a movie."

01:31:12 ◼   ► The movies that get you, you forget you're watching a movie and you feel like you're

01:31:15 ◼   ► experiencing what you're experiencing.

01:31:18 ◼   ► Journey relies entirely on that happening, because not every game does.

01:31:21 ◼   ► If you're playing a first-person shooter in super hard mode, you could feel like you're

01:31:24 ◼   ► playing a game the whole time and it's totally satisfying.

01:31:26 ◼   ► Oh yeah, you have to be engaged.

01:31:28 ◼   ► thing about Journey was that I felt like when sometimes you're going through some of the areas,

01:31:32 ◼   ► I felt like I could just let go of my controller and it would just go there for me. You know,

01:31:36 ◼   ► like I was just being funneled through the game. Well, yeah, like it is you can't feel like you're

01:31:41 ◼   ► playing a game. Like because yes, if you think about it for two seconds, you realize, oh,

01:31:43 ◼   ► this section is on rails, I can let go. But it's like, you know, you can't think that way. You have

01:31:49 ◼   ► to think I am here, this is happening to me and why would I ever want to let go. So for the

01:31:53 ◼   ► sections where you're sort of surfing down the hill and stuff, the reason I found those levels

01:31:57 ◼   ► so amazing is because like early in the game you realize you can slide and it's just it's just like there's no point there

01:32:03 ◼   ► There's no you're not you don't get extra points for going through the little stanchions

01:32:08 ◼   ► You know the little gates that you can go that's that's not there like oh you have to do that in your extra bonus

01:32:12 ◼   ► No, it's all about just having fun

01:32:14 ◼   ► You have to feel like you're there and enjoying the little sort of

01:32:16 ◼   ► Dolphin things diving through the sand and going the why because do you get some reward when you jump on their heads?

01:32:22 ◼   ► No, you jump in their heads because it's fun and you really have to just feel like you're there

01:32:25 ◼   ► And if the entire time you feel like you're playing a game, then it's like well this game is not challenging

01:32:29 ◼   ► This game is not interesting this game doesn't offer me anything doesn't offer me

01:32:33 ◼   ► Extra power-up extra weapons all the type of things that you would you know in a game in a game where you can be okay?

01:32:38 ◼   ► feeling like you're playing a game you have to

01:32:40 ◼   ► Not feel like you're playing a game for this and feel like you're having experience and realizing there are other people probably helps you with

01:32:45 ◼   ► That because then it's like oh

01:32:46 ◼   ► I'm not playing a game in the same way that suddenly you feel like you know you ever see people like texting walking down the street

01:32:52 ◼   ► And the expressions on their face are you know?

01:32:54 ◼   ► angry or joyful. Oh, yeah, yeah, yeah, they're holy, they're going through all the emotions of

01:32:59 ◼   ► what's happening, but nothing's happening to them, really. And what they're using is an app that has

01:33:03 ◼   ► a text box and a bunch of little text box over and over and over, but their experience is like,

01:33:08 ◼   ► oh, no, no, they're connected to the person on the other end of this. Like, and Journey does that

01:33:12 ◼   ► with significantly better graphics than the messages app on iOS, right? And can elicit the

01:33:16 ◼   ► same type of emotions because like I said in the article, they shave off every possible way that

01:33:21 ◼   ► person can be a jerk to you and all that's left is the good and in general people are

01:33:25 ◼   ► good to each other. And even if they try to be a jerk, it just comes off as like, you

01:33:29 ◼   ► know, craziness or disinterest, like they can't do anything to harm you. So it's like

01:33:32 ◼   ► a relentlessly positive experience.

01:33:34 ◼   ► With it with that whole argument, though, that's exactly why I disagree with you that

01:33:37 ◼   ► when you first play the game that someone should know that you're playing with other

01:33:41 ◼   ► people because the game was infinitely more enjoyable for me when I realized that it was

01:33:46 ◼   ► just another person with me, instead of just an AI. And so I didn't enjoy the game when

01:33:51 ◼   ► I thought it was AI, but I love the game after I found out it was a real person. And I've

01:33:56 ◼   ► gone back twice now and played it once I played under Marco's name and I had to go get my

01:33:59 ◼   ► own name so I can get my own scarf stuff. And I didn't have Tumblr more. What was it?

01:34:04 ◼   ► In the paper Marco. They're like, you played with instapaper Marco. It's like, no, you

01:34:09 ◼   ► played with Tiff Arman, whoever I am. I don't know. But like, that's, you know, that that

01:34:15 ◼   ► whole experience of with other people made it so much better. And I would have never

01:34:20 ◼   ► played the game again, if it was anything like the first time I played it.

01:34:23 ◼   ► I was hoping that you would not know that think they're AI initially, but fine, but

01:34:28 ◼   ► figure it out after like five minutes, because that realization that it's another person

01:34:32 ◼   ► is a fun realization. Because, you know, again, if you don't like that, that's the common

01:34:37 ◼   ► experiences. If they don't know it's AI, they initially think it is AI, but then figure

01:34:40 ◼   ► out that it's not

01:34:41 ◼   ► the it's kind of like the reverse Blair Witch Project. Like if you see the movie, and then

01:34:45 ◼   ► you and you think it's real, it's great the first time you see it, and then you realize

01:34:49 ◼   ► it's not real and it sucks. It's the worst movie ever. So journey is reverse Blair Witch Project.

01:34:55 ◼   ► Yeah. People do go back and play journey. I mean, do you have your white robe yet? Are you

01:34:59 ◼   ► on the path to get that? Well, no, because now I've got cheated out of a, out of an entire

01:35:03 ◼   ► playthrough because I played on Marco's account. And so then I had to create my own account.

01:35:08 ◼   ► And so I only have two play throughs on mine. Although I found like this crazy shortcut where

01:35:14 ◼   ► you can like jump to any level. And I accidentally did that tonight. And all of a sudden I skipped

01:35:17 ◼   ► half the game and I was like, well, yeah, that's the hub. That's the hub level you you

01:35:21 ◼   ► want to get the white, right? The white robe is the one thing that is game like in this

01:35:25 ◼   ► and it's to reward multiple playthroughs. You don't have to play a certain number of

01:35:28 ◼   ► times although there is something that happens as you play a multiple number of times but

01:35:31 ◼   ► you can play through once and have the white robe. Well, your robe gets all like in much

01:35:34 ◼   ► more intricate on the bottom. So right now I have like a whole bunch of intricate orange

01:35:38 ◼   ► bits. Yeah, and that keeps going until it reaches like it fills up entirely but the

01:35:42 ◼   ► white robe is not dependent on that filling up you could fill up that entire thing and

01:35:46 ◼   ► still want to have a white robe and you could do one play through and get the white robe.

01:35:49 ◼   ► Yeah, I still got to find all the little symbols everywhere because yeah, I think I could do

01:35:53 ◼   ► much better. And there's not much you know, this that's the only collectible in the game

01:35:57 ◼   ► is little symbols and like I said in the article like when you get the symbol and you're playing

01:36:00 ◼   ► with someone else they get it too. You don't steal it from them. Oh, yeah. Yeah, I think

01:36:04 ◼   ► that that's great because then you can do that's what the one part of the second time

01:36:06 ◼   ► I played through the person was showing me where everything was and helping me get them

01:36:10 ◼   ► so we like both had the same length scarf and that was just fantastic and just tonight

01:36:14 ◼   ► I found like a little new person and I was like, come with me.

01:36:17 ◼   ► I know where these things are.

01:36:18 ◼   ► Yeah.

01:36:18 ◼   ► And then once you, once you have those long scarves, like when you get into

01:36:21 ◼   ► the levels that have danger, then it's like, I mean, even though the stakes

01:36:24 ◼   ► are like, like you said, you can't die.

01:36:25 ◼   ► You don't want to lose your scarf.

01:36:26 ◼   ► Right.

01:36:27 ◼   ► Yeah.

01:36:27 ◼   ► And you don't want your new, you want to protect your new person.

01:36:29 ◼   ► If there's someone new with you, or if you have a long scarf yourself,

01:36:31 ◼   ► you're like, look, I, I worked for the earlier part of this game to get this.

01:36:34 ◼   ► I don't want it to be taken from me.

01:36:36 ◼   ► And then, so exactly.

01:36:37 ◼   ► So what do you think of the ending part where you're in the snow and going up

01:36:40 ◼   ► the mountain and everything and how that ends is that I thought I died. I was like, No,

01:36:44 ◼   ► I thought I don't know. I thought I failed the first time I played I was I was alone.

01:36:49 ◼   ► And it was cold and windy. And like, no, I'm like, that's it. It's over. I need to do something

01:36:54 ◼   ► again. I need to go back and do it again. But then you know, I started flying and I'm

01:36:57 ◼   ► such a spaz sometimes and you know, when you go up after you die in the snow and or whatever

01:37:03 ◼   ► happens to you in the snow and you go up to the mountain, I crash right into the side

01:37:08 ◼   ► of the mountain, I'm stuck under the mountain and I'm jumping. Oh, it was horrible. I'm

01:37:12 ◼   ► like, this is supposed to be my moment. And I'm like, I'm all caught underneath the mountain.

01:37:17 ◼   ► So the for the story part of the game, were you? Were you on board with that? Were you

01:37:21 ◼   ► paying attention to the cutscenes and the first or second playthrough and like deciphering

01:37:25 ◼   ► the story and then you know, connecting it to the ending?

01:37:27 ◼   ► Oh, yeah, definitely. And like how all the little markers are all really graves and you

01:37:32 ◼   ► know, all this stuff and I liked how just like we talked about in the incomparable where

01:37:36 ◼   ► know, you when you have a companion with you, it shows it on the wall that there's someone

01:37:40 ◼   ► else with you or you're alone in certain parts. And yeah, I thought it was I thought it was

01:37:45 ◼   ► a really nice story that it was simple. It was uplifting, you know, like it, it moved

01:37:49 ◼   ► you along and I liked it a lot. Yeah. For my one playthrough that you know, you only

01:37:55 ◼   ► have one chance to make the first playthrough. I'm so happy that I happened to that I played

01:37:59 ◼   ► alone in the beginning to get the feel of the game and then found someone that had a

01:38:02 ◼   ► sort of experience with them in which I disappointed them but that we did end up going through

01:38:06 ◼   ► the light together in the end. You know, where like you get up to that final sort of snowy

01:38:10 ◼   ► part where it's quiet and your scarf is gone and the light is right there, the thing you've

01:38:15 ◼   ► been trying to get through the whole time. Oh, so you had someone going with you? I like

01:38:18 ◼   ► keep waiting at the time. I waited at the top the last game and... Yeah, that's the

01:38:22 ◼   ► thing. Waiting at the top, you're like, "Is that person going to come?" Nobody showed

01:38:26 ◼   ► up! I was like, "Beep, beep!" Like, you know, beeping my little... And no one's coming,

01:38:30 ◼   ► because I keep crashing into the mountain when I get up there, so I'm always behind.

01:38:33 ◼   ► Yeah, I've waited at the top several times for people.

01:38:37 ◼   ► Sometimes they show up, sometimes they don't.

01:38:38 ◼   ► But the first time we kind of like—I mean, the first time, you know, you don't know

01:38:41 ◼   ► what's going to happen in that snowy thing.

01:38:42 ◼   ► And I was with the person going up the snowy thing, and we both pitch over into the snow.

01:38:46 ◼   ► I thought, like you, that that's the end of the game, because, like, you know, I mean,

01:38:49 ◼   ► I don't know if you saw the ending of The Sopranos, but all sorts of things.

01:38:52 ◼   ► And that's one way it could end.

01:38:55 ◼   ► But that wasn't the end.

01:38:56 ◼   ► Like, you know, you were in—you end up in that other place, and for me it was with that

01:39:02 ◼   ► of the person and we sort of swoop to the top of that mountain together joyously, not

01:39:07 ◼   ► crashing to the side of the mountain, luckily.

01:39:08 ◼   ► Yeah, because I listened to you and you were saying that, "Oh, this is your freedom moment,"

01:39:13 ◼   ► and you get to play around. And I'm there crashing to the side of the mountain, stuck

01:39:17 ◼   ► underneath this snowy bank, and I'm walking. I'm like, "This is so pathetic. Everyone

01:39:21 ◼   ► else is flying. It's like happy music, and the scarves are everywhere, and the friendly

01:39:26 ◼   ► little carpet dogs are like woof woof woof," and everything's happening, and I'm stuck

01:39:30 ◼   ► under this mountain. I'm like, what kind of a hell gamer am I? It's just terrible.

01:39:34 ◼   ► The carpet dogs. I always call them dolphins, but since you have hops around, I can see

01:39:39 ◼   ► how they might look like carpet dogs to you. Yeah, well, they're all loyal and they come

01:39:42 ◼   ► by, especially that sad one in the snow. Oh, it kills me. So sad. It's like, "I'm so cold

01:39:47 ◼   ► and you can't take it with you." Yeah. Terrible. Well, so I'm sad that your first playthrough

01:39:52 ◼   ► wasn't everything that it could be, but like you can't control. Well, I guess I wanted

01:39:55 ◼   ► it to be so much because you're like, "It's going to be so good." And you know, I had

01:39:59 ◼   ► all these expectations. And I'm like, Really? John likes this game. It's just kind of like

01:40:02 ◼   ► a funnel game. That's what I call them. They just you just want to let go of the controller

01:40:05 ◼   ► and let it happen to you. And you're like, this is right, but you need to be engaged

01:40:09 ◼   ► to not like you need to not be thinking that's a possibility because there is no controller.

01:40:12 ◼   ► What do you mean to control you're walking through a desert? There's no controller like

01:40:14 ◼   ► you need to get into that headspace. Yeah, appreciate it. But I'm glad you haven't said

01:40:18 ◼   ► and that's why I said like in the episode, it was titled one of our past atps was a journey

01:40:22 ◼   ► would be wasted on Marco or something like that. Oh, yeah, he would just be like, this

01:40:26 ◼   ► was why did I just do this? He's like, he would be like, Why did I just play this game?

01:40:30 ◼   ► Because he would never get into that headspace like it like, you know, you're getting into

01:40:32 ◼   ► it on second playthroughs and now you're having like the actual experience of the game or

01:40:35 ◼   ► whatever you have to you have to be there in the game and you have to like be absorbed

01:40:41 ◼   ► in the story and be appreciative of a pretty desert and a cute little frozen cloth thing

01:40:46 ◼   ► and the scary dragons and like, if you're never going to get into that space and you're

01:40:50 ◼   ► always just think I'm sitting on a couch holding a controller, you'll never enjoy the game.

01:40:54 ◼   ► Yeah. And you know, honestly that I would have been, I would have just stopped right there if

01:40:57 ◼   ► it were my first playthrough and that's what then, but I did all my homework. I, you know,

01:41:02 ◼   ► listened to the uncomfortable. I read your article and that, that sucked me back in. So

01:41:05 ◼   ► it ends up, uh, I really do like the game. So what are we playing next? Huh?

01:41:10 ◼   ► Yeah. I was super sad when, when it seemed like you didn't like it, because I'm like,

01:41:13 ◼   ► Oh, all these other recommendations I had for her now, I'm not even going to give them because

01:41:17 ◼   ► if she didn't like, if she didn't like journey, it's clear, it's clear our taste in games differs.

01:41:21 ◼   ► No, it's like, you know, some people don't like the same type of games and you're unlike Marco

01:41:25 ◼   ► It's totally clear that you're a gamer

01:41:26 ◼   ► But maybe you just don't like the same type of games that I do or you know