PodSearch

ATP

33: A 30-Minute Skip Button

 

00:00:00   Who's got a fan on over there? Is that you, Casey?

00:00:02   Yes.

00:00:02   Uh...

00:00:03   [HONK]

00:00:03   Oh my god.

00:00:05   It's October! You still need a fan? This is how you can tell you're in the South.

00:00:08   Yeah, he lives in like a swamp down there.

00:00:10   Oh, that is so not true, you big jerk.

00:00:12   It was so humid in July.

00:00:14   Yeah, in July.

00:00:15   [laughter]

00:00:17   No, actually, all kidding aside, we're going through a little bit of a heatwave.

00:00:19   Although I will say, and this is just between us and whoever's listening on the livestream,

00:00:25   I definitely sweat to death and turn my fan off when I record any other podcast because

00:00:30   I don't have confidence that any of them will be able to pull out my fan like you can.

00:00:36   That might have sounded naughty.

00:00:37   Oh, so this is like when the guests get to sit on the good furniture, but the family

00:00:40   doesn't.

00:00:41   Right, exactly.

00:00:42   Our podcast, turn off the fan for our podcast, put it on everyone else's, make their podcast

00:00:46   worse.

00:00:47   How about that?

00:00:48   Well, no, but because I know that Marco can strip it without issue.

00:00:50   I had to edit out John's Crickets a few episodes ago.

00:00:53   Yeah, that I can't actually help.

00:00:55   You can't turn the crickets off for the show.

00:00:57   No, I wish I could turn the crickets off, but they do not have off switches, and they're very hard to find.

00:01:01   [

00:01:04   How's the review, John?

00:01:05   Just dandy.

00:01:06   Did you see the rumor that they're gearing up for a late-October release of Mavericks?

00:01:12   I was so excited when I saw the Apple Insider post that said there was a new build-out that wasn't developer preview,

00:01:18   but it was typical Apple Insider sounding exciting, but really, nothing there.

00:01:22   So I was sad.

00:01:24   I thought today might have been the GM.

00:01:27   Nah.

00:01:28   I mean, what do you think will get the GM, like a week ahead of time maybe?

00:01:32   If that?

00:01:33   Uh, probably.

00:01:34   I don't know.

00:01:35   I mean, the GM, like, that matters for developers, and it matters a little bit for me, but at

00:01:43   this point I don't think they're changing anything, they're just fixing bugs.

00:01:46   That's good.

00:01:47   I mean, if they're aiming for a release, quote, "this fall," well, as we record this, it's

00:01:52   October 3rd, so we're getting into fall pretty significantly, and they shouldn't

00:01:57   be making any noticeable changes at this point. They should just be fixing things. We will

00:02:02   see. Yeah, I mean, you would expect at a certain

00:02:04   point they're going to announce the date for the iPad event, too, right? And whether

00:02:07   those are—they could do that at Maverick stuff at the same time, they could do the

00:02:12   Mac Pro stuff, or they could just do an iPad event and then do a separate announcement.

00:02:16   But anyway, those are all coming this month, right, I would imagine.

00:02:20   Did you see, I don't know when this was posted, it might have been more than a week ago, but

00:02:25   here I'll paste this in the chat.

00:02:29   This is a Geekbench result from what appears to be the new Mac Pro.

00:02:35   Running a CPU that I didn't even think to look at, the E5-1 series, so it's the 1680

00:02:42   V2.

00:02:44   This is it.

00:02:44   This link is it compared to my Mac Pro, the 2010 3.33 gigahertz,

00:02:49   which is the fastest at a lot of things that

00:02:51   aren't extremely parallelizable.

00:02:54   And it does substantially better.

00:02:56   What's interesting, though, is-- and I looked around.

00:02:59   I tried to find earlier.

00:03:00   I tried to find other Geekbench results from the other Xeon

00:03:06   entries that are probably going to be in the Mac Pro.

00:03:08   And this was by far overall the best

00:03:12   with single-threaded stuff.

00:03:13   it came very close to the best with multi-core. It looks like a really, really nice CPU. And

00:03:20   it's about a thousand bucks from Intel, so it's probably going to be, I don't think this

00:03:24   would be the base CPU, but this might be like a plus 500 or plus a thousand dollar option.

00:03:29   So this could be good. Now if you two, well when you two buy the

00:03:34   new Mac Pro, are you planning on just going, I don't know, full arm-end on it and just

00:03:39   maxing this thing out? Like what did you do with your current boxes?

00:03:42   Well, what's interesting about this one--

00:03:45   and this, I believe, was a little bit

00:03:47   of the case the last generation, but not nearly as much--

00:03:49   Intel is really hitting thermal walls here.

00:03:52   Like, they've shrunk their process a lot,

00:03:54   but they haven't really been able to really destroy

00:03:58   single-threaded performance compared

00:03:59   to previous generations.

00:04:01   So that's why they just keep adding cores.

00:04:03   However, with these new Xeons, the whole new E5 V2 line,

00:04:07   which is what the Mac Pro is probably going to be exclusively

00:04:10   using. Actually, I think it has to be exclusively using. That whole line, if you look at the

00:04:17   core counts and the clock speeds, as the core count goes up, the clock speeds go down, because

00:04:22   they're limited by thermal capacities of the surrounding enclosure and everything.

00:04:25   So these all have the same TDP. And they can all do this turbo boost thing, but with reduced

00:04:33   capacities as the core count goes up and with reduced benefit. So what you have, basically,

00:04:39   is a whole bunch of CPUs where there's

00:04:42   like five different top of the line models that all

00:04:44   have different characteristics.

00:04:45   So if you're doing more single threaded stuff,

00:04:49   MP3 encoding, or hey, everything Adobe makes,

00:04:54   if you're doing a lot of single threaded stuff,

00:04:58   you'll be better off with a higher clocked CPU,

00:05:02   even if that means fewer cores.

00:05:03   Whereas the top of the line one is going to be this 12 core

00:05:06   model that's only 2.7 gigahertz.

00:05:08   And that, if you do the multi-core test,

00:05:12   it does substantially better than the CPU.

00:05:16   This one is scoring a 24,000 on the new Geekbench,

00:05:19   and the 12-core one is, I believe, 29,000.

00:05:22   So it's not twice as fast, but it's certainly faster.

00:05:26   So if you're doing things like video encoding all day,

00:05:28   then that will be noticeable.

00:05:30   But for most things, the top core model

00:05:34   is actually not going to be fastest.

00:05:36   It'll be better off going with something like this,

00:05:38   has fewer cores, but it can boost them up higher.

00:05:42   What about you, John?

00:05:43   Well, I would get it maxed out if I could afford it, but I'm imagining that the top

00:05:47   end one of these, just like the top end of most macros, is going to be, you know, silly.

00:05:52   Silly pricing.

00:05:53   But nobody—even if you can afford it, you feel like you don't want to buy it because

00:05:55   you're like, "Come on.

00:05:56   I'm going to spend as much money as a car on this little black trash can on my desk."

00:06:00   So yeah, I'm looking for that sweet spot of, like, you know, maybe I'll splurge in

00:06:05   in one particular area, but I basically want some kind of balance of performance.

00:06:09   So normally I was forced to buy something close to the top of the line because I always

00:06:13   wanted the best video card they had.

00:06:15   But with a BTO, they usually let you configure that.

00:06:17   And I don't think... the only time I ever got the fastest, fastest CPU was when the

00:06:21   original Power Mac G5 came out, and I got the top of the line one of those, because

00:06:25   it was reasonably priced.

00:06:26   And so, you know, get the fastest one you can get.

00:06:28   You know, it was 2 gigahertz or whatever it was.

00:06:30   And then, you know, 3 gigahertz was coming in a year, they said.

00:06:32   So here we are in 2013, and we're looking at the Geekbench score of a 3 GHz Apple processor.

00:06:39   So yay, IBM finally didn't ever deliver.

00:06:43   Ever.

00:06:44   Yeah.

00:06:45   But no, I'll be looking at the same thing Marco was.

00:06:47   I'm probably going to pick a system with a higher clock speed and fewer cores, because

00:06:52   I think that will be better for the things I do on the computer.

00:06:55   At this point, it's not like it used to be where it's like, well, you get a single CPU

00:07:00   or a double CPU, and you have to worry if you need that second one.

00:07:03   The one with fewer cores has eight cores in it, and 16 threads because of hyperthreading.

00:07:10   I usually tended to go multicore in the old days, because I'm like, "Look, this Unix operating

00:07:15   system has tons of processes running all the time.

00:07:18   I don't want all these little background demons and stuff, and even just in a single processor

00:07:21   multithreaded waiting around for cores.

00:07:23   If I could get some with more cores, I'll let them grind away.

00:07:26   Plus, there's multiple processes running all the time.

00:07:28   want to have multiple cores so there's less waiting for CPUs. And even though they're

00:07:31   not stressing the CPUs, just hey, more cores to pass around. But everything has tons of

00:07:36   cores now. So the only reason to go to the super multi-core ones is I guess if you have

00:07:40   an application that uses them, like some kind of application that's massively, takes advantage

00:07:45   of every core you have, like you get almost linear scaling if you double the number of

00:07:48   cores. But I don't use any of those applications, so I'm going to be looking for, maybe I won't

00:07:54   buy the fastest, fastest, like low number of cores processor because they might charge

00:07:58   a premium for that too. This has a lot to do with how Apple prices things. But until

00:08:03   I see the prices, maybe I'm getting the cheapest one because the cheapest one is going to be

00:08:06   five grand and then, you know, that's what's going to happen.

00:08:09   Right, yeah. I mean, the pricing is still such a big question mark on these. We have

00:08:13   no idea what the entry price will be, what the CPU upgrades will cost, anything. I mean,

00:08:18   we don't even have a ballpark except that we can assume that it's probably not going

00:08:22   be that much different than the previous Mac Pro, maybe a little more to cover those graphics

00:08:26   cards, but I'm still betting entry price of $3,500, which is higher than the current ones,

00:08:34   because I think they do have to cover those graphics cards with a healthy margin.

00:08:38   And certainly the U.S. Assembly is going to increase their cost a little bit, but I don't

00:08:45   think it's going to be enough relative to the profit margin of this product to make

00:08:49   make any difference in the retail price. But I don't know.

00:08:53   I still think they can put the bottom of the line pretty cheaply. Intel charges 500 retail

00:08:58   or whatever for their thing. Apple's not getting that price. And that's probably the most expensive

00:09:03   component in the system. Maybe the video card gives it a run for its money. There's not

00:09:07   that much else in the system, just in terms of physical goods and the amount of stuff

00:09:11   in there. If you were to build your own PC with this processor, the processor would definitely

00:09:16   be the most expensive thing in the system. And then you'd buy a retail gaming video card

00:09:21   and a motherboard and a case and you'd slap it together and you'd do that good old PC

00:09:24   Mac price comparison. You could see, what is the real price of this stuff? Take that

00:09:30   home-built PC cost and then cut a whole bunch off that. That's Apple's actual price. So

00:09:34   I think there is room for Apple to get a healthy 45-50% margin and still offer this machine

00:09:41   entry level under 3K.

00:09:43   Yeah, we'll see. I mean, really, those GPUs are such a big question mark, because at retail,

00:09:48   of course, they're very expensive, but they're also now going to be like a half generation

00:09:52   or one generation old by the time we actually get them, right?

00:09:55   Yeah, they pre-announced the Mac Pro, and it's like, "But no, now the video card is

00:09:58   crappy. AMD announced their next-gen GPUs. We waited so long for this great machine,

00:10:03   and we don't even have it in our hands, and the GPU is already, you know, AMD's already

00:10:07   got their next-generation architecture." I'm like, "Oh, we've got to buy this previous-generation

00:10:11   architecture in my brand new Mac Pro.

00:10:14   One thing I'm curious, though-- and I admit I don't know enough

00:10:17   about the various low-level workings of the motherboard

00:10:20   and the bus and the chipset and everything.

00:10:23   Is there much of a chipset?

00:10:24   Or is it all in the CPU now?

00:10:26   Anyway, one thing I noticed, though,

00:10:31   looking on Geekbench, looking at the scores,

00:10:33   comparing the Mac Pro to other CPUs that

00:10:37   should have similar scores, the Mac Pro does a lot better.

00:10:41   And I'm wondering, we know that the new Mac Pro is designed

00:10:45   in this crazy way to have-- to basically use

00:10:48   every possible PCI Express lane, to use

00:10:50   every possible amount of bandwidth through that CPU

00:10:52   and through the chipset.

00:10:54   I wonder if they've been able to make some tweaks because they

00:10:58   can assume there's going to be nothing else on the bus

00:11:00   besides what they have in there, stuff like that.

00:11:04   Is there enough headway for them to tweak

00:11:06   that to make their systems like 10%, 15% faster

00:11:10   than you just buying a Supermicro server and sticking a Z on it? Is that even possible?

00:11:17   I don't even know. I'd love to hear from listeners who know.

00:11:19   I would think that they would definitely make it 10 to 15 percent cheaper, because as you

00:11:22   said, they don't have to worry about internal expansion pretty much at all.

00:11:27   They do have to cover the cost of a very custom-designed motherboard.

00:11:31   Yeah, I know. But I don't think it's outside the expertise of the people who make motherboards.

00:11:38   It's not a standard size or shape, but I don't think that's the, you know, probably the stupid

00:11:42   round case and that triangular piece of metal thing is more expensive than just getting

00:11:46   motherboards cut to the look.

00:11:47   They're rectangular shapes, it's not like they're, you know, and they do custom motherboards

00:11:50   for all their laptops anyway with those crazy shapes to work around the components.

00:11:54   So this is a much, this is relatively a freer environment.

00:11:57   Look how much room!

00:11:58   Like, think of what Apple has to stick motherboards into.

00:12:01   Laptops where there's not a millimeter to spare and everything is like, every capacitor

00:12:05   is laid out, so it just barely clears the little post that's holding the screw that

00:12:08   holds the case together. The Mac Mini, which is tiny. Phones and iPads, forget it, there's

00:12:13   no room in there to breathe. The iMac, which is relatively luxurious, but they keep squeezing

00:12:17   it thinner and thinner, and it's just laptop parts in there anyway. And then finally the

00:12:21   Mac Pro, where you can have actual rectangles. Three of them! Three actual rectangle printed

00:12:26   circuit boards. But yeah, the lack of any other expansion in there, I would imagine

00:12:31   would allow them to just trim every single component that's not absolutely necessary

00:12:35   to run what they know will be in the box. This machine, this new Mac Pro, is designed

00:12:40   electrically and physically a lot like a couple generations ago game console, albeit with

00:12:45   much more expensive components, right? But, you know, it's got, it's trying, you know,

00:12:50   one fan, you're trying a game console to have one fan to cool the whole thing, no internal

00:12:54   expansion, everything's all just like shoved in there. There's a CPU, a GPU, memory,

00:12:59   you know, I guess the internal hard drive flash replaced with SSDs.

00:13:05   It's like a big tubular game console that costs tons of money and probably doesn't run games that well.

00:13:10   Now do you think that it's going to, that the fact that it's going to be made in the US is going to have any

00:13:17   empirical difference on the price?

00:13:19   Nah, I don't think so. I mean,

00:13:22   For US customers anyway, the extra amount they have to...

00:13:27   Well, it really depends on what's shipped where.

00:13:29   Like you say, "Oh no, they don't have to ship from China to me."

00:13:35   But they have to ship probably every component that's in the thing from China to Kentucky

00:13:39   or wherever the heck the thing is built, and then ship the finished product from Kentucky.

00:13:42   It's being assembled in America, but the places that the components come from.

00:13:47   And in terms of assembly, this is probably an easier thing to assemble.

00:13:51   They're not machining aluminum and jamming little pieces in there and everything.

00:13:54   It's just a metal frame, slap the printed circuit boards on it, make a plastic case,

00:14:00   a fan on top, a little power supply, slap it all together, you're done.

00:14:03   Right, and the tolerances amongst those major components are a lot better.

00:14:08   I mean, obviously the tolerances within a component are just as bad as always, but like

00:14:12   you were saying earlier, it's a lot harder to squeeze a bunch of stuff in a phone than

00:14:15   it is in the trash can.

00:14:17   Yeah, plenty of room in there.

00:14:19   Although I was having waking nightmares while I was idly thinking earlier today about

00:14:24   What will the fans sound like when I'm playing a game on that thing?

00:14:28   Because I know when I play on my current Mac Pro

00:14:30   it's the only time I ever hear the fans is and it's because the fans in like

00:14:33   Section A9 of the 15 wind sections that are in a Mac Pro start cranking up

00:14:38   And normally you don't hear them at all because the GPU is doing nothing

00:14:41   But you know start playing a 3D game for an hour or so and you hear this whine

00:14:44   And it's not the giant CPU fans the it's the smaller ones that are blowing across the video card area plus

00:14:50   Also, the you know the actual active cooler that's on the video card itself

00:14:53   And so I'm like well, that's kind of noisy

00:14:57   Well this new one be better because there's only one fan or it will be worse because the only alternative when things get hot in

00:15:03   There is just to crank up the one fan they have to even higher speeds well the one you have now

00:15:07   Is that the 8800 with the stock fan?

00:15:09   Yeah, yeah, so I'm very familiar with that fan that is by far the loudest part in that computer

00:15:14   Even at idle speeds, eventually mine got, I guess, a little bit of dust in it somewhere,

00:15:19   and so it became a little bit louder. And even at idle speeds, like, I would take the

00:15:24   lid off and stick my finger on the hub so it would stop that fan. It was a massive difference

00:15:29   in noise. Not to the point where you'd have to replace it, but to the point where you'd

00:15:33   notice it and you're like, "Wow, that's kind of inelegant. The little fan is so loud."

00:15:36   Yeah, I had that on my G5 where that fan, it started to go bad. And when they started

00:15:41   they start to go bad, you know, it's just like nails on a chalkboard, you gotta get

00:15:44   rid of it. So I bought an aftermarket cooler to put in there. It was probably actually

00:15:48   noisier than the old one, but so far on this 8800, for many years of service, it has not

00:15:54   started to pick up that telltale little scritch that you know is just gonna eventually get

00:15:58   louder and louder. And for a video card fan that I assume you leave your computer on most

00:16:02   of the time or all the time? Sleeps at night. Okay, that's pretty good, but still, for a

00:16:06   For a video card fan from 2008 until now, that's pretty good.

00:16:12   Those things always die.

00:16:14   This is a champion machine.

00:16:16   This has had the fewest hardware problems of any Mac I've ever owned, including all

00:16:19   of my old classic ones.

00:16:21   Yeah.

00:16:22   I'm looking forward to seeing what happens with the new one.

00:16:27   If it comes out…

00:16:28   I'm curious to see benchmarks.

00:16:29   I'm curious to see when Barefeet gets a hold of it and tests it against the other

00:16:35   Max and against other models within itself.

00:16:38   I really wanted to see how these CPUs perform in real world use

00:16:42   or with real benchmarks once we have the retail machines

00:16:44   and see what we can do.

00:16:46   What's really going to be telling to me

00:16:48   is whole system performance.

00:16:50   Like the PCI SSDs, are they super fast SSDs,

00:16:55   or are they just like, nah, the SSDs are fine.

00:16:57   Because it's going to feel fast if when you launch the thing

00:17:00   and you click on System Preferences

00:17:02   and it runs all those preference panes or whatever,

00:17:04   "Wow, I can notice this is faster."

00:17:06   Is it going to feel faster than an existing Mac that's all SSD?

00:17:10   Or will it just feel like, "Yeah, it's about the same"?

00:17:13   What's crazy is, I don't even think it has a serial ATA controller.

00:17:17   - No, why would it? - Exactly.

00:17:19   There's so much stuff that by getting rid of all the internal expansion,

00:17:22   there's so much stuff that it just doesn't need to have.

00:17:25   And it can dedicate tons of bandwidth to the stuff it does have,

00:17:30   if it can use it, and then just have three Thunderbolt controllers

00:17:34   and be like, "All right, done."

00:17:36   I don't know, could be interesting.

00:17:37   - What's up with the JPEG decompress result

00:17:39   in this benchmark?

00:17:40   - I saw that too.

00:17:41   - Oh yeah, look at that, I don't know.

00:17:43   I have no idea.

00:17:44   But who knows?

00:17:46   - Dude, did you see like the,

00:17:48   this is kind of like testing the A7 and text benchmark,

00:17:52   or the AES multi-core score is like twice as high.

00:17:55   - Yeah, exactly, 'cause it's harder to accelerate.

00:17:57   - Because they had a hardware instruction.

00:17:59   That's the best way to win benchmarks.

00:18:00   - Oh yeah.

00:18:01   - Add a hardware instruction

00:18:02   for whatever it is that they're benchmarking.

00:18:03   "Wow, it's ten times faster!"

00:18:05   And one of the reasons why I keep looking at these benchmarks is like,

00:18:09   even though my current computer has a three or even probably four year old CPU by this point,

00:18:17   this actually isn't that much faster than it in single core stuff.

00:18:22   And so I'm kind of like, I don't want to buy a whole new Mac Pro and go through all that expense

00:18:29   if there's not going to be a retina display to go with it, and therefore, you know,

00:18:32   another whole reason to get it.

00:18:33   If there's not gonna be the retina display

00:18:35   and there's not gonna be a massive upgrade

00:18:38   in CPU performance available.

00:18:40   Like it's kind of stupid to have a three year old,

00:18:42   to have a three and a half year old CPU

00:18:45   that's almost as fast as what you're about to launch today.

00:18:48   But that just shows how little progress

00:18:49   Intel has made with the Xeons.

00:18:51   - You've already got a PCI Express SSD, a big one, right?

00:18:54   So like you should definitely look at, you know,

00:18:56   does this, if we put a black sheet over your desk

00:19:00   didn't tell you if you're using your old Mac Pro or your new one and you can't tell the

00:19:03   difference then maybe wait for the second generation. I kind of wish I could wait for

00:19:06   the second generation system, you know, especially knowing all the things that we know now about,

00:19:11   well maybe the retina displays won't be ready yet and the new GPU architecture from AMD

00:19:16   is out and those architectures don't change every year, you know, so it would be nice

00:19:21   if I could wait for a second, but I can't wait, I mean I've had this computer wait too

00:19:23   long.

00:19:24   Exactly.

00:19:25   Now John, how old is the one that you have at work, because you have a Mac Pro at work

00:19:29   as well, don't you? Yeah, I've got, I don't know, I think it's around five years old,

00:19:34   but it was the first, it was, what do you call it, the Intel chip whose name starts

00:19:38   with an N that I can't pronounce. The Nihalad? Yeah, the first one with integrated memory

00:19:43   controller. It's the single CPU socket version of those, with the stupid one with triple

00:19:49   channel memory and four RAM slots. And I got it because it was the cheapest Mac Pro that

00:19:54   was available at the time, and so I, you know, give work break and, you know, instead of

00:19:58   Instead of getting a $300 Dell crap box, they're going to buy me this super expensive, you

00:20:02   know, whatever it was, $2,000 Mac.

00:20:04   But the replacement cycle for machines at work, I think now is like 18 months or something

00:20:09   for the crappy Dell laptops that everybody gets.

00:20:12   And I've had this machine for five years, so I feel like I really—they got their money's

00:20:15   worth.

00:20:16   Oh, yeah.

00:20:17   And that's what I was going to ask is, are you replacing both home and work, or do you

00:20:20   think you're going to do one and not the other?

00:20:22   I don't have any control over what happens at work.

00:20:23   I have put an SSD in the one at work, so like it's, you know, I kind of got a new machine

00:20:27   when I put in that, you know, it was like a 500 gig SSD or something. So I'm not hurting

00:20:31   at work for, you know, for CPU, but I'll definitely get one at home and see how it goes. And then

00:20:37   see how much at work I might even just ask for an iMac next or something. I don't know

00:20:41   if I really need to have the Mac Pro at work.

00:20:43   Do you want to tell me about something that's awesome?

00:20:46   I would love to. This episode is once again brought to you by our friends at Squarespace.

00:20:54   Squarespace is the all-in-one platform that makes it fast and easy to create your own

00:20:58   professional website or online portfolio.

00:21:01   For a free trial and 10% off your first purchase, go to squarespace.com and use offer code ATP10

00:21:07   because now we're in the month of 10, October, so ATP10 for 10% off.

00:21:13   Squarespace is always improving their platform.

00:21:14   They have new features, new designs, and even better support all the time.

00:21:18   Their designs are beautiful to start with, and you can even customize them.

00:21:22   You can hack the CSS, you can inject JavaScript, you can do whatever you want, and they have

00:21:25   all sorts of stuff that doesn't require coding that you can customize as well.

00:21:29   So you can create a unique website for you or your business.

00:21:32   They've won numerous design awards, and it's incredibly easy to use.

00:21:36   If you ever want some help though, they have an amazing customer support team that works

00:21:39   24/7, with over 70 dedicated employees here in New York, just dedicated to customer support.

00:21:46   So and even they, by the way, have won numerous awards, most recently a Gold Stevie Award.

00:21:51   So Squarespace starts at just $8 a month,

00:21:54   and it includes a free domain name if you sign up for a year.

00:21:57   And every design comes with a unique mobile experience.

00:21:59   It's a responsive design that matches your overall style

00:22:02   of your site.

00:22:03   So you don't have just one generic responsive mobile

00:22:05   design for every theme.

00:22:07   Every theme has its own design.

00:22:08   Looks really nice.

00:22:09   You can check out our site, actually,

00:22:11   atp.fm to see an example of one of these.

00:22:13   We've customized it a little bit, but it's mostly stock.

00:22:15   And it's great.

00:22:17   So go to squarespace.com.

00:22:19   You can start a free trial with no credit card required.

00:22:23   You can start building your website, see how you like it,

00:22:25   never even give them a credit card just to try it out.

00:22:28   See if you like it before you pay them anything.

00:22:29   It's really great, you don't have to like try to remember

00:22:31   to cancel it before your trial's up or anything,

00:22:33   otherwise you get charged, nothing like that.

00:22:35   So when you do decide to sign up for Squarespace,

00:22:38   make sure you use offer code ATP10

00:22:41   to get 10% off your first purchase

00:22:43   and to show your support for ATP.

00:22:45   Thanks a lot to Squarespace for sponsoring the show.

00:22:47   They're everything you need

00:22:48   create an exceptional website.

00:22:49   I just saw a tweet from somebody a little bit before the show extolling the virtues

00:22:56   of Squarespace and how they loved it so much.

00:22:58   And the thing is, there was a tweet from a person who I know doesn't listen to my podcasts

00:23:03   and wasn't tweeting in response to a sponsored message from Squarespace, but it was just

00:23:07   a user of Squarespace who was a non-technical person who wanted to have a website.

00:23:11   And he said, "Oh, Squarespace, I love you so much.

00:23:13   Everything is so easy."

00:23:15   That makes me think that it's not, you know, like, we hear about Squarespace and lots of

00:23:18   podcasts and they do sponsor a lot of podcasts and advertising and other venues or whatever,

00:23:22   but regular people who use the product also actually like it.

00:23:26   That helps.

00:23:27   Absolutely.

00:23:28   Going back a second, before I forget, first of all, there's a good question in the chat

00:23:33   room that I want to get to in a second also, but going back to the Mac Pro for a second,

00:23:37   the reason I asked earlier what video card you had and everything, I forgot to actually

00:23:40   finish this point, but have you heard the fans spin up on the Retina MacBook Pro?

00:23:48   Probably in the office where everything's noisy, but not in terms of I'm using one

00:23:53   and I hear it because I've never really used one for an extended period.

00:23:56   So that was the one where they first did their asymmetrical blade thing.

00:24:00   And it's actually really, really good.

00:24:03   It's very quiet, even when you'd expect it not to be.

00:24:07   And then when it does ramp up to full speed, you can hear it, but it almost sounds like

00:24:11   white noise, or like pink noise, one of those various expensive office noises.

00:24:16   It sounds kind of like air whooshing, not as much like a whirring high-pitched noise.

00:24:21   It's hard to describe it, you really have to hear it.

00:24:23   But it sounds better, and it is less noticeable than a fan of a regular design spinning at

00:24:31   roughly the same speed.

00:24:33   So I think the new Mac Pro, it also has that same design.

00:24:38   I believe they mentioned that specifically.

00:24:39   It also has the asymmetrical blades

00:24:41   and it's one giant slow fan.

00:24:42   - You sure about that?

00:24:44   - I think so, I'll have to double check.

00:24:45   - Chat room can go look it up for us.

00:24:47   - Yes, please do.

00:24:48   (laughing)

00:24:49   Anyway, but if they do the same thing,

00:24:51   I would imagine that having one giant fan

00:24:55   that has the capacity to cool two cranking GPUs

00:24:59   and one cranking Xeon up to 12 cores,

00:25:02   plus the power supply to the whole thing. That's a lot of capacity. So if you're

00:25:05   just stressing part of it, like just the CPU, I would imagine it doesn't have to spin

00:25:09   that fast to cool that adequately. And so I would guess it would probably be pretty

00:25:15   quiet.

00:25:16   Now was that the, it wasn't the retina MacBook Pro that you and I, well really you, stole

00:25:20   from Jason Snell and ran the...

00:25:23   Yeah, yes, that was it.

00:25:24   Okay, because that I heard, I was there with you, this was WWDC last year, and you're

00:25:30   absolutely right. You can hear them for sure, but it sounds a lot less offensive, which

00:25:35   is weird. And if I was listening to this and hadn't heard the fans on a Retina MacBook

00:25:38   Pro, I'd think that we were all crazy. But it really is different, and it really does

00:25:42   make a difference.

00:25:44   Yeah. Oh, and John Solo in the chat had just confirmed from Apple's Mac Pro promo site

00:25:50   that it does have asymmetrical spaced blades.

00:25:52   They're impellers, not propellers.

00:25:54   Yeah, it's this weird giant thing anyway. So yeah. And then the question that I wanted

00:25:59   to get to from a few minutes ago is from Brad hyphen hyphen colon in the chat room. He said,

00:26:04   "Are you," the letter U, "supposed to put the Mac Pro on top of the desk or on the floor?"

00:26:10   The new one. So what do you think? That's actually a good question. I don't know what

00:26:13   it is.

00:26:14   You're supposed to put it on top of the desk.

00:26:16   Right. But I mean, everyone who's had a Mac Pro up until now puts it on the floor.

00:26:19   Right. Because the Mac Pro is supposed to go on the floor. This is another thing I was

00:26:22   thinking about when I was thinking about a Mac Pro. As I reached down to plug in my podcast

00:26:26   microphone, I plug it into the front of my Mac Pro and I say, "Well, I'm not going to

00:26:29   be doing that anymore."

00:26:30   Yeah, there's no more front ports.

00:26:31   Right, but that's why it rotates, but that makes for an interesting demo, but you can't

00:26:37   really rotate the thing with cables sticking out of it, and they all yank.

00:26:39   I don't know.

00:26:40   All the dust comes forward.

00:26:42   It's like the diagram of Mac Pro 2010, where it shows our big tower, and then it says Mac

00:26:46   Pro 2013, and it shows the little trash can with a million peripherals hanging off of

00:26:49   it with a bunch of cables.

00:26:50   Yes.

00:26:51   Please stop sending that to us.

00:26:52   We've seen it a million times.

00:26:53   But anyway, I'm thinking I already have a USB hub attached on a Mac Pro.

00:26:58   I have everything attached on a Mac Pro.

00:26:59   But yeah, I'm going to have to use a USB hub or something, because I won't be able to plug

00:27:03   something into the front of it anymore.

00:27:05   For putting it on the floor, if you put it on the floor, it would look like you had a

00:27:12   trash can.

00:27:13   I think it's too small and lonely to be on the floor by itself.

00:27:16   Imagine if you put a Mac Mini on the floor.

00:27:18   Yeah.

00:27:19   That would look ridiculous.

00:27:20   It's like, "Aw, what's it doing down there?"

00:27:21   And also, it's got little vent holes on the bottom.

00:27:24   And so if you put it on-- I think it does, right?

00:27:26   Yeah, it does.

00:27:27   And I want to put that on a carpet.

00:27:28   Like my Mac Pro has got the little feet,

00:27:30   and it's up off the carpet.

00:27:31   I really don't want to put anything

00:27:33   that's going to sink into the carpet.

00:27:34   And would it even be steady down there,

00:27:36   like knock it over with my foot?

00:27:38   Right.

00:27:39   It's not big.

00:27:39   People who haven't seen it in person,

00:27:41   who it looks big in the advertising pictures,

00:27:43   it's not a big computer.

00:27:44   So I kind of like it on the floor,

00:27:46   because it gets the noise below the level of the desk

00:27:49   and out of earshot for me.

00:27:51   but this is probably gonna have to go on my desk.

00:27:52   I mean, I guess it won't take up a lot of room,

00:27:54   but it'll take up more than zero room,

00:27:55   which is what my current Mac Pro does.

00:27:56   - Right, and all the cables have to go into the back of it.

00:27:58   And you will hear the fan more than you would

00:28:00   if it was on the floor, but then again,

00:28:02   the fan is also now the only moving part

00:28:04   in the entire computer.

00:28:05   - Yeah.

00:28:06   - Are we done with Mac Pro, 'cause I wanted to--

00:28:09   - We probably should be.

00:28:10   - Yeah, can I ask the arbiter of all things follow-up,

00:28:13   will you allow me to do one brief piece of follow-up

00:28:16   so we stop getting spammed?

00:28:18   - It depends on what it's about.

00:28:19   - It's what you got spammed about.

00:28:20   about the scrubber? Oh my god, we get this, we've been getting this like every hour.

00:28:24   We? Who's we? I've been getting it. Oh no, well we've been getting our fair share

00:28:28   as well. Yeah. Do you want to cover this or shall we?

00:28:31   Yeah sure, no. So we talked about, I suggested a feature for Marco's podcast app last week,

00:28:37   half jokingly, about, you know, improving the audio scrubbing experience and many many

00:28:41   people wrote in to tell me that I should use the built-in feature of the regular audio

00:28:46   scrubber where you slide your finger up to and change the speed from like half speed

00:28:51   scrubbing, quarter speed scrubbing, or fine scrubbing or whatever.

00:28:54   A lot of people thought that was actually a new feature in iOS 7.

00:28:56   It's not.

00:28:57   It's been around for a very long time.

00:28:58   I think maybe even back to iOS 3.

00:29:02   And a lot of people asked, like, "Doesn't that do what you want?"

00:29:04   No, it doesn't.

00:29:05   It doesn't do what I want.

00:29:06   Like, what I was trying -- I did a bad job explaining it because it's, you know, off

00:29:09   the cuff and it's difficult to explain visual stuff like this, but the key thing that that

00:29:13   feature is lacking for when you're trying to scrub through a two-hour podcast or something

00:29:17   really long is that it doesn't change the visual feedback. Even though you supposedly

00:29:22   have more fine control, like when you move your thumb an inch, it doesn't really move

00:29:25   the little thing an inch, it doesn't change how far the little playhead moves. So in a

00:29:29   two-hour podcast, a single retina pixel would be like 10 seconds. And so how do you fine

00:29:34   scrub through 30 seconds of audio when that's represented by three retina pixels? Yes, you

00:29:40   can move your finger and you're not jerking the thumb along, but you don't know how far

00:29:43   You've gone kind of like you what I'm looking for was trying to express last time with my video game analogy is you want a

00:29:47   Connection between what you do with your finger and an immediate clear visual response

00:29:51   Showing what you're doing and a clear visual response isn't like play head doesn't move play

00:29:56   It doesn't move play it moves one retina pixel represents 10 seconds that is not a good connection between thumb and finger

00:30:02   Which is why I'm looking for something

00:30:03   But zooming or some kind of other thing where it's more like playing a video game

00:30:07   and it's you know a complete 60 frames per second very responsive experience of

00:30:12   zooming in and zooming out and zooming in when you're doing fine movement and zooming out so that the movements the visual movements are

00:30:18   Connected directly and always always, you know, you always see immediate to scale feedback of your finger

00:30:25   But the amount of time that it represents changes because we're zooming in and out of the time

00:30:29   It's very difficult to explain and the other thing I want to say about this feature is a lot of people like oh

00:30:33   That's a cool feature. I would love that or Marco should do that or shouldn't the thing about any features especially ones

00:30:37   You're just like wouldn't it be cool if

00:30:40   You can't tell until you implement them. So I'm not saying that this is gonna

00:30:43   This would be the best thing in the world

00:30:45   It could very well be that if you went and implemented it

00:30:47   It would be terrible or the first three tries would be terrible you'd find out

00:30:50   This is you know a different approach would be better like a lot of people think as soon as you know

00:30:55   You just ascribe an idea or maybe you draw something on an app

00:30:57   You're like oh just make that I know that will be good

00:30:59   You don't know until you actually implement it and sometimes especially in this case

00:31:02   Implementing it can be very difficult and complicated especially if you're not like a game programmer used to trying to do

00:31:07   you know, responsive control systems. So you really have to, not that I think Marco was

00:31:11   weighing this heavily on his mind, but you really have to know what you're in for in terms of,

00:31:14   "I'm going to try to implement this really complicated, difficult feature that may not

00:31:18   even work out. I won't know until I get it implemented and I've already wasted a week trying

00:31:21   to get it done." So that's the calculus for doing this. And that's why, despite the fact that people

00:31:26   are like, "Oh, I would totally buy this program." Remember I said there'd be like five people who

00:31:29   would buy this program if you added this feature? I was off by a factor of two of, I think,

00:31:33   ten people. So they would do it. But yeah, and each one of them thought, "Man, I bet

00:31:38   you're getting a lot of replies." Yeah, I got like ten. And so, yeah, ten people would

00:31:41   do it. But they don't know they'd like it either, because you can't even tell if an

00:31:47   app like Twitterific or something is going to be good that uses more or less simple scrolling

00:31:52   and gestures. You can't even tell that's going to be good until you implement it, let alone

00:31:55   this thing. Not that I'm saying nobody should do this, and I think it would be a cool feature

00:31:58   if someone's got to be in their bonnet about making this really cool. But I am under no

00:32:02   illusions about whether the thing I described would actually be awesome and how difficult

00:32:07   it would be to get it to be awesome. There's a reason making awesome games is difficult,

00:32:11   because you can get something working in a game engine, even that's hard enough, but

00:32:14   then it's probably terrible until you tweak it to death to try to get it just so.

00:32:18   Oh yeah, I mean, there's been so many things, like when I was making Instapaper, there were

00:32:24   so many things that I tried and threw away because I couldn't get them right. Or they

00:32:28   sounded cool in theory. Here, I've dug up this link. I'm paging the chat now. Where

00:32:36   it sounds cool in theory. Oh, let me use the accelerometer to do anti-shake on the screen

00:32:43   so that, because whenever I was like riding the subway in New York, trying to read on

00:32:48   my phone, you get jostled so much on the train that it's kind of hard to keep your eyes on

00:32:52   the screen. So I tried, oh, let me do this anti-shake thing. And I just, I just could

00:32:56   not get it right. And if you don't get it exactly right, it actually makes it worse

00:33:00   and gives you motion sickness. You have to get it exactly right as to be extremely responsive.

00:33:05   And I just couldn't do it. As far as I could tell, I don't think the hardware was accurate

00:33:09   enough to really nail it.

00:33:11   Yeah, it probably wouldn't have been. That's the other thing you might find out. Maybe

00:33:13   this feature I have in my head isn't actually possible on the hardware. Or maybe it's only

00:33:17   possible on a 5S but not possible on most of the phones people have, right?

00:33:21   And so I'll have to see.

00:33:23   I mean, that's the kind of thing I'll tell you right now.

00:33:26   I'm probably not going to make it into version 1.

00:33:31   But I have wanted to play with the scrubbers for a while.

00:33:37   And so I do want to attack that problem sometime.

00:33:40   But it's probably not important enough to put it in 1.0.

00:33:42   If that was going to be a 1.0 feature,

00:33:44   it would be the flagship feature of an app.

00:33:46   Like an app that includes that kind of thing

00:33:48   as a 1.0 feature, that would be what the app is known for.

00:33:50   for, because that's the type of thing where you would sink all your time into this, like,

00:33:53   "I'm going to make whatever it is you're going to make, and it's going to have a scrubber,

00:33:56   and my whole experience is going to be focused around the scrubber. It's going to be what

00:34:00   my app is known for, the app with the awesome scrubber." You know what I mean? And that's

00:34:04   not really what you're making.

00:34:06   Well, it might be. I talked last week, I think, a little bit about how it's challenging to

00:34:11   make the now playing screen. It's challenging to design that, because the whole rest of

00:34:17   the app, honestly, I don't care about the design.

00:34:19   Like, I'll make it work, I'll make it look good,

00:34:21   but the whole rest of the app, any kind of navigation

00:34:24   structural thing, it doesn't really matter.

00:34:26   It's probably going to be a bunch of table views

00:34:28   because it doesn't really matter.

00:34:30   What matters to me the most by far

00:34:32   is the playing experience.

00:34:34   When you're using the controls on the playback screen

00:34:37   and doing the most common actions

00:34:40   that you're likely to do when you interact with your podcast

00:34:42   app, and that's not navigation, and it's not

00:34:44   or moving shows, it's playback controls and seeking and stuff like that. So that is the

00:34:50   kind of thing that I do want to get really right.

00:34:54   And is there anything else that's going on with Overcast that you'd like to share? Anything

00:34:57   from last week? What am I forgetting? You're asking that as

00:35:02   if you have something in mind. No, no, no, no. Not at all.

00:35:05   No, I'm rewriting my... I'm rewriting half of my sync protocol tonight.

00:35:11   How?

00:35:11   You haven't even released it.

00:35:13   You're already chucking it all?

00:35:15   No, no, just the ideas are the same.

00:35:17   But I was using-- OK, so I have to figure out-- so the server,

00:35:23   it has server-side crawling.

00:35:24   I'm pretty sure I've said that before,

00:35:26   so I don't think I'm revealing anything new here.

00:35:27   It has server-side crawling, of course,

00:35:29   because that's the obvious way to do it in this day and age.

00:35:32   And there's a lot of benefits to it.

00:35:34   So I have to figure out, when the server tells the client

00:35:38   what episodes should be in this person's account.

00:35:43   Does it include-- obviously, there's

00:35:45   the list of feeds that you are subscribed to.

00:35:47   But does it include every episode that's in those feeds,

00:35:52   or just the ones that are new to you,

00:35:55   like just the unplayed ones to you?

00:35:57   So in other words, do you include all the back episodes?

00:36:00   And if you do, you can do some cool things.

00:36:03   Like you can instantly toggle over to the list of all

00:36:07   and add certain ones back, or make certain things faster in navigation and management.

00:36:11   Right after I tell you that I don't care about navigation and management.

00:36:15   It improves things there. And the downside

00:36:19   though is that there's then a lot of objects for

00:36:23   the sync engine to manage. Like, alright, you have to somehow keep

00:36:27   track and keep in sync this much larger

00:36:31   set of items as you're communicating between these two things. And so I didn't want it to take up a whole

00:36:35   a whole lot of bandwidth.

00:36:36   So I made the first sync protocol totally binary

00:36:42   on the way up.

00:36:43   So when the device communicates to the server,

00:36:47   it sent, as the request body, it sent

00:36:52   a binary stream of a whole bunch of integers, basically.

00:36:57   And it worked fine, and it was really, really small.

00:37:01   It used very little bandwidth to communicate information

00:37:03   for a lot of items, because I really did fairly intelligent

00:37:06   packing work there.

00:37:07   And this is, by the way, I tweeted a couple weeks ago

00:37:10   that I hit my 64-bit bug, where I had a struct align issue.

00:37:15   That's what it was.

00:37:16   It was in that binary protocol.

00:37:17   Are you just taking native C structures

00:37:19   and sending them over the wire?

00:37:21   No, I was packing them into an NS data, but otherwise, yes.

00:37:24   Which I learned is terrible.

00:37:26   You should use protocol buffers or something.

00:37:28   Google's got them sitting there waiting for you to use them.

00:37:30   I bet they work in 64-bit.

00:37:31   Anyway, so as I was designing the system

00:37:35   and as I was using it, it became increasingly clear

00:37:38   that it was going to be very hard to expand.

00:37:42   It was fairly error prone.

00:37:44   And it was just being a pain in the butt.

00:37:46   And so I figured, you know what?

00:37:48   This is 2013.

00:37:49   I don't need to be using a binary protocol.

00:37:51   So now tonight I'm changing it to just a JSON dictionary

00:37:54   that I run through gzip.

00:37:56   And that gets it most of the way down to the original size,

00:37:59   because the dictionary is just the same few keys followed

00:38:03   by the same few-- followed by digits 0 through 10.

00:38:07   It Huffman codes pretty well.

00:38:08   And so it actually compresses fairly well.

00:38:14   So I'm changing that part of it to just be a little bit less

00:38:18   technically cool and a little bit more functional and less

00:38:22   fragile and less prone to things like endian changes

00:38:26   being a problem.

00:38:28   All this stupid stuff I was doing on the server to interpret this that I really didn't need

00:38:31   to be doing.

00:38:32   I think you made it more technically cool, not less.

00:38:34   It is uncool to send binary data over the wire.

00:38:40   So anyway, oh, you know what?

00:38:44   If you guys don't mind, do you see Greg Minton's question?

00:38:46   I would like to address that.

00:38:47   Yeah, I actually had just added that in the show notes tonight, but I didn't know if you

00:38:52   would want to address that or not.

00:38:54   Yeah, sure, I'll address that.

00:38:55   All right, so Greg Minton's question is--

00:38:58   and he asks this via the contact form.

00:38:59   He said, do you anticipate any conflicts of interest

00:39:03   between ATP and overcast?

00:39:04   For example, would you feel resistance

00:39:06   towards implementing a potentially cool feature

00:39:08   like a more seamless ad skip button out of fear

00:39:11   that it would devalue ATP's ads?

00:39:13   OK, so out of all possible conflicts of interest,

00:39:17   I can't think of them all right now.

00:39:19   Maybe something might come up someday.

00:39:21   I don't know.

00:39:21   But I will address that one in particular,

00:39:23   because I have thought it would be really cool, since I'm going to have server-side

00:39:27   infrastructure, it would be really cool to do basically the Amazon Kindle popular highlights

00:39:35   feature for things that are skipped in podcasts. And that would be an easy way to build an

00:39:41   automatic ad skipper or an automatic skip Merlin's comic book section thing.

00:39:46   [laughter]

00:39:47   I love Merlin, sorry.

00:39:49   So, you know, it would be, it would be interesting for that.

00:39:54   However, I love podcasts, and I love podcast creators, and I am a podcast creator.

00:40:00   And so, I wouldn't want to, because I know in reality what that would do if this got popular.

00:40:07   Honestly, even just the existence of this in an app that any sponsor had heard of,

00:40:12   that would to some degree devalue podcast ads.

00:40:16   And even if I didn't have my own show,

00:40:20   just caring about the ecosystem as much as I do,

00:40:24   I don't think I'd want to do that.

00:40:25   I think I would feel bad doing that.

00:40:27   And if you want to skip ads on your own, that's fine.

00:40:31   I've skipped ads before.

00:40:33   It's not the end of the world if a few people skip ads.

00:40:36   But to enable it to make it much better in mass like that

00:40:40   would do some damage.

00:40:41   And I don't want to do that.

00:40:43   The last thing I want-- this medium is not very big.

00:40:46   It's mostly very small producers, often single people or small groups of people like us.

00:40:53   It's very small producers.

00:40:54   We're not talking about ripping off NBC here.

00:40:56   We're talking about ripping off small people like us.

00:40:59   And so I just would not feel good doing that.

00:41:04   And yeah, that probably has something to do with me having my own show, but the reality

00:41:08   is if I thought about it at all, even if I didn't have my own show, it would still feel

00:41:12   a little bit wrong to do.

00:41:13   So that's the kind of feature that I'm going to almost definitely not do.

00:41:18   Now, but there's a gray area there, is there not?

00:41:20   And what I mean by that is, you could have a 30-second skip button, but you could avoid

00:41:27   the cool kid coalescing of all this data to get the communal skip button.

00:41:32   Does that make sense?

00:41:33   Right.

00:41:34   I do have a 30-second skip button.

00:41:35   I think that's an important feature to have for any client.

00:41:37   And that mostly comes back to Merlin's comics, right?

00:41:40   Oh, entirely.

00:41:41   No, just sometimes.

00:41:42   30 seconds doesn't put a dent in the comment section.

00:41:44   Are you kidding?

00:41:45   You got a 30 minute skip button.

00:41:49   Right.

00:41:50   So yeah, I mean, it's important to have things like that.

00:41:52   But I'd rather leave that to the user.

00:41:57   Leave it to them to skip what they want to skip.

00:41:59   And don't massively enable automatic ad skipping at scale.

00:42:04   That would just be kind of a dick move

00:42:06   for the whole industry.

00:42:08   The highlights feature is still a good idea, though.

00:42:10   like to know which sections... It's kind of like only allowing upvotes on comments instead

00:42:16   of downvotes, where instead of people marking the sections that they want to skip, how many...

00:42:20   You know, who... Or even just having the equivalent of dropping a marker, when they're listening,

00:42:25   if they think this part is particularly good, hit a star or something.

00:42:27   Yeah, like a like button. It expands out in a minute in either direction and makes a centrally

00:42:33   located little point. And then you can see where the points cluster of, "Oh, this is the funny part

00:42:36   or this is the part with the most likes and stuff." That sounds like a good idea.

00:42:39   Yeah, that's the kind of thing I could do.

00:42:41   I thought about, too, whether I want

00:42:44   to go into this whole area of detailed stats tracking.

00:42:49   Because that would probably anger some people

00:42:52   if I just snuck it in, that like, oh, by the way,

00:42:54   I'm recording the sections that you skip on my server

00:42:57   to accumulate with other people's data

00:42:59   and then present this to people in some anonymous but

00:43:02   accumulated way.

00:43:03   Or even just what they listen to and when.

00:43:05   Right.

00:43:06   I do want to start tracking that and get

00:43:08   into things like, I want to be able to tell publishers,

00:43:13   how many people who use Overcast subscribe to your feed?

00:43:17   How many people of those listened to this episode?

00:43:19   And how many of those people started it

00:43:23   but didn't finish it?

00:43:25   Or how far did they get on that?

00:43:26   I want to be able to tell publishers that.

00:43:27   First of all, as a publisher,

00:43:29   it'd be interesting to know that.

00:43:30   And second of all, if you're in a position

00:43:33   where you have a bunch of people using a client

00:43:35   and you control the server side of it,

00:43:36   why not have that data and look at it

00:43:35   try to gain some kind of insight and try to share that with the world. There's not a lot

00:43:39   of reason not to do that except that it will anger some people to be collecting that. So

00:43:44   I want to, if I do choose to do that, I don't even know if it will make it into 1.0, probably

00:43:49   not, but if I do choose to do that I'm going to be very upfront about that happening because

00:43:53   that would be weird if I wasn't.

00:43:55   Now would you as a means to make Overcast have a different income channel, would you

00:44:01   consider not selling the data but perhaps selling access to that data? So, I don't

00:44:08   know, either other podcast owners, I mean, I guess you kind of, maybe you already answered

00:44:13   that a moment ago, but would you, would that be free if I had a different podcast and I

00:44:20   wanted to know all the detailed statistics about what Overcast users were doing with

00:44:24   my podcast? How do you envision spreading that?

00:44:27   Well, I kind of want to be the anti-Stitcher here.

00:44:32   Stitcher basically, if you want your podcast

00:44:37   to be on their network or if they somehow

00:44:39   add it without ever telling you, all your hits go to them

00:44:43   because they cache the file and they transcode it

00:44:45   and everything.

00:44:47   So all those hits don't go back to your server.

00:44:49   So there's a bunch of people listening to your show

00:44:50   that you can't track unless you work with Stitcher, which

00:44:53   means signing up with them and then agreeing

00:44:56   all their crazy terms, and some of them are pretty crazy, and all this stuff, it doesn't

00:45:02   make me feel good to even think about working with them ever.

00:45:06   And so the last thing I want to do is, if I am collecting these stats, even though I'm

00:45:10   not going to be transcoding people's files, but if I collect any stats about shows, I

00:45:15   think I just make them public.

00:45:16   Just on the site, just here, you know, on the page that shows the show, just here's

00:45:20   all the stats about it that I know of.

00:45:21   I would feel better about doing that, I think.

00:45:24   But is that dicey in the sense that to some degree listener numbers are kind of like a

00:45:28   salary and oftentimes podcast producers don't like to share that kind of data?

00:45:35   Oh, certainly. But, you know, the reality is my one client on this one platform of iOS

00:45:42   is very unlikely to ever get big enough that these numbers represent the entire market

00:45:47   or even, you know, even like a uniform subset of it.

00:45:51   Sure.

00:45:52   to every case. So I don't think it would cause that kind of problems. You can look at--you

00:45:56   can already look at things like our friend underscore David Smith, his pod wrangler thing.

00:46:00   You know, he's told us about how our show ranks in his app relative to other shows.

00:46:06   But that's just his app, and it's different--like, that's a very different ranking than, like,

00:46:09   the iTunes top podcast chart, you know? And you look at things like Instacast and Pocketcast

00:46:15   that today already have directories, and they will list, like, their most popular shows

00:46:19   among people using their apps. And it's all a pretty different set. I mean, the tech podcast

00:46:23   ranked pretty well on all of them, but it's all very different from the iTunes top podcast

00:46:28   directory and things like that. So obviously we're not looking at a uniform, random subset

00:46:33   of the overall market with any of these clients. Yeah, because tech nerds listen to tech nerd

00:46:37   podcasts and buy podcast apps. But on David Smith's stuff, I think This American Life

00:46:44   was like mid-pack, and like tech podcasts.

00:46:48   In the reality, this American life gets slightly more than all the tech podcasts in the universe

00:46:53   combined.

00:46:54   So, yeah, it's not a representative example.

00:46:56   That's why I don't think it's probably a big deal to, you know, because it's like any show

00:47:01   that you are on is going to probably have a disproportionate number of listeners in your

00:47:06   podcast app that you talk about on your show.

00:47:09   So it's like, you know, it's not good.

00:47:11   You think the information would be interesting, but you certainly don't want to sell it.

00:47:14   You're just like, "Look, here it is.

00:47:15   Take it for what it is."

00:47:16   It's kind of like when—this is the same thing happens with websites when they show

00:47:19   like, "Oh, iOS 7 adoption numbers," and they show the hits to their website.

00:47:23   It's like, "That's not iOS 7 adoption numbers.

00:47:25   That's people who go to your website's iOS 7 adoption numbers, and your website is about

00:47:29   iOS and Apple stuff."

00:47:32   So it's not—yeah.

00:47:36   It's very difficult to pick any single website that would be representative of iOS adoption

00:47:39   as a whole.

00:47:40   But certainly it's not going to be some Apple tech news site

00:47:43   that is going to be representative.

00:47:45   Right.

00:47:46   And to answer C.O.Laff_ in the chat,

00:47:51   he's saying listing rankings is very different from listing

00:47:54   numbers of subscribers.

00:47:57   And that's true.

00:47:58   However, I think if I did list numbers and not just

00:48:00   relative ranks, the worst that would show--

00:48:03   because I'm not going to be taking over half the market

00:48:05   or anything.

00:48:06   The most this is going to show is how few people use my app.

00:48:10   Like that's that's the most likely outcome is like wow like your app is you know

00:48:15   You only have like 500 people who listen to this American life in your app. Like wow, you have nobody using your app like that's

00:48:22   That's that would be the bigger risk there. It's not like it's not that I would like

00:48:27   Become so huge that all of a sudden you'd be able to tell how everyone's doing just based on my my stats

00:48:35   So, I don't know. It's it's a tricky thing. I

00:48:38   I probably won't even do it in version one just because it's not worth the time,

00:48:42   but I would like to look into that in the future.

00:48:44   Anyway, let's move on to anything else except all me.

00:48:47   Because this isn't the Marco Show.

00:48:50   You want to do another sponsor before we move on?

00:48:53   Yeah, might as well. Let's do that.

00:48:55   Tell me about something else that's awesome.

00:48:57   This really is awesome. They're pretty cool. This is one of the coolest people,

00:49:01   or some of the coolest people that sponsor podcasts.

00:49:04   No offense to all the sponsors, but these guys really hit it out of the park.

00:49:07   It is, of course, Igloo software, a.k.a.

00:49:10   Igloo.

00:49:11   Igloo is an intranet you will actually like.

00:49:14   Go to igloosoftware.com/atp so that Igloo

00:49:17   knows that we sent you.

00:49:19   That way they keep buying sponsorships,

00:49:20   and we can keep buying new Mac Pros

00:49:22   so that we can have the credibility to tell you

00:49:24   to buy the new Mac Pro, and then you

00:49:26   can rationalize your upgrade.

00:49:28   With Igloo, you can share content quickly

00:49:30   with built-in apps.

00:49:31   They have built-in blogs, calendars, file sharing,

00:49:33   forums, Twitter-like micro-blogs and Wikis.

00:49:37   Everything is social.

00:49:38   You can comment on any type of content.

00:49:40   You can @mention your coworkers.

00:49:41   You can follow content for updates.

00:49:43   And you can use tags to group things around the way you work.

00:49:46   You can also add on rooms, like mini igloos,

00:49:49   for each of your teams to work in.

00:49:51   It's very easy to use.

00:49:52   The whole thing is drag and drop.

00:49:54   It features responsive design and uses beautiful fonts

00:49:57   from Typekit.

00:49:58   Your igloo is enterprise-grade security.

00:50:00   And you can start using it right away.

00:50:02   This is really cool.

00:50:03   It's all the stuff that you get from big social networks

00:50:06   and big social products that normally your intranet is stuck in the 80s. All this stuff

00:50:11   that we've been finding out that works and is cool on the consumer web, they take a lot

00:50:16   of this stuff, they take the best of this stuff and they make it work for enterprise

00:50:19   intranet with all privacy and security that you need when you're working in the enterprise.

00:50:23   So Igloo is free to use for teams of up to 10 people. That's pretty cool. That'll cover

00:50:28   a lot of companies that our listeners have out there. And when your Igloo grows past

00:50:32   It's only $12 per person per month.

00:50:35   So if nothing else, check out the website

00:50:37   just to see their really cool, funny videos

00:50:40   about their service.

00:50:42   And they are made by none other than

00:50:43   Woman's Sandwich, Adam Lisagor, Sandwich Videos.

00:50:46   So they have their own sandwich videos,

00:50:48   really cool, check them out.

00:50:49   Go to igloosoftware.com/atp to start building your igloo.

00:50:54   Thanks a lot to igloo for sponsoring the show.

00:50:56   - So I wanted to ask you, Marco,

00:50:58   since you, well the world conspired against me

00:51:02   and you got your 5S just before the show last week. Not that I'm still bitter about that.

00:51:06   Now that you've had a week with it, any new and interesting thoughts? To be honest,

00:51:10   I really don't have anything new and interesting. I still love the thing. But you only had brief

00:51:15   moments with it before the show, so I didn't know if you had anything interesting to add.

00:51:19   I don't really. No, it's my new iPhone. It works. I like it.

00:51:24   Have you given up on Touch ID?

00:51:27   Yes, I have. I'm going to Singleton with you, actually. All of us except John are going

00:51:33   to Singleton next week.

00:51:35   Indeed.

00:51:36   And I'm probably going to turn it on there just to try it out with a conference setting.

00:51:40   Like I said earlier, "Oh, maybe I should lock my phone when I'm at a conference." But the

00:51:43   reality is I don't really leave my phone anywhere. I never take it out of my pocket. I never

00:51:50   leave it on a table. My phone is either in my pocket or in my hand. That's it. So I've

00:51:56   never really had the need. No one's ever picked up my phone and mess with it. Not once.

00:52:01   So the only reason I would really need it would be if somebody actually stole my phone

00:52:04   out of my hand, which does happen, but that's a lot less common than people tweeting "pooping"

00:52:10   when you get up from the bar and leave your phone there.

00:52:13   So I really am not sure that I'm ever going to stick with it, but we'll see.

00:52:18   Oh, and to answer CoLaugh_ again in the chat room, how do you charge it? I charge it next

00:52:24   to me at night, so somebody would have to break into my house or my hotel room when

00:52:28   I'm traveling and come up next to my head and take my phone off of its charger or cable

00:52:34   to get it when it's charging. So again, it's really not... I'm not one of those people

00:52:39   who always has a dead battery and has to be plugged in at parties and stuff. I'm never

00:52:43   that person. I'm never in that situation. So there's just really very rarely a time

00:52:48   when anybody could get my phone without me knowing, without me knowing immediately and

00:52:54   being able to react immediately. So I don't know. I don't think it's much of a problem.

00:53:00   And the worst case scenario is they get into my email or something. What are they going

00:53:05   to do? Read the stuff I have saved in Instapaper? That's okay. Fine. Just don't delete it,

00:53:12   please. What else are they really going to do on my phone besides read my email or my

00:53:18   or try to do a password reset and get the email from that.

00:53:21   So the email is really the big security there.

00:53:25   But how long would they have my phone

00:53:28   without me knowing about it

00:53:29   before I could start doing something about it?

00:53:31   - And to answer head lead in the chat,

00:53:34   are we seeing, are you or I seeing any issues

00:53:37   with the accelerometers or gyroscopes

00:53:39   since I was making the rounds today?

00:53:40   I have not, but I'm not really sure

00:53:42   that I've been in a position that I would have noticed it.

00:53:45   I don't know if you've noticed anything.

00:53:47   - No, I'm the same way.

00:53:48   I mean, as I said last show, I think, and multiple shows,

00:53:51   I've never had the compass work properly

00:53:53   in any of my iPhones ever.

00:53:54   Since they added the compass, which was in, I believe, the 3G,

00:53:57   I've never-- or no, it was in the 4, I think, or the 3GS.

00:54:00   Anyway, I've never had it work properly, not once.

00:54:03   And so I haven't tried it with the 5S yet.

00:54:06   I've only had it for a week.

00:54:08   But I'm guessing it still works about the same, which

00:54:10   is it works sometimes, which for a compass isn't good enough.

00:54:15   So I'll probably never rely on it.

00:54:16   The accelerometer, you know, I-- the accelerometer has always been accurate enough that it'll

00:54:23   work okay, but inaccurate enough that I'm never going to use it for anything particularly

00:54:26   sensitive.

00:54:27   So, again, I haven't noticed.

00:54:29   Right.

00:54:30   Now, Jon, forgive me.

00:54:31   You said Tina is getting one, but not yet.

00:54:33   Is that right?

00:54:35   She kept-- she couldn't decide which color she wanted, it comes down to, and kept fretting

00:54:38   and asking me, "What do you think I should get?

00:54:40   I don't know."

00:54:41   She liked the gold, and she was afraid I was going to make fire.

00:54:43   I was like, "Just get the phone that you want.

00:54:46   Whatever you want, get the one that you want.

00:54:49   Don't worry about what I think."

00:54:50   And then she kept pressing me.

00:54:51   I was like, "Look, if it was up to me, I would tell you to get Space Gray."

00:54:53   And she's like, "No, I don't like that."

00:54:55   Eventually she finally decided, and she ended up getting a silver one.

00:54:59   Oh, so it is in the house?

00:55:01   Well, no, she ordered it online.

00:55:04   She went to the Apple Store once after work one day, and they didn't have any.

00:55:08   They only had Sprint or something or whatever.

00:55:10   And so she didn't want to keep going back to the store.

00:55:12   I was telling her, I heard on the latest talk show that Gruber was talking about how nice

00:55:16   it is to be able to order it from your phone, and you could tell it replaced the phone that

00:55:20   I'm currently ordering from.

00:55:21   So when the new one comes in the mail, it'll be like all set to replace that other one.

00:55:25   You don't have to enter all your information, which I thought was neat.

00:55:28   And so I told her to do that, and she did.

00:55:30   It was a little bit tricky in that when you order from your phone, you are logged into

00:55:34   the Apple Store application with your iTunes Apple ID, and we share an iTunes Apple ID

00:55:40   for the households for the apps that we buy, so had to sign her out of her iTunes Apple

00:55:44   ID and into her own Apple ID that she uses on her Mac and stuff, then buy the thing and

00:55:49   then sign back in.

00:55:50   So it was a little bit cumbersome.

00:55:51   It would be nice.

00:55:52   Apple's usually pretty good about giving you a separate Apple ID for each thing that you

00:55:56   use.

00:55:57   Like, "This is the Apple ID I want to use to do my iCloud syncing.

00:56:00   This is the Apple ID I want to use for the App Store."

00:56:02   And this is, especially on the phone, but on the phone, apparently, the iTunes and App

00:56:07   App Store Apple ID is shared with the Apple Store application.

00:56:11   But anyway, she ordered it.

00:56:13   It says "delivery in October," kind of like "Mavericks is due in fall."

00:56:18   So we don't know what that means yet.

00:56:21   And she picked silver because in the end, I think she decided that her color coordination

00:56:26   options are best with a more neutral color like silver, even though when you put the

00:56:31   gold in a case, it is really hard to tell that it's gold, especially depending on the

00:56:35   color temperature of the lighting to tell, you know, is that silver or is that gold or

00:56:39   is it just warm lighting in this room?

00:56:41   It's really hard to tell.

00:56:42   So she got silver and she got the red leather case with it, and she got the 64 because apparently

00:56:45   I convinced her with my talking on the past show about how you're regretting if you get

00:56:49   the smaller size.

00:56:50   So.

00:56:51   Nice.

00:56:52   Yeah, I'm curious to see what the leather case is like.

00:56:57   I got a, so I had a stock bumper on my 4, this is so boring but I'm already committed,

00:57:02   I got a stock bumper on my 4S and I liked it a lot until I destroyed it over the course

00:57:08   of a year.

00:57:09   And then I rolled without a case for about a year and I didn't like shatter anything.

00:57:14   But by the end of the year, the second year with my 4S, it was looking a little rough.

00:57:18   And so I got this literally $3.50 Monoprice bumper for the 5S, which is okay, but I handled

00:57:25   the leather case for literally five or 10 seconds when I was in line on launch day and

00:57:30   And my recollection of it, after having been up since five, and this was like eight or

00:57:33   nine in the morning, was that it was really nice, but $40.

00:57:38   And so I want to go back to the store and see if maybe that's worth it a little bit

00:57:43   more, a little bit nicer than this cheap bumper that I didn't realize says "Monoprice" across

00:57:49   the side on one side until it was already here.

00:57:52   But I don't know, we'll see what happens.

00:57:53   So I'm curious to, in summary, I'm curious to hear what you guys think of it once you

00:57:55   have it.

00:57:56   with the leather case on other people's phones.

00:57:59   And the slams I heard against the case was that it's hard to get off, which I don't really

00:58:02   care about because it's not like I'm ever going to take it on and off, and I doubt my

00:58:06   wife would either, and that it made the buttons hard to press.

00:58:08   And so that was the first thing I tested when I saw someone's phone with a leather case.

00:58:14   What it does do is it's kind of like if you have wood detailing in your house and you

00:58:19   put lots of layers of paint over it, the detailing kind of goes away.

00:58:22   That's what it does to the buttons.

00:58:23   They used to be prominent, easy to find and feel.

00:58:26   With the leather over it, they become less prominent, stick out less or whatever.

00:58:31   But the actual pressing of them, it felt fine to me.

00:58:34   I don't know what they have inside there, but it's not...

00:58:36   I've had plenty of crappy cases from my iPod touches over the years, and some of them it's

00:58:39   like, you feel like you're squishing your way through just like this big jello blob

00:58:43   and somewhere under there is a button.

00:58:46   It felt pretty positive, the connection between I press here, the button goes in, and the

00:58:50   little click of the actual button could be felt through the leather thing.

00:58:53   And my main thing is, she said, "Yeah, but they said the buttons are hard to press."

00:58:57   I'm like, "Look, I've tried it.

00:58:58   It's not that bad."

00:58:59   But how often do you hit the volume up/down buttons?

00:59:00   I don't hit them that often on my iPod Touch.

00:59:02   I don't know if she hits them often.

00:59:04   And the power button, when do you ever use the power button?

00:59:06   I've long since switched to using the home button to wake my thing up, especially with

00:59:11   the touch ID thing.

00:59:12   Do you guys ever touch the power button on the top of your iOS devices?

00:59:15   All the time.

00:59:16   Yep.

00:59:17   Oh, I always wake it up with the home button.

00:59:19   That's what I hit when I'm taking it out to check the time or something.

00:59:22   Yep.

00:59:23   And if I ever want it to sleep, you know, because I'm done using it, then I won't let

00:59:27   it time out.

00:59:28   I'll just turn it off.

00:59:29   Yeah, I guess maybe I do grab it for when I turn the thing off.

00:59:32   Yeah, I don't know.

00:59:34   I don't do it when I'm waking it up, but maybe when I put it down.

00:59:36   I don't know.

00:59:37   But anyway, the buttons seemed like they were good enough to hit.

00:59:40   And, you know, whatever.

00:59:41   Like, yes, it's expensive, but all I can think back to is the leather iPad cases, which I

00:59:46   think were like 70 or something.

00:59:50   It's all ridiculous.

00:59:51   can imagine what the margins are on that thing. The worst part about the leather cases is

00:59:54   they don't even feel like leather. If you're going to use real, genuine leather, it should

00:59:57   feel like leather. But it doesn't. It doesn't even smell like leather. It smells like something

01:00:01   else. Someone was saying it smelled like original NES manuals, but that person may have been

01:00:08   having a stroke, so it's hard to believe.

01:00:11   Wow. I'm good on the 5S. Anything else, though?

01:00:14   No, I think I'm good.

01:00:17   All right. John, tell me about how to charge a battery.

01:00:19   Oh no, not this again.

01:00:21   Yeah, it's not about charging the batteries.

01:00:23   This is another thing we got a lot of replies about.

01:00:24   I was thinking that someone at Macworld was talking to me about what features do you want

01:00:30   to see, or maybe it was ours, I don't remember, what features do you want to see in Mavericks

01:00:33   or something.

01:00:34   This was before Mavericks had been announced.

01:00:36   It was like, what features do you want to see in 10.9 or whatever.

01:00:39   And one of the ones I suggested that's been bothering me for a long time is, if you, like

01:00:45   my wife, have a laptop but keep it plugged in pretty much all the time

01:00:49   that's terrible for your battery and it would be nice if the OS took care of that

01:00:54   and like, you know, extended the battery's life

01:00:58   and why is it terrible for your battery to be plugged in all the time?

01:01:01   well it's not the fact that it's plugged in that's terrible, it's the fact that

01:01:05   the battery is charged to full capacity all the time, it's very bad to keep a

01:01:10   lithium-ion battery charged to 100% capacity all the time

01:01:13   and that will shorten the life of your battery so that when you do unplug it after, say it's been plugged in for two years straight,

01:01:19   and you unplug it and you want to use it, you don't get the kind of battery life that you would, you know,

01:01:22   if you had just taken it out of the box on day one and charged it to full and then gone to a cafe or something.

01:01:27   Oh look there, in the new Mac Pro, what we'd like to see? I didn't think that was it.

01:01:31   Anyway, someone put a link in the chat room to that article that I was thinking of that I tried to Google for but couldn't find.

01:01:36   So this came up again in a Wired article, and it's always these same guys from like Battery University,

01:01:41   bunch of battery scientists promoting this idea that, hey, it's really bad for lithium-ion

01:01:46   batteries in particular to be kept at full charge all the time.

01:01:50   And it would be nice if OSS took care of this and kept it at a lower charge level.

01:01:54   And they say that the optimum charge level for storing lithium-ion batteries is 40%.

01:01:59   So if you're going to put it on a shelf and let it sit there, don't charge it to full

01:02:02   capacity and put it on a shelf, and don't drain it to nothing and put it on a shelf.

01:02:06   Charge it to 40% and put it on a shelf.

01:02:08   And their suggestion for active use things is charge it to 80%, discharge to 40%.

01:02:14   That is the best way to extend the life of the battery.

01:02:19   The current Mac OS doesn't do any of these things, but it does in more recent versions,

01:02:24   as everyone has been telling me, and I have no reason to disbelieve them, it does make

01:02:27   it oscillate between 100% and 95%.

01:02:30   So it doesn't just leave it capped off at 100%, because that would be terrible.

01:02:33   It does kind of, I think they call it conditioning the battery, where it lets it discharge to

01:02:37   95, then charges it back to 100, then lets it discharge to 95.

01:02:41   And what I'm suggesting is a feature that would let it discharge all the way down to

01:02:46   40% while it's plugged in, and then either keep it at 40% or bring it back up to 80,

01:02:50   whichever one is actually better for the life of the battery, I'm not sure.

01:02:54   And the problem with this feature, and the reason it hasn't been implemented, is not

01:02:58   a technological problem.

01:03:00   It's a user experience problem, because if you do that, then when you need to take your

01:03:04   laptop, you're like, "Oh, well, now the stupid thing's at 40%."

01:03:07   And now the battery life I was supposed to have in my laptop because it was in this middle

01:03:11   of this crazy cycle where it always drains it down to 40% or charges it back to 80% or

01:03:15   remember keeps it at 40%.

01:03:16   Now, "Oh, I got to run.

01:03:17   I got to go out.

01:03:18   Let me take my laptop with me."

01:03:19   It's at 40%.

01:03:20   And people would hate that, and it would be terrible.

01:03:22   So I kind of understand why they haven't implemented it as a feature, and I kind of understand

01:03:26   when they chose to do it they just had it oscillate between a hundred and ninety-five percent, but at the same time

01:03:30   for people like me who know that

01:03:33   you know I'm going to this computer I

01:03:36   It just sits there on that desk plugged in day after day after day after day

01:03:40   It's like it's behind the 27 inch cinema display like you have to reach around behind it even get it

01:03:45   We don't even look at the screen even though it is open for cooling reasons

01:03:48   It would be nice if the OS at least just api's or something where you could tell it don't

01:03:55   cycle between 90 and 95. Go down lower. Don't even charge past 90. And as many people who

01:04:01   are tweeting back and forth have been either telling me or asking me, "Isn't that what

01:04:05   electric cars do?" And yes, it is what electric cars do, because they are optimizing for the

01:04:10   life of the battery, because you don't want to buy a $90,000 Tesla, and two years later

01:04:15   the battery is fried like it would be on a laptop that you use constantly.

01:04:18   Well, but if you do the math, then if you count all these tax credits, and if you count

01:04:23   the time that you would spend getting gas and repairs for your other car, then it ends

01:04:27   up being only a dollar a month.

01:04:29   Yeah, but then you're out of $90,000 for the car when you've got to get a new one, because

01:04:32   the cost of the battery is like half the cost of the car. But yeah, those batteries charge

01:04:38   to full capacity, and most of those batteries, they don't let you go past $80,000 in most

01:04:41   ... I don't know about Tesla's exact policy, but the same thing with Prius. Any car with

01:04:46   a battery, they pretty much don't let you charge that battery to 100% capacity, because

01:04:50   It's terrible for the battery. They let you charge to 80 and they don't let you discharge all the way either

01:04:54   So you're kind of using this this middle band of power in the battery and when they give you the power ratings for the battery

01:04:59   And like they tell you you know how much mileage you got up

01:05:01   They're telling you how much that middle is worth because they want you to avoid

01:05:04   Pushing the thing to 100% capacity

01:05:07   Unless you do some big override and say like I need super duper range for this one time and the same thing for draining down

01:05:12   They don't let it go down to zero if they can possibly help

01:05:15   But they want you to recharge it before it gets up to there

01:05:17   and zero isn't really zero. And Apple's laptops are just starting to do that now.

01:05:22   When you have it plugged in, even if the battery—it'll lie to you. It'll say, "Oh, battery is full,"

01:05:28   and it'll show the little whatever, the plug symbol that shows that the battery isn't charging.

01:05:31   But maybe it's at like 96%, right? It's just trying to tell you, "We're not going to charge

01:05:37   it anymore because we're in the middle of this conditioning cycle," or whatever.

01:05:40   So I kind of understand why this feature doesn't exist, but I think at this point, especially

01:05:44   in like an energy saving release like maverick so that would have been the time to say either

01:05:49   provide APIs or provide some optional mode for like click this checkbox that says yes

01:05:54   I'm always plugged in and accept the fact that if you need to leave in a hurry and grab

01:05:57   your laptop it might not be a hundred percent capacity but I don't think that's coming.

01:06:03   Now I've seen some of this feedback fly by and I know a lot of people have pointed to

01:06:07   fruit juice which is apparently an app and do I have this right so what it does is it

01:06:13   tells you to unplug to kind of force you to take charge of this whole situation, is that

01:06:19   correct?

01:06:20   Yeah, and that's not what I'm looking for.

01:06:22   I've tried to do that manually, and every time I've tried to do that manually, accidentally

01:06:25   the computer unplugged and come back and find it has gone into hibernation mode, right?

01:06:31   The whole point is, you can't trust me to plug and unplug, I will forget.

01:06:35   I want to leave the little MagSafe thing connected all the time and then have the computer say,

01:06:39   "I'm not accepting power now because I'm in this..."

01:06:41   like it does with conditioning between 195, but for a much bigger range, treating the

01:06:47   battery inside this plugged-in thing more like the battery in a car, or the battery

01:06:51   in a car that never goes anywhere. So again, maybe keep it at 40%.

01:06:54   Well, one thing they're doing with iOS 7, which we can now talk about finally, is that

01:07:01   they have, in iOS 7 they have this background fetch thing, which is awesome. And they talked

01:07:05   to be able to see about how one of the ways it works is, you know, your app says, "Here's

01:07:11   how often I'd like to be--or I'd like to be woken up for background updates." And the

01:07:16   system actually decides when to actually do it. And what it does is it actually measures--it

01:07:22   keeps track of when you tend to launch certain apps or when the phone tends to be completely

01:07:27   idle for a long time. And then it tries to predict when you're going to launch the app

01:07:32   next and do the update shortly before that time, during a time when you're likely to

01:07:39   be on Wi-Fi and have power. Like if you're charging overnight and you wake up in the

01:07:42   morning and you launch your apps, it'll launch it like short, it'll background refresh it

01:07:46   shortly before that so it has fresh data. If they brought something similar to that

01:07:51   over to the Mac, which they probably can do pretty easily, then they can do things like

01:07:59   automatically do a charge cycle down to 40% and then back up to 80% every third day in

01:08:06   the middle of the night when you aren't using your computer. Because you never use your

01:08:10   computer during that time or you've used it like once in the last year during that time

01:08:13   and so it'd be worth it. They could offer a feature like that based on heuristics and

01:08:18   analysis of how you actually use your computer so that it could be doing this stuff when

01:08:22   you don't even notice because you're asleep.

01:08:24   Exactly. That's what I was going to say. If you were going to do this, that's the way

01:08:29   way you would try to do it. And with the heuristics of background apps on iOS, it's not the end

01:08:33   of the world if it doesn't, like, it didn't get my RSS subscriptions because it hasn't

01:08:36   learned my--it's kind of like the Nest thermostat, where you want it to learn, but the consequences

01:08:39   of it not learning aren't that big. In the case of charging your laptop battery, the

01:08:43   negative consequences are a little bit more severe in terms of user experience, where,

01:08:47   like, I guess Nest, if it makes your house the wrong temperature, pisses you off too.

01:08:50   But like, you know, oh, the new MacBook Pros and the new, you know, Apple operating system

01:08:56   will learn your habits and keep your battery life longer or whatever. I still think it

01:08:59   would have to be opt-in because there's going to be, especially during learning period,

01:09:01   and even outside of learning period, like that time when you need to grab your laptop

01:09:05   and go and it's at half capacity and now you're pissed and you're like, "I paid all this money

01:09:09   for this laptop with 12-hour battery life and I get six because I picked it up at the

01:09:13   wrong time." And the final thing is that Apple probably is the only one that knows how many

01:09:17   of their laptops spend their time plugged in all the time.

01:09:21   Maybe I'm an outlier and maybe that's why, you know, they don't care because most people

01:09:24   who buy a laptop use it.

01:09:26   I mean, our laptop is plugged in 99.9% of the time,

01:09:30   but the reason we got a laptop and not an iMac

01:09:32   is because when we go on vacation,

01:09:34   we just grab a little 13-inch air,

01:09:35   and it's much easier than trying to lug an iMac.

01:09:38   And even when I'm podcasting in here

01:09:40   and my wife wants to use her computer,

01:09:42   she unplugs the laptop and brings it into the other room

01:09:45   so she can use her computer with all her stuff on it

01:09:48   in a little portable form.

01:09:49   So we'd have to see what the numbers on that are,

01:09:52   and it could be if there's not enough people do it.

01:09:55   Certainly iOS would get that type of thing first

01:09:57   and it has with the background updates,

01:09:58   but eventually they'll get around

01:10:00   to bringing that back to the Mac

01:10:01   and that's definitely a feature I'd like to see.

01:10:03   - All right, anything else?

01:10:06   Are we good for today?

01:10:07   - Yeah, let's keep it there for today.

01:10:08   We're pretty good.

01:10:10   - This might actually be a little bit shorter.

01:10:11   Is it safe to say that over an hour in?

01:10:14   - No.

01:10:14   All right, thanks a lot to our two sponsors this week,

01:10:16   Squarespace and igloo, and we will see you next week.

01:10:20   - See ya.

01:10:20   Accidental, accidental.

01:10:22   Now the show is over, they didn't even mean to begin

01:10:29   'Cause it was accidental, oh it was accidental

01:10:36   John didn't do any research, Marco and Casey wouldn't let him

01:10:41   'Cause it was accidental, oh it was accidental

01:10:46   And you can find the show notes at ATP.fm

01:10:52   And if you're into Twitter, you can follow them @C-A-S-E-Y-L-I-S-S

01:11:00   So that's Kasey Liss, M-A-R-C-O-A-R-M

01:11:05   Auntie Marco Arment, S-I-R-A-C

01:11:10   USA, Syracuse, it's accidental

01:11:16   They didn't mean to, accidental

01:11:21   Tech not cast so long

01:11:25   People in the chat room are still obsessing about this battery thing of like, "Why don't

01:11:29   they just stop sending energy to the battery once it's fully charged?"

01:11:32   They do.

01:11:34   They do.

01:11:35   Like, the problem—this is a problem we should get like, you know, an electrical engineer

01:11:40   on even though my major was electrical computer engineering.

01:11:42   Like somebody who works with electricity every day to give whatever is the current best analogy

01:11:47   because the way people think electricity works and the way it actually works are not the

01:11:51   same.

01:11:52   I think people visualize, probably because this is one of the metaphors they use in school,

01:11:56   they visualize electricity like water going through a hose, and that somehow it's bad

01:12:00   because the water is always pressing into your battery, like puffing it up like a water

01:12:03   balloon, and that's not why it's bad.

01:12:05   It's bad to keep a battery at full charge with no electricity going into it.

01:12:10   Just charge it up to full, disconnect it from everything, and suspend it in a vacuum tube.

01:12:14   It's still bad.

01:12:17   the analogies of water flowing into things and

01:12:21   uh... like a

01:12:22   what is the norm on the use? The current is the

01:12:26   the volume of water and the voltage is like the speed of the water like all those

01:12:29   analogies lead people astray and make people think about

01:12:32   their electrical

01:12:33   components in ways that are not healthy

01:12:36   so yeah

01:12:37   don't worry people, the electricity is not pushing into your battery really hard

01:12:41   and causing it to bulge out

01:12:42   [BLANK_AUDIO]