127: Not a Cactus in Sight


00:00:00   How do you pronounce S-T-A-H-P?

00:00:02   What? That's not a word.

00:00:04   Yes it is. Well, not really.

00:00:06   But it's what the kids say when they really want you to stop doing something.

00:00:09   Is it "STAP"?

00:00:10   I guess.

00:00:11   "STAP"? I don't know.

00:00:12   Something like that. I don't know.

00:00:14   I'm gonna go with "STAP".

00:00:16   Like maybe, maybe like the Great Lakes accent.

00:00:19   "STAP".

00:00:20   Oh god.

00:00:21   Yes, because that's what this podcast needs, is more talk about regional accents.

00:00:24   Yes, definitely.

00:00:25   Definitely.

00:00:26   I actually got called out at work today in a happy way.

00:00:31   Some co-worker asked me, a co-worker that I just started working with, asked me if I

00:00:34   was from Jersey.

00:00:35   Well, that part wasn't happy at all.

00:00:36   That was miserable.

00:00:38   But point being, I said mash on something as in to like hit emphatically or repeatedly,

00:00:45   and apparently that's a northeastern thing or so I'm told.

00:00:48   Well that's a drink in Great Britain, right?

00:00:50   Smoosh or whatever?

00:00:51   Oh, that was squash that you're thinking of.

00:00:53   Squash!

00:00:54   I thought it was smush.

00:00:55   [laughter]

00:00:56   So Tiff's opinion of that, when she heard this, she said it's probably like getting

00:01:02   like frozen fruit concentrate here, like the fruit juice concentrate.

00:01:05   I think that's correct.

00:01:06   From the most I can piece together, but as we talked about on the last episode of Analog,

00:01:12   I sent an emergency text to underscore David Smith, the Internet's arbiter of all things

00:01:18   UK versus American and also the Internet's historian.

00:01:24   And he indicated to me that we do not have anything that's really equivalent to squash.

00:01:28   Well, we have things that have that name, but it's different.

00:01:31   It's like football.

00:01:32   It's like, "Well, we have that, but that's not what you think it is."

00:01:34   Yeah, well, it's either football or vegetable.

00:01:36   Yeah.

00:01:37   So anyway, I was asking you what's going on with cellular downloads on Overcast.

00:01:43   And this is okay because I see in the show notes it says "pre-follow-up" and then

00:01:48   Marco talks.

00:01:49   I did not type that this time.

00:01:51   I don't know who did.

00:01:52   Clearly, it was John.

00:01:53   - I just keep adding sections.

00:01:54   What about pre-pre-follow-up?

00:01:56   - We'll get there, maybe next episode.

00:01:58   - What happens if the pre-follow-up generates follow-up?

00:02:01   What do we do with it next episode?

00:02:03   Where do we put it?

00:02:04   - Then it gets lumped into regular follow-up.

00:02:06   - It doesn't get pre-follow-up follow-up?

00:02:08   - I don't think so.

00:02:09   This is getting recursive, I'm uncomfortable.

00:02:12   No, I was asking you what's going on

00:02:14   with cellular downloads and overcast

00:02:16   because you seem to be stressing out about it quite a bit.

00:02:19   - Well, this is a constant battle I have

00:02:22   with designing Overcast is like,

00:02:25   where do I strike the balance between offering settings

00:02:29   for customization versus complexity of the app?

00:02:33   Because if you look, so many well-featured,

00:02:37   well-regarded podcast apps have just these

00:02:40   massive setting screens.

00:02:42   And it's not because they're badly designed by idiots,

00:02:45   it's because it's a hard problem,

00:02:47   and when you make a podcast app,

00:02:48   everybody asks for all these features.

00:02:50   And it's really hard to please a large number of people

00:02:53   without adding a ton of options

00:02:56   and customizations and features.

00:02:58   I've always tried to differentiate Overcast

00:03:01   from the other well-established clients

00:03:04   by being simpler in those regards.

00:03:07   It's really, really hard to keep that up, to manage that.

00:03:10   'Cause right now I'm working on version 2.0.

00:03:12   One of the big features of 2.0

00:03:14   that I'm willing to talk about is streaming.

00:03:16   I've been working on streaming for a long time,

00:03:18   as I mentioned on this show a couple times

00:03:19   in various after shows, it does finally work.

00:03:23   It is working, I've been using it myself

00:03:24   for the last couple of weeks, I don't really like it.

00:03:26   (laughing)

00:03:28   It works well, it is solid, and it is awesome.

00:03:33   - And to be clear, it works with Smart Speed.

00:03:35   - Of course, I wouldn't do it without it.

00:03:36   - That's what I thought, just wanted to make sure.

00:03:38   - Now I have to face the other issues

00:03:40   of having streaming in the app,

00:03:42   so there are certain options that it needs.

00:03:46   Things like, what do you do when a new episode comes in,

00:03:49   do you market for streaming

00:03:51   or do you download it automatically?

00:03:53   There has to be some choice there.

00:03:56   There has to be an issue of like,

00:03:57   what do you do if you're playing a playlist

00:03:59   and the episode that you're on ends

00:04:01   and the next episode is a streaming episode

00:04:04   but you are offline, what do you do?

00:04:07   No matter what you pick,

00:04:09   you're gonna anger some people

00:04:11   and you're gonna please some people.

00:04:13   And it's really hard to know a lot of times ahead of time

00:04:16   which one of those sides will come out ahead, you know?

00:04:18   (laughing)

00:04:19   So the question I have now is,

00:04:22   should I permit cellular data usage for streaming?

00:04:27   'Cause right now, I have cellular downloads.

00:04:30   And so in a world where you,

00:04:31   in an app that you have mixed streaming and downloads,

00:04:35   now you have to have a little more control over that.

00:04:37   So it's questionable.

00:04:39   So no question, the thing that I agree on, no question,

00:04:44   is that cellular data not only will be

00:04:47   allowed for streaming, but it will be allowed by default and

00:04:50   it'll be on by default because in the world we live in today, I

00:04:55   think people widely expect things to just work when you

00:04:58   tap them and lots of other media apps stream by default

00:05:02   over cellular and don't ask you first things like YouTube or

00:05:06   most of the streaming music services. Although I think

00:05:09   Apple Music does not. I think they ask you first. But anyway,

00:05:12   almost all media streaming apps, they will just do it. If

00:05:16   If you say play this video and you're on cellular

00:05:19   on your phone, it'll just do it over cellular

00:05:21   and that's fine.

00:05:22   So no question streaming has to be there

00:05:25   and it has to be enabled by default over cellular.

00:05:28   The problem is there are lots of,

00:05:30   and when I posted these tweets, it's funny,

00:05:32   everybody thinks that they have the most advanced

00:05:35   world view on this issue.

00:05:37   (laughing)

00:05:38   And of course I'm the stupid American,

00:05:40   so I don't know what the rest of the world is like.

00:05:41   So I have heard from about equal numbers today

00:05:45   of people saying both in my country, we have very bad data

00:05:49   plan, so you have to have restrictions and the rest of the world works this

00:05:53   way and therefore you need to have very strong restrictions and I've also heard

00:05:57   from other people saying well in my country everybody has unlimited data and

00:06:01   that's that's the world we live in now and so you don't need to even bother

00:06:03   with this.

00:06:04   I've heard both ends of this a lot, but right now I have I have two cellular

00:06:11   options in the app and I wish I could get away with just one.

00:06:14   But right now I have two.

00:06:16   So the big one, the main one is whether I allow background

00:06:21   or foreground, whatever, when the app downloads full

00:06:24   episodes ahead of time or when it gets a notification

00:06:26   or whatever, when it downloads full episodes,

00:06:28   do you download the whole thing over cellular

00:06:30   or do you wait for wifi?

00:06:32   So that is an option and I think that option is good.

00:06:35   That deserves to stick around because if you just stream

00:06:39   and you say all right, play this right now,

00:06:41   that's an action you're taking.

00:06:42   You're deliberately saying, play this episode right now

00:06:45   and I'm recognizing that I'm playing a podcast

00:06:48   while I'm on my phone on a cell network,

00:06:50   this might use data, fine, you know.

00:06:52   But if you background download things that just come in,

00:06:55   like overnight somebody could blow their data cap

00:06:57   if a whole bunch of episodes are released

00:06:59   by their favorite podcast, maybe their feed messes up

00:07:02   and it releases like 10 new episodes accidentally

00:07:04   and then the phone downloads 10 things over cellular

00:07:06   without them even knowing or initiating it.

00:07:08   So it makes sense to differentiate

00:07:10   between automatic background downloads using it

00:07:12   and things that you initiate as the user

00:07:15   for playback using it.

00:07:16   So that makes sense.

00:07:17   You should have an option to disable cellular

00:07:20   on background downloads, fine.

00:07:21   But then there's an iOS annoyance that I have.

00:07:26   It's, I consider it a bug.

00:07:28   I'm not sure that Apple does.

00:07:30   For every app that you have on your phone

00:07:32   that uses data at all, you can go into settings

00:07:35   and you can turn off cellular data usage

00:07:37   for the app at the iOS level.

00:07:40   So the app doesn't have to have an option for that.

00:07:42   You can just go in and disable it,

00:07:43   and many people do who are on limited plans

00:07:45   who really need to watch their usage.

00:07:49   So the problem is, in previous versions of iOS,

00:07:53   you've always had this so-called reachability API.

00:07:56   It actually has this long CF network, whatever name,

00:07:58   but it's shortened to everyone says the reachability API.

00:08:01   And this is the API that can notify you

00:08:04   of changes in internet connectivity

00:08:06   so that you can, for instance, your app can know whether the connection's offline and

00:08:11   when it goes back online. And you can also tell whether it's online via cellular or Wi-Fi

00:08:16   or neither, right? So you can make a whole bunch of intelligent decisions and it isn't

00:08:19   perfect. Sometimes it isn't that reliable, but it's pretty close. It's pretty good most

00:08:22   of the time. In previous versions before iOS 8, if somebody turned off that cellular option

00:08:29   for you in system settings, so that they say this app can't use cell data anymore, your

00:08:34   would see when it asked the reachability API, "Am I online?" The system would say, "No,

00:08:40   you're offline." So the app couldn't tell this switch was on. It just seemed like it

00:08:45   was offline. And so if the app tried something, it would show that annoying dialog box that

00:08:52   everyone has seen that says, "Cellular data is disabled for app name. You can enable it

00:08:57   in settings." And there's no way for an app to initiate a network connection that does

00:09:03   does not show that box.

00:09:04   So there's no way that I can say like,

00:09:06   you know what, I'm gonna do a background sync here.

00:09:08   If it fails, the user doesn't even need to know.

00:09:10   Doesn't matter, it's not that important.

00:09:12   Or I'm gonna download some artwork for the show

00:09:14   that's showing in this table cell.

00:09:16   If that fails, it doesn't really matter either.

00:09:18   Just, you know, you don't have to alert the user

00:09:19   with a modal dialog box saying cell data's off

00:09:22   for me downloading artwork to show in a table cell, right?

00:09:24   Doesn't matter.

00:09:25   The smart thing to do as a developer in that case

00:09:28   was to, before you do something optional

00:09:32   that the user didn't really initiate,

00:09:34   you know, something like a background sync

00:09:35   or an artwork download.

00:09:36   Check reachability, and if it says it's offline,

00:09:39   don't even try.

00:09:40   And then that dialog will never show up.

00:09:42   'Cause if you don't do that,

00:09:44   then every time somebody goes to your app

00:09:46   and it tries to do a background sync,

00:09:47   and they've disabled cell data for your app,

00:09:49   the connection will not only fail,

00:09:51   but it'll show that box and it'll annoy the crap out of them.

00:09:53   And so then they will write you saying,

00:09:55   why does your app keep showing this box?

00:09:56   'Cause they don't know it's a system box

00:09:58   that you don't have any control over.

00:09:59   So that's bad.

00:10:01   Anyway, with iOS 8, there's what I consider a bug,

00:10:05   which is if someone has disabled cellular data

00:10:07   for your app in iOS settings,

00:10:09   Reachability API will tell you that you are online.

00:10:13   It will not say you're offline anymore,

00:10:16   so that you have no way to tell

00:10:18   whether you are online unrestricted

00:10:21   or whether you are online but with cell data disabled.

00:10:24   And so you can't avoid making those requests.

00:10:28   and you also still can't make a request

00:10:31   that's marked as some kind of optional

00:10:32   so it doesn't show that dialog box.

00:10:34   So therefore, if someone disables that for your app

00:10:39   and you try to make a connection,

00:10:40   it will show that box every time,

00:10:42   or at least once every 10 minutes or whatever.

00:10:44   There's some kind of throttling on

00:10:45   so it doesn't show them constantly,

00:10:46   but it still shows them enough

00:10:48   that it annoys people and they email me.

00:10:50   So in order to work around this slightly,

00:10:53   I had to add the second option called Sync Over Cellular,

00:10:58   and it's in my nitpicky details settings level,

00:11:01   which is where I bury all my other options.

00:11:03   And the reason why this is there

00:11:05   is not to let people save data,

00:11:07   because it doesn't save that much data.

00:11:09   It's to let people avoid that dialog box

00:11:11   if they've disabled cell data for my whole app.

00:11:14   Because with that, if you trigger that option,

00:11:16   then I won't even attempt

00:11:18   to make those connections over cellular.

00:11:20   Because again, I can't distinguish between

00:11:22   cellular that I'm allowed to use

00:11:23   and cellular that I'm not allowed to use.

00:11:26   So with that setting, that's what that whole thing is for.

00:11:28   It's to work around this giant bug in iOS 8

00:11:31   that I think is still there in 9.

00:11:34   You know, I'll have to do some tests.

00:11:35   Maybe, I really hope 9 fixed it, but I don't think it did,

00:11:37   but I'll double check before next week,

00:11:39   and I'll, well, before our fake next week show,

00:11:42   and I'll report back.

00:11:43   Because if that fixed it, I can remove that whole option,

00:11:46   and that would be awesome.

00:11:48   Anyway, sorry for the very long-winded thing here.

00:11:52   Anyway, so now, faced with the question of adding streaming

00:11:57   to this and having some kind of cellular control

00:11:59   over streaming, do I have three cellular data options?

00:12:03   That's terrible.

00:12:05   I'd have to just make a whole separate screen for them

00:12:07   and explain, and that's, I mean, I will if I have to,

00:12:10   but that sounds terrible.

00:12:12   Or my best idea so far is to keep the two options

00:12:16   I have now and just attack streaming to the second one

00:12:20   so that the background downloads

00:12:22   are still separately controllable.

00:12:23   'Cause that's something, I can see somebody

00:12:25   wanting to do background downloads only when they're on WiFi

00:12:28   but be willing to stream wherever they are

00:12:30   if they actually ask for it.

00:12:32   The second option, rather than calling it sync-over-cellular,

00:12:35   rename it to stream-end-sync-over-cellular.

00:12:38   Does that make sense?

00:12:40   I know this is very long, probably very boring.

00:12:42   We'll cut all this out.

00:12:44   - I think it does make sense.

00:12:46   I would probably agree with you that sync-over-cellular

00:12:49   become stream-in-sync over cellular.

00:12:53   I don't know, it's interesting because I have an unlimited data plan from AT&T on my iPhone.

00:12:58   I've gotten grandfathered in and I've never had a terribly compelling reason to walk away

00:13:03   from it.

00:13:04   I know that I could probably save a few bucks a month if I didn't keep it anymore, but whatever,

00:13:08   it is what it is.

00:13:10   However, on my iPad, as I've said numerous times in the past, I have the free 200 megs

00:13:17   a month from T-Mobile.

00:13:19   And that's typically how I use cellular on my iPad, if I'm

00:13:23   going to use it at all.

00:13:24   And occasionally, if I'm on vacation, for example, I'll

00:13:27   actually pay for a few gigs for that month or whatever.

00:13:30   But generally speaking, I just live on the 200 megs a month.

00:13:35   And so I have gone in on my iPad and turned off cellular

00:13:38   data at the iOS level pretty much everywhere.

00:13:41   And I can assure you that if I saw that dialogue all the

00:13:45   time when using Overcast, it would drive me nuts.

00:13:46   So I think you need to stick around with the nitpicky

00:13:49   detail version.

00:13:51   And I see no reason not to put sync in there--

00:13:53   or excuse me, stream in there.

00:13:56   Especially since any time you're streaming, from what

00:13:58   I've gathered from you, any single time you're streaming,

00:14:01   like you've said a couple times, it's

00:14:03   based on user action.

00:14:04   So at that point, I should know what I'm doing.

00:14:07   And I should know, no matter how advanced a user I am, if

00:14:11   I'm trying to play a podcast that is not on my device, it's

00:14:14   going to have to come from somewhere.

00:14:16   and if I'm not on Wi-Fi, if I'm not on the fan,

00:14:19   as we jokingly call it around these parts,

00:14:22   then I know that it's gonna have to go via cellular,

00:14:26   and I should be able to figure that out.

00:14:29   So I think you've got the right approach,

00:14:30   but I'm curious to hear what Jon has to say.

00:14:32   - Well, and there is one little complexity that you said,

00:14:35   as long as you're streaming and you always chose that,

00:14:37   there are scenarios in which you can unexpectedly stream.

00:14:41   - That sounds like a condition.

00:14:43   - Yeah. (laughs)

00:14:44   Wow, all right.

00:14:45   Oh, my so, for instance, suppose you are on a trip that you know, suppose you

00:14:52   have it set so that new episodes come in in streaming mode, so that basically you

00:14:56   disable auto download right so so new episodes come in and their marks for

00:15:00   streaming only they're not automatically downloaded by default. Okay, so you have

00:15:04   a set that way.

00:15:04   You are listening to a playlist and you've downloaded everything on the

00:15:08   playlist anyway because you were going to go somewhere and so you're like a lot

00:15:11   of want to burn all my data when I'm on this trip. So I'm going to download

00:15:13   everything in advance.

00:15:14   While you are listening to something and the screen is off,

00:15:17   a new episode comes in, it's marked for streaming by default

00:15:21   and it's inserted right after the thing you're listening to.

00:15:23   So as you're listening, the episode that you're listening to

00:15:26   ends, it starts up the next episode,

00:15:29   which is a streaming only episode.

00:15:31   So it is possible for streaming to happen to you

00:15:35   somewhat unexpectedly.

00:15:37   It's not gonna be the common case, but it is possible.

00:15:39   - It's like the web, it just happens to you.

00:15:41   - Right. (laughs)

00:15:43   - Exactly, so you see like there are lots of little

00:15:45   complexities, like a lot of people said,

00:15:48   why don't you just pop up a dialogue asking people

00:15:50   to approve streaming when you're on cellular

00:15:52   or have an option to do that.

00:15:53   The problem is, lots of times when I have to make

00:15:57   that decision, the display of the phone is off

00:16:00   'cause it's in somebody's pocket or it's in somebody's dock

00:16:02   somewhere in a car or in their house and a lot of times

00:16:04   like if you're driving, I really shouldn't be asking people

00:16:07   to interact with the UI, that's dangerous.

00:16:10   or you're like exercising and it's in your pocket.

00:16:15   There's all these situations where I have to make

00:16:19   a decision without asking the user with a dialog box.

00:16:23   So I think the way to do it is to just have,

00:16:26   is to just never prompt the user for this decision.

00:16:29   You just have a setting that you can change,

00:16:31   either on or off.

00:16:32   - I don't know, John, what do you have to say about this?

00:16:34   - You talk about all these settings makes me think

00:16:36   that like the key point here is not the raw count

00:16:40   of little toggle switches you have in your settings.

00:16:42   I know there's sort of a combinatoric explosion

00:16:45   of possibilities that you have to debug,

00:16:47   so keeping the counter under control

00:16:48   is a good first approximation

00:16:50   of how you should keep your app from going crazy.

00:16:52   But for the cellular stuff, setting aside the bugs,

00:16:55   which, you know, what can you do?

00:16:56   Just file the radars and keep your fingers crossed.

00:16:59   The settings that come to mind

00:17:02   when you're describing all this functionality

00:17:04   that I can imagine users feeling comfortable with,

00:17:06   and the way I'm conceptualizing this,

00:17:08   somewhere in the settings screen,

00:17:09   There's gonna be a list of either split into Wi-Fi

00:17:13   and cellular or just do cellular if you're not gonna

00:17:16   disable anything in Wi-Fi.

00:17:17   Here are the things I want this app to be allowed

00:17:19   to do on cellular.

00:17:20   And it's just a fairly long list of things.

00:17:23   I wanna say yes to that, no to that, yes to that.

00:17:25   You can get increasingly specific about them.

00:17:27   You could start off with a few number.

00:17:29   That's when I hear like stream and sync over cellular.

00:17:32   I'd rather just see a cellular section.

00:17:33   What am I allowed to do over cellular?

00:17:35   Download new episodes.

00:17:36   Stream episodes that I've asked you to stream.

00:17:38   download, you know, stream episodes that come in while I'm listening to it, but I don't

00:17:42   know what the options are.

00:17:43   But it's like, you can go through all the possibilities and just, because I think, yeah,

00:17:46   that's a lot of options and it's like, who wants to answer all these questions?

00:17:50   You're going to come up with some set of defaults that you think are appropriate for most people.

00:17:53   But when someone goes to settings, I can imagine them going down the list and going, yes, no,

00:17:58   no, yes.

00:17:59   And if you want to help them, you could even put like average of X number of megabytes

00:18:03   over the past month, which probably won't help them initially because you can't pre-populate

00:18:06   that because you don't know what the usage is unless you collect it for everybody.

00:18:09   But after a month of usage, they can go down that list and say, "What's using all my cell

00:18:12   data?"

00:18:13   It's like, "I want Overcast to..."

00:18:15   You know, they can only guess, like, "What things?

00:18:17   I don't understand.

00:18:18   Does this take up a lot of data or not?"

00:18:19   And you can see over the past month, because you allowed us to download new episodes over

00:18:23   cellular, it's used this amount of data, right?

00:18:26   And then they can, you know, just a bunch of switches going, "Yes, no, no, no, yes,

00:18:29   no."

00:18:30   I feel like six, seven, eight, nine, even 10 options and that type of thing isn't crazy,

00:18:33   especially when you look at the actual iOS settings screens,

00:18:36   they're just giant walls of toggle switches

00:18:37   and people are okay going, who can use location data?

00:18:40   Not you, you, not you, you, you, not you.

00:18:42   Like, I think that's a reasonable interaction,

00:18:46   even though it seems really complicated.

00:18:47   If you're gonna have this complexity anyway,

00:18:49   because you've got to have all these conditionals

00:18:50   in the code of like, when do I do this?

00:18:51   When do I do that?

00:18:52   And you're gonna pick defaults for them,

00:18:54   just throw in the switches, bury it in a section

00:18:56   if it makes you feel better.

00:18:57   But I think that is actually the most straightforward way,

00:19:00   As long as you sort of organize these fleets of switches

00:19:03   into logical groups, then people feel like they're,

00:19:06   you know, that each one, because like there's a topic,

00:19:09   the topic is, I don't know how you'd phrase it,

00:19:12   but like this is what Overcast is allowed to do

00:19:14   over your cell connection.

00:19:15   Like when they're in that frame of mind,

00:19:17   I feel like they're in the mode to go down those questions

00:19:21   and just sort of, you know, give yes or no's,

00:19:23   rather than just going down a list of options

00:19:26   and having to read each one and see what the topic is,

00:19:28   what is it discussing, and what is the decision I have

00:19:30   make them go to the next one which is entirely unrelated what is this one

00:19:32   talking about what is it discussing you know whatever I don't know I maybe I

00:19:37   haven't thought of all the options but of all the ones you listed so far I'm

00:19:40   like those are all reasonable things that people could conceivably want

00:19:45   settings for but I think there's no way to to express that in the UI without

00:19:51   just a really big long list of options I think when you try to combine them into

00:19:55   a single line item, it gets confusing because it's less clear what's going on or people

00:20:01   might want different options for the two things that you've combined into one of your settings.

00:20:05   - Well, and there's more complexity to it than that.

00:20:08   It isn't as simple as do I have combo settings or do I have a whole bunch of settings and

00:20:13   a big long list?

00:20:14   'Cause one of the biggest problems is if presented with a big long list, there is a cost to having

00:20:20   that in the app in that people will see that screen.

00:20:24   will confuse them and it will give them the general impression that either this app is

00:20:29   too complex for my taste or I don't understand this app and therefore I don't feel good about

00:20:35   it and it's, you know, people don't like to be made to feel dumb or to be confused.

00:20:38   Well that's why you're burying it. You bury it in the advanced or nitpicky or whatever

00:20:41   section like I don't think anyone's gonna go to settings period unless the app doesn't

00:20:44   do what they want by default so it's all about picking the good defaults as you discussed.

00:20:47   I think you have a good handle on what the reasonable defaults are. You just want to

00:20:51   have a go-to section, and when someone gets into that mode where they feel like this app

00:20:55   is doing something that I don't, that's why I mentioned the sizes on the things, because

00:20:59   what they really want is stop using so much of my data.

00:21:02   And if they just see a bunch of switches, or even one or two options, even if it's just

00:21:06   two options, like which one of those, if I change, will make it use less of my data.

00:21:10   That's why stats on like, because this thing has been enabled, we have downloaded this

00:21:15   number of things over your cell connection over this period of time.

00:21:18   even if it's just two settings and two numbers,

00:21:20   then go, okay, well that's the one that's using my data,

00:21:22   so I wanna turn that one off.

00:21:23   - Right, and I probably, stats are tricky,

00:21:27   because I don't know if I can really guarantee them.

00:21:30   Like, I think it'd be hard to measure

00:21:33   total bytes used accurately.

00:21:35   There's things like redirects and header sizes

00:21:39   and everything that like, you know,

00:21:41   I'd have to go very low level in the APIs

00:21:43   to be able to actually count all of those accurately.

00:21:45   So like, I wouldn't, I don't really want to be

00:21:48   game of being incredibly specific.

00:21:50   'Cause also there's some degree of liability there.

00:21:53   And maybe not, I'm sure legally it wouldn't cause me

00:21:55   huge problems, but it might anger people.

00:21:57   If I say something that is wrong.

00:22:00   So if I say, for instance, if I have in the app

00:22:03   a big toggle that says all cell data off,

00:22:07   I don't offer that.

00:22:08   Because I can't guarantee that my app will use no cell data.

00:22:11   The iOS at the system level option can.

00:22:14   I can't, because there are things like WKWebView.

00:22:17   and a web view that tries to load images.

00:22:20   I can't prevent that from hitting the cell network easily.

00:22:23   Like I can't guarantee that my app will use no data,

00:22:26   so that's why I don't offer that option.

00:22:27   Also, I don't show the file sizes of the podcasts

00:22:32   before they're downloaded.

00:22:33   A lot of people request this.

00:22:34   The big problem with that is that it basically would

00:22:37   require doing a head request in the file

00:22:38   before you download it, and like right before you download it

00:22:41   to really get it accurately.

00:22:43   'Cause there is a field for in feeds

00:22:45   where people are supposed to specify that.

00:22:46   we do in our feed, but that is manually entered by a human being. So it is unreliable and

00:22:52   it is often absent. So the only way to do it is to do a head request and I can do that

00:22:55   server side for every single thing that's complicated and that's also kind of unreliable

00:23:00   because then what if the server, you know, what if one of these things blocks me, then

00:23:03   I report zeros for everything. What if my information is out of date, where like the

00:23:06   first crawl, maybe it was just one megabyte because they uploaded only part of the file

00:23:10   and then you go to download it, you know, two hours later and it's a hundred megs and

00:23:14   and then you're angry at me 'cause I said it was only one meg.

00:23:16   There's all these weird complex edge cases

00:23:19   that I don't want to make a promise I can't keep.

00:23:23   - Well, you can fudge it.

00:23:24   Like the people who say in the chat room,

00:23:25   low, medium, high, percentages.

00:23:27   I'm trying to just reconceptualize this whole cellular thing

00:23:30   from a user's perspective.

00:23:32   What they're concerned about is,

00:23:34   they can get the idea from the settings thing

00:23:37   of which apps are using most of my data,

00:23:39   like Apple provides that, but then if they find out,

00:23:42   Okay, Overcast is a big consumer of things.

00:23:44   I subscribe to a lot of podcasts

00:23:45   and the settings app says it does a lot.

00:23:48   I like podcasts, I don't want to uninstall Overcast.

00:23:51   I want to continue to listen to them,

00:23:52   but now I have to go into Overcast

00:23:53   and say, how can I make Overcast use less data?

00:23:56   And other than the big giant switches

00:23:57   that says no cell data at all,

00:23:59   they're going into Overcast with a mission.

00:24:01   And the mission is figure out what little thingies

00:24:04   I have to flip to make it use less data.

00:24:07   And anything to guide them in terms of

00:24:10   which one of these options will have the most effect

00:24:12   rather than relying on people reading the descriptions

00:24:14   and understanding which one is bigger.

00:24:17   And you know, it can't give sizes in megabytes.

00:24:19   Maybe you can give percentages,

00:24:20   maybe give low, medium, high, I don't know.

00:24:23   I'm just, that's what I'm trying to think.

00:24:24   But like, rather than worrying about it

00:24:27   from the people who are like super picky

00:24:28   and wanna control every aspect of your app

00:24:30   and wanna like be able to write their own code

00:24:32   to make decisions on each decision point,

00:24:34   most people just wanna know how to make the app

00:24:38   use less data and know what the consequences are.

00:24:41   Here's what the option says,

00:24:45   here's some sort of rating for if I were to turn it off,

00:24:47   how much might I save,

00:24:49   and then I'll make that decision to understand

00:24:51   because I turned that off,

00:24:52   now I won't get that anymore from on a cell network.

00:24:53   I don't know.

00:24:54   It is a hard problem,

00:24:55   but I'm trying to simplify it in two ways.

00:24:58   One, changing the way you think about it

00:24:59   with terms of what is the goal?

00:25:01   The goal is to use less data

00:25:02   and I feel like sizes or ratings

00:25:03   or something like that have to be in there.

00:25:05   And then two, the other extreme is, you know,

00:25:09   just introduce the topic,

00:25:11   give a big fleet of things that can go

00:25:12   and then people will just go down the list

00:25:14   and go yay or nay.

00:25:16   And like that sync over cellular thing,

00:25:18   I've never even seen that option.

00:25:19   So I don't, you know, like I'm trying to think of my usage.

00:25:21   I don't think I've ever been to the overcast setting screens.

00:25:23   Maybe it's because you just happened to pick the defaults

00:25:25   that are right for me.

00:25:26   Maybe it's because I'm always on wifi, but yeah.

00:25:30   The people who write you in to complain about cell usage,

00:25:35   They may be noisy, but I think most of your job is just picking good defaults and then

00:25:40   after that I think you have to worry less about the intimidation of your settings screen

00:25:43   because I don't think many people even visit it.

00:25:45   Maybe.

00:25:46   Well, the other problem is, this is what I affectionately call the power user problem,

00:25:51   that if you give people settings they will use them and then they will forget that they

00:25:56   use them and then the app will behave differently from the default because they change settings

00:26:02   and then they forgot that they changed them.

00:26:04   And then they will write in or complain on Twitter

00:26:06   or complain in public that my app is not working properly

00:26:08   because of a setting they changed.

00:26:10   This happens all the time

00:26:12   with the continuous playback setting,

00:26:15   where in the playback effects pane thing,

00:26:19   I have an option that most people would consider

00:26:22   called continuous play, and I title it

00:26:24   When This Episode Ends, and at the bottom,

00:26:27   I have two buttons, Play Next or Stop.

00:26:30   And I thought this was very clear.

00:26:33   Many people, when they're poking around in the options,

00:26:36   they will toggle that.

00:26:37   Then they will forget that they toggled that.

00:26:39   And then they will write in saying,

00:26:41   my app used to just automatically move

00:26:42   to the next episode, and now it doesn't.

00:26:44   What happened?

00:26:45   It's a bug.

00:26:46   (laughing)

00:26:47   And every time I have to explain,

00:26:49   this is kind of like a blow on the power cord

00:26:52   kind of solution.

00:26:53   It's like, I don't want to embarrass them in my response.

00:26:55   So I have to be very gentle in how I say this.

00:26:58   (laughing)

00:26:59   but it's just a bad thing for everybody

00:27:01   because if you give people settings, they will use them.

00:27:03   Many of them will forget about it or use them badly

00:27:05   or in ways they don't understand.

00:27:07   And then the app will break for them.

00:27:09   And regardless of what it does to me or my reputation,

00:27:11   it makes the app suck for them.

00:27:13   So that's also something I want to avoid.

00:27:15   - Aren't they happy when you gently lead them to the option?

00:27:19   Aren't they like, oh, my problem is solved.

00:27:21   I found a setting, I've plus a little title switch,

00:27:23   and now I feel like, I mean, I feel like it turns around

00:27:26   on them at that point where it's like,

00:27:28   I thought that this app that I used to like was ruined for me forever, but really all

00:27:32   I needed to do was tap a little toggle switch.

00:27:34   Yes, but a lot of times those people are not writing in.

00:27:38   So like this is all, these are always indicative of like, well, if these five people on Twitter

00:27:43   said this happened to them and they didn't understand that's why it happened, think of

00:27:47   how many people there are who didn't get in touch with me, who just thought the app was

00:27:51   broken.

00:27:52   Why do people like to, I don't understand why someone would turn that off while they're

00:27:55   wandering through settings or why they'd be wandering through settings.

00:27:58   I've also never seen this setting.

00:27:59   I don't even know, I guess the default is to go

00:28:01   to the next track, 'cause I think that's what mine does.

00:28:03   - Oh, Jon, you are a responsible power user.

00:28:06   You are the power user that developers wish

00:28:08   all power users were like.

00:28:09   But unfortunately, that's not the case.

00:28:12   And there are so many people who,

00:28:14   if you give them a way to customize or change something,

00:28:17   they will, and they will always demand more of those things,

00:28:20   and then they will become support problems for you.

00:28:22   - You know what, don't go to settings,

00:28:23   I just launched the app just to figure out.

00:28:25   It's because it's not a gear icon.

00:28:26   and I'm like, where the hell is settings in this thing?

00:28:28   We've been through this before.

00:28:29   No wonder I never go to settings, it's hidden.

00:28:31   - That was one of those things, like I said,

00:28:34   can I get away with not using a gear icon?

00:28:37   - You can, it keeps me away from the settings.

00:28:39   - That's why maybe, you know,

00:28:42   regular people are so confused by everything

00:28:45   because everything is horrible,

00:28:46   that they'll push anything and not think about it.

00:28:50   Power users look for a gear, and they don't see a gear,

00:28:53   they just assume, well, I guess there's no settings.

00:28:54   - Yeah, so that's perfect.

00:28:55   Maybe I should make all my icons make less sense.

00:28:58   - Have it squirm out of the way when you try to tap it.

00:29:00   - Wasn't that a thing?

00:29:01   - Yeah, like on Windows.

00:29:02   - You can't do proximity detection with the...

00:29:04   - Yeah.

00:29:05   - Wait for the iPad that has the stylus support,

00:29:08   then you'll have proximity detection.

00:29:09   - Yeah, for all my iPad users, yeah.

00:29:12   - Totally, hey, I'm one of them.

00:29:13   So before we leave this topic--

00:29:14   - It's like less than 5%.

00:29:15   - Yeah, well, before we leave this topic,

00:29:18   maybe you mentioned this and I blanked,

00:29:20   but why not just have one universal,

00:29:23   can I do anything on cellular, yes or no?

00:29:26   - Because I can't enforce that, basically.

00:29:30   Because that's why, like I said earlier,

00:29:32   if a web view for show notes loads an image,

00:29:35   there's stuff that is very hard to enforce.

00:29:38   - Well, you could rephrase it.

00:29:40   Can I download any podcast-related materials over,

00:29:45   well, that's okay, that's still a little weak,

00:29:46   but you know what I'm driving at?

00:29:47   You could wordsmith it to get the wording right.

00:29:50   Do you just think that's not granular enough

00:29:53   for your average user?

00:29:55   - Well, I think people either want the app

00:29:57   to literally use zero, in which case they have

00:29:59   to use the iOS system toggle,

00:30:00   which they probably already have

00:30:02   without even looking in the app,

00:30:03   'cause most apps don't have these options.

00:30:04   Most apps just do it, and it's up to you as the user

00:30:07   to minimize the usage or to go in settings and disable it.

00:30:10   So if somebody wants to literally use no data,

00:30:15   they will use the system pane for that.

00:30:17   I think some data, 'cause every app

00:30:21   that syncs with the web or get status or info from the web,

00:30:25   like you're losing a few hundred kilobytes here and there

00:30:29   to all these apps and you just have to not care about

00:30:33   that level of usage if you're gonna have cell data enabled

00:30:36   for anything.

00:30:37   So some data usage is fine, but again,

00:30:40   I don't want to make a promise that I can't really keep.

00:30:44   And especially with something like this

00:30:45   where making a mistake here could cost people money.

00:30:48   - Yeah, yeah, I guess that makes sense.

00:30:50   It's a tough thing for sure.

00:30:52   But before we move on to start the show with follow-up,

00:30:57   you should probably tell us about something that's cool.

00:31:00   - I agree.

00:31:01   So this week, we have back long-term sponsor,

00:31:03   lynda.com, L-Y-N-D-A dot com.

00:31:07   lynda.com is an easy and affordable way to help you learn

00:31:10   with high quality, easy to follow video tutorials.

00:31:13   You can instantly stream thousands of courses

00:31:15   created by experts on software, web development,

00:31:18   graphic design and more.

00:31:20   Go to lynda.com/atp, that's L-Y-N-D-A dot com slash ATP

00:31:25   to see for yourself.

00:31:27   lynda.com has fresh new courses added all the time.

00:31:31   They work directly with the industry experts

00:31:33   and software companies to provide timely training,

00:31:35   often the same day new versions or releases hit the market

00:31:38   so you can always be up to speed.

00:31:40   If you're a developer, they have programming languages,

00:31:42   they have different skills, media production,

00:31:45   they have so much stuff, and they have all these courses

00:31:48   that are offered at all experience levels,

00:31:50   whether you're beginner or advanced.

00:31:52   These courses are offered at the highest quality.

00:31:54   This is not like the homemade YouTube videos,

00:31:57   which can be very inconsistent in their quality level

00:31:59   and it can be pretty hard to find good stuff.

00:32:01   Every lynda.com course is very high quality

00:32:04   and just high production value, everything.

00:32:06   Courses are also broken up into bite-sized pieces

00:32:08   so you can learn at your own pace

00:32:10   and learn from start to finish or just find a quick answer.

00:32:12   You have all these choices.

00:32:13   They have little transcripts on the side,

00:32:15   you can scan through the transcript, you can search,

00:32:17   You can jump right to the point they're talking about when you find something.

00:32:20   It's really an incredibly advanced platform with very well produced videos.

00:32:24   So you can even learn while you're on the go with lynda.com apps for iPhone, iPad and

00:32:28   Android.

00:32:29   And you can, you know, the great thing about lynda to me is that you can watch whenever

00:32:33   you want with no pressure or commitment because you don't pay per video.

00:32:37   You pay one low monthly price of just $25.

00:32:41   This gives you unlimited access to over 100,000 video tutorials.

00:32:45   I learned things like, for example, editing this podcast.

00:32:48   My podcast editing skills improve dramatically

00:32:52   when I watch the videos on Logic and on Level Compression

00:32:54   and EQ and Voice, really great courses here on lynda.com.

00:32:58   So they also offer an annual premium plan.

00:33:00   If you wanna download courses for offline viewing

00:33:02   or download sample projects and practice along

00:33:04   with the instructor, really great stuff here.

00:33:07   Again, they have programming, they have management skills,

00:33:10   creative apps like Adobe's Creative Suite, Logic, Final Cut,

00:33:14   negotiation skills, all sorts of great courses

00:33:17   that almost everybody can find something here

00:33:19   that they would like.

00:33:20   So anyway, lynda.com is offering a 10-day free trial

00:33:24   to access all courses for free

00:33:25   if you visit lynda.com/atp.

00:33:28   Once again, that is a 10-day free trial

00:33:31   with access to all courses, lynda.com/atp.

00:33:36   Thanks a lot to lynda.com for sponsoring our show once again.

00:33:38   Highly recommended.

00:33:39   - All right.

00:33:42   So now that we're, what, 30, 40 minutes in,

00:33:44   let's do some follow-up.

00:33:45   (John laughs)

00:33:47   - See what you've done here with this pre-follow-up?

00:33:49   You've just broken the show.

00:33:50   (John laughs)

00:33:51   Not a complicated format.

00:33:53   - No. - It's really straightforward.

00:33:55   - All those years of follow-up ruined.

00:33:58   - Just like that, just like that, I'm sorry.

00:34:00   Some men just like to see the world burn.

00:34:02   - Yeah, it's true.

00:34:03   - That's a reference, John.

00:34:05   Except I think that's actually watch the world burn,

00:34:07   but that's okay.

00:34:08   - Is it, wait, is this like Nero or?

00:34:10   Yeah, Marco just learned about the cowbell SNL skit today.

00:34:15   So he's coming down.

00:34:17   So when so no, this is how this went.

00:34:19   I wanted to find myself here.

00:34:20   Tiff said she made like the cowbell fever joke, and I knew there was a joke about more

00:34:25   cowbell and I wasn't quite sure where it was from.

00:34:27   So I said, oh, is that from Spinal Tap?

00:34:30   And it turns out no, but I at least knew that the joke was about having more cowbell in

00:34:35   that Blue Oyster Cult, whatever song that is.

00:34:40   The thing is that it's something that you heard about but didn't know the original of.

00:34:45   That's correct.

00:34:46   That's what we're getting at.

00:34:47   It's not whether you get references, it's that you haven't actually experienced the

00:34:51   original piece of media from pop culture.

00:34:56   I can't believe you didn't recognize that quote from Dark Knight.

00:34:59   Now I know how Jon feels with me.

00:35:01   You know, I've seen that.

00:35:02   So good.

00:35:03   It's one of my favorite movies.

00:35:06   Oh, God.

00:35:07   Anyway, so we should probably correct our incorrect statements about Half-Life for like the third week in a row.

00:35:13   Half-Life just keeps coming. It's the real one is out for the Mac.

00:35:16   Has been for a long time, not just the source port, but the plain old cruddy-looking original Half-Life.

00:35:22   So Half-Life is completely and fully available for the Mac.

00:35:25   I can't believe this game came out when I was like 16 and we're talking about it still.

00:35:28   Kind of like Halo, also completely available for the Mac. Too little, too late.

00:35:33   Wait, what? The original Halo is on...

00:35:35   That's that's where I played the original hell I played it on the Mac what I genuinely did not know that I

00:35:40   Assumed it was always Xbox one. Oh, thank God

00:35:43   I was gonna email you I know cuz I knew there was like a history with like bungs

00:35:47   He was making it for the Mac and then Microsoft bought them and then they canned it, right?

00:35:50   Yeah, you guys I guess you guys weren't in Apple circles when they were when they the Halo trailer was shown to Macworld or whatever

00:35:56   That was no was it Macworld? Yeah, I guess I'm had to been a Macworld. Anyway on stage

00:36:03   Look at this amazing new game coming for the Mac. I don't want to talk about it

00:36:06   It hurts too much to talk about it's dark time. Yeah

00:36:10   All right, so thank you Christian for that correction

00:36:16   This is from hunter mentioned last week about on-demand resources

00:36:21   Where if your app doesn't come with a bunch of stuff it can download them on demand and you can sort of spec that out

00:36:26   When you build your application which resources?

00:36:28   Won't be bundled with the application but can be downloaded by the application

00:36:33   at any point and I mentioned that it didn't seem like there was enough UI hooks to give

00:36:38   a good user experience.

00:36:39   The Hunter says that the on-demand resource API does support NSProgress for tracking progress

00:36:44   and displaying it to the user.

00:36:45   So that's nice, although you think you have it bad.

00:36:49   Think about games that have like 500 megalevels that download levels over cellular?

00:36:54   Yes, no.

00:36:55   What's the right answer there?

00:36:56   Probably no, but then if they're playing a game in a car ride and they get to the next

00:37:00   level, the only thing you really do is say, "Sorry, you can't play the next level because

00:37:05   you have cellular downloads, blah, blah, blah, blah, blah," and the kid immediately goes

00:37:08   in, changes his setting, turns it on, and uses all of his parents' data. Or just waits

00:37:12   a really long time for the next half an hour in the car for the 500 meg level to download

00:37:16   while he travels along the highway. Anyway.

00:37:17   - Right. Or any game that uses it gets so many one-star reviews from people having this

00:37:20   problem that it stops using it.

00:37:23   - Yeah. But I think it's still a good option. You just gotta break it up into smaller chunks,

00:37:28   you know, like, instead of, hopefully no levels 500 megs, but if games are like 3 gigs and

00:37:35   they have a small number of levels, each one's got to be pretty big.

00:37:38   Anyway, I hope game developers figured this out.

00:37:40   I kind of like the PS4 thing where if you download a game you can start playing it before

00:37:46   the game is entirely downloading because they just download the early parts of the game

00:37:49   sooner.

00:37:50   It still takes a long time to download, but it's nice that you don't have to wait for

00:37:52   the whole thing to download because you're not going to outrun it by playing.

00:37:56   Like some of these games take a certain minimum amount of time to get through, especially

00:37:59   when you're playing for the first time.

00:38:00   They just let you start playing as soon as you've got like the first, you know, hours

00:38:04   worth of levels or something.

00:38:05   And by the time you grind through those, even if you're an expert in the game, the rest

00:38:08   will be downloaded, or so they hope.

00:38:10   Yeah, I don't see this working out.

00:38:13   We'll see.

00:38:14   Someone's gonna implement it.

00:38:16   I do think it's interesting also that like there seems on the iOS app store, there seems

00:38:20   to be no correlation between game size and commercial success.

00:38:27   Okay. Think about it. Man, you guys are off tonight. Either everyone else is crazy or

00:38:35   I am. Alright, so keep going. You expect there to be? Like, you expect big

00:38:40   games to be more successful? Yeah, I'm kind of saying like maybe it's not worth making

00:38:44   games so big. Well, it really depends on the type of game.

00:38:48   Like if you're making a puzzle game, how big could it possibly be, right?

00:38:51   I mean, talk about having no assets, how big is letterpress?

00:38:54   I don't know if there's any graphics in that entire game.

00:38:56   There's just like the launch icon.

00:38:57   That's it.

00:38:58   Right.

00:38:59   But other types of games, like if you're making a 3D game, there's going to be geometry, there's

00:39:02   going to be textures, and like there's no getting around that, and it's going to be

00:39:05   much bigger than a game with a static screen.

00:39:09   I don't know.

00:39:10   I don't think people have the same expectations of mobile games, but I think genre plays into

00:39:16   how big things are.

00:39:18   Art style if you do things with texture mapping textures take up room if you do everything with flat shading you just have geometry then

00:39:24   You know I?

00:39:25   Think it's reasonable on a small platform like this desert golf not be enough for anybody

00:39:29   Yeah, you love that game so much. It's really good. I haven't played that since the Alto came out

00:39:36   I realized so I still haven't installed, but I'm stuck at whatever yeah me too

00:39:40   I'm not stuck stuck if I played it. I will continue, but it like it's not not making progress

00:39:44   But I do not install because I'm afraid that if I delete it. I'll lose my progress

00:39:48   I don't know if that's actually true, but I'm afraid it is yeah

00:39:51   I've never played it, and I'm pretty happy about that get out those adventure. It's better game

00:39:55   I have it. I've had it for weeks still never played it

00:39:57   Disagree you think you don't think out those adventures better than desert golfing I

00:40:02   Think everyone else thinks it is, but I don't think it is

00:40:05   This is like you as a gamer in a nutshell

00:40:12   It's too much like a game. I don't like that gamey game part.

00:40:15   Can you put me in a desert with a ball? That's all I can handle.

00:40:17   No, I mean, they're both good games.

00:40:19   I would say they're both excellent games, but I got a lot more time out of desert golfing

00:40:24   just because of the kind of zen commentary qualities I got out of it,

00:40:30   where Alto is a great game, but once I played, like, I played it on one plane flight,

00:40:38   and then I was like, "All right, well, every time I play this game, it's basically the same thing."

00:40:42   over and over again.

00:40:44   And I know desert golfing is similar,

00:40:45   but for whatever reason,

00:40:47   I didn't get any depth out of Alto.

00:40:49   So it was cool for the one day I was playing it a lot,

00:40:53   but then I tried playing it here and there afterwards,

00:40:54   and I just like, okay, it just felt too repetitive to me.

00:40:57   - You don't have any competitive juices flowing.

00:40:59   Alto is all about competing with yourself,

00:41:01   trying to best your previous best run,

00:41:03   or competing with other people,

00:41:04   whereas there's no competitive aspect

00:41:06   in desert golfing, really.

00:41:08   - That's what makes desert golfing so great,

00:41:10   is that it really shows you how pointless

00:41:13   everything else in the universe is.

00:41:16   - This is a man whose wife plays all console games

00:41:18   in hard mode, by the way.

00:41:19   - Yes.

00:41:20   - Could not be more different.

00:41:21   - That's correct, yeah.

00:41:21   No, she is the gamer, I am totally not.

00:41:24   She's playing Prison Architect now, thanks to CGP Grey.

00:41:27   - That sounds awful.

00:41:29   Like, it just sounds like work to me.

00:41:31   - I mean, it looks, it's like Theme Hospital, you know?

00:41:33   Like, it's one of those kind of games.

00:41:35   Like, you know, it's basically Sim Prison.

00:41:37   If I was like more of a gamer right now,

00:41:39   I would probably play it too,

00:41:40   'cause it looks pretty cool.

00:41:42   I just don't, I can't, you know,

00:41:44   like I'm doing great with work stuff,

00:41:46   with Overcast and with the show and everything else,

00:41:49   and I'm like, the last thing I wanna do

00:41:51   is take time playing games.

00:41:52   Like I just, I don't, whenever the opportunity comes up

00:41:55   to spend time doing something,

00:41:56   I'd rather work than play games.

00:41:59   - And can you tell everybody exactly what level

00:42:01   you're on in Desert Golfing?

00:42:02   So they can add some context to your previous statement.

00:42:06   - Yeah, give me a sec.

00:42:08   It's been a while that played it so I gotta take it up the man who doesn't want to spend too much time playing games

00:42:12   Go ahead. What level?

00:42:14   3054 all right, and I've been stuck on this one for a long time. That's

00:42:18   3054 holes people yeah, I think we've solved that I don't know though. I haven't seen prison architect, but

00:42:25   So clearly I'm unqualified to talk about it with that said let me talk about it

00:42:29   It sounds like so much work

00:42:31   And I stopped doing work like that at Sim tower like Sim towers when I pretty much stopped doing work

00:42:37   for games. I remember vividly trying World of Warcraft when it was relatively new and

00:42:44   all I did was run around killing like boars or something like that and I lasted two hours

00:42:49   before I decided I'm just doing work. This isn't a game, this is just me doing work.

00:42:53   Eric Michael Rhodes President Architect is more like Transport

00:42:56   Tycoon than WoW. I mean…

00:42:59   Jared Ranere Yeah, it's more like the building sim games,

00:43:02   But even, I love Transport Tycoon.

00:43:06   I haven't played it in probably five years.

00:43:08   - Oh yeah, I haven't played it in more than that.

00:43:10   - Right, because like, and OpenTTD is awesome.

00:43:13   Like it was such a fantastic re-implementation

00:43:16   of the game engine and added so much stuff

00:43:18   and it's fantastic.

00:43:19   I lost a lot of time to it like, you know,

00:43:21   in 2007 or 2008 kind of era, but again,

00:43:25   I just, like there's never a time when I want to spend

00:43:29   hours and hours and hours playing games

00:43:31   instead of producing something that I want to be working on.

00:43:35   And that's just me, that's just like,

00:43:36   it's what motivates me right now.

00:43:38   And maybe that'll change over time.

00:43:40   I'm sure it'll be different once my kid's old enough

00:43:42   to enjoy games and I wanna do it with him.

00:43:44   But for now, I just don't care about games.

00:43:48   - Over 3,000 holes, people.

00:43:49   Over 3,000. (laughing)

00:43:52   Doesn't wanna spend a lot of time playing games.

00:43:54   He'd much rather be doing something else.

00:43:55   - That was almost a meditation for me.

00:43:58   Desert Golfing, it was-- - Oh, yeah, yeah.

00:43:59   You're lucky that game does not have a count of hours played inside it.

00:44:04   No it doesn't, but I mean I know it took like 8,000 strokes.

00:44:07   I mean that's, that kind of, you can figure some idea, like you know, well if each one

00:44:11   takes a few seconds or whatever, but you know, so obviously I know I spend a lot of time

00:44:14   there but how much time do I spend on Twitter?

00:44:17   It's way more than that.

00:44:18   Alright, what do you have to show for that?

00:44:20   There's no whole count on Twitter.

00:44:21   We don't even know, there's no background color change, no nothing.

00:44:24   No rocks, not a cactus in sight.

00:44:27   My god, I don't even know what kind of things you're seeing today at three whole three thousand

00:44:31   This show has been I had no idea when we started before follow-up

00:44:37   Do it or doing things before follow-up that that would really take this entire show right off the I had an idea

00:44:42   Our second sponsor this week is hover hover is the best way to buy and manage domain names go to

00:44:55   and use promo code passinglane for 10% off your first purchase. I'm so sorry, Casey.

00:45:02   - Hey, buddy.

00:45:03   - When you have a great idea, you want a great domain name that's catchy and memorable.

00:45:06   Hover gives you exactly what you need to find the perfect domain for your idea so you can

00:45:10   get started actually working on it. Hover gives you easy to use powerful tools to buy

00:45:14   and manage domains so anyone can do it, advanced or beginner. And they're actually, they're

00:45:18   known for their incredible support. Not only do they have, you know, email options and

00:45:22   but they also have no wait, no hold,

00:45:24   no transfer phone service.

00:45:26   So you can call them up, and a real live human being

00:45:29   will pick up the phone, not after a big hold time,

00:45:32   they won't transfer you to a million different departments

00:45:34   to figure out who can help you, no.

00:45:36   You call, and a human being picks up the phone,

00:45:38   and they can help you.

00:45:39   If you have a registered domain name anywhere else,

00:45:41   you know that this can be a tricky business.

00:45:44   You know, at best, you get a place that means well,

00:45:48   but their site is really clunky and hard to use

00:45:51   and ugly and the control panel is really clunky.

00:45:54   At worst, it can feel like you're being scammed

00:45:56   by a lot of these registrars.

00:45:57   It can be pretty bad.

00:45:58   It can feel like you're being tricked

00:46:00   into buying upsells and cross promotions and stuff

00:46:03   that just doesn't make you feel good about them.

00:46:05   Hover is above all of that.

00:46:06   Hover is, they respect you.

00:46:09   They respect you as a user, they respect your time,

00:46:11   and they have really good design and really easy usability.

00:46:14   So that they don't try to upsell you with crazy stuff

00:46:17   or try to trick you into buying extras.

00:46:20   They include, anything that should be included for free,

00:46:23   they include for free.

00:46:24   Things like who is privacy, DNS,

00:46:26   they have all sorts of stuff they just include

00:46:28   right there for free.

00:46:29   The control panel is the best domain control panel

00:46:32   I've ever used, and I've used now,

00:46:34   I mean throughout my past life when I had a real job

00:46:36   and through all my personal stuff,

00:46:38   I've used probably 10 different registrar's control panels

00:46:41   at least over the years, and Hubber's is by far the best,

00:46:45   the nicest, the easiest to use.

00:46:46   It's not even close.

00:46:47   If you, like me, have had domains in other places,

00:46:50   and you wanna transfer them into Hover,

00:46:52   they will even, if you want to,

00:46:54   they will take your old registrar's login info

00:46:58   and do the transfer for you.

00:46:59   They call this a value transfer service

00:47:01   that I don't know anyone else who offers this.

00:47:02   And so that way they'll move over everything

00:47:04   without making a mistake, like DNS entries,

00:47:06   stuff like that that's kinda tricky to get right

00:47:08   and has pretty big ramifications if you mess anything up.

00:47:12   They will do it all for you if you want them to,

00:47:14   so you don't have to worry about the stress of moving.

00:47:15   and even just annoyances like dealing with the transfer codes

00:47:19   and the unlocks and all that crap,

00:47:20   they will deal with it all for you if you let them.

00:47:22   So it's great.

00:47:24   They also have email features, just 20 bucks a year

00:47:27   if you want a fully functional email account

00:47:28   on a domain with 10 gigs of storage,

00:47:31   29 bucks a year gets you a terabyte of storage,

00:47:34   and then if you want just email forwarding,

00:47:36   let's say you already have email hosting,

00:47:38   you just want a domain that can receive email

00:47:40   at that domain, just five bucks a year.

00:47:42   So you can keep using the host you already have

00:47:43   if you want that.

00:47:44   Great options here.

00:47:45   Go to hover.com.

00:47:47   H-O-V-E-R.com.

00:47:48   For people who don't understand the word I'm saying.

00:47:50   Go to hover.com and use promo code passinglane this week.

00:47:54   That's passinglane, one word,

00:47:56   for 10% off your first purchase.

00:47:58   Thanks a lot to Hover for sponsoring our show.

00:48:00   - Little bit more followup.

00:48:01   Jon, tell us about Satya Nadella.

00:48:03   - No, I'm saving it for next week.

00:48:04   We're moving on.

00:48:05   - Good cock.

00:48:06   All right, we're done with followup.

00:48:07   - Really?

00:48:09   - This show's out of control.

00:48:10   - We're almost an hour in.

00:48:10   - We are really only an hour in?

00:48:12   We're done with followup?

00:48:13   - Boy, I wonder what made that take so long.

00:48:16   - It must be too much follow up.

00:48:18   - God, I'm staying out of this.

00:48:20   Okay, so are we talking about the end of TikTok

00:48:23   or is that also not for right now?

00:48:25   - Oh, I think that's a good one.

00:48:26   - Oh yeah, I don't know why that has that prefix on it.

00:48:27   I'll fix it.

00:48:28   - So this says in the show notes,

00:48:29   next week end of TikTok,

00:48:31   which kind of makes it sound like TikTok

00:48:33   is ending next week.

00:48:34   - It could be.

00:48:35   - All right, so what is--

00:48:37   - This is the end of the computer age.

00:48:39   Oh, I messed that up too.

00:48:40   Damn it.

00:48:41   It was welcome to the end.

00:48:43   We might as well just stop the show now

00:48:45   'cause we're all out of control.

00:48:47   - Welcome to the end of the Computer Age.

00:48:50   (laughing)

00:48:53   (drumming)

00:48:58   - Jon, try to bring us back if you please

00:48:59   and tell us about TikTok and why it's ending.

00:49:02   - Oh no, the one part I didn't put in the show notes

00:49:04   is the part I always forget.

00:49:05   So I'm gonna wing it.

00:49:06   Intel has this strategy that they call TikTok

00:49:09   where they make a new microarchitecture

00:49:12   And then the next round of chips that they offer for sale is,

00:49:17   oh, I screwed it up already.

00:49:20   (laughing)

00:49:21   - Since 2007, Intel has been operating

00:49:23   on a staggered release schedule

00:49:25   that alternates manufacturing process shrinks, TICs,

00:49:28   with major micro architectural changes, TOCs.

00:49:31   - Yeah, if you don't help me know which is the which.

00:49:33   Sometimes they just do a shrink.

00:49:34   - That would be a TIC.

00:49:36   - And sometimes they change the micro architecture.

00:49:39   - That's a TOC.

00:49:40   - All right, there you have it.

00:49:41   How am I the voice of reason between the three of us?

00:49:44   What is going on?

00:49:45   - I can never, they picked a bad naming scheme.

00:49:47   The point is they alternate.

00:49:48   And so this has been sort of regular schedule.

00:49:50   What it meant was that every other round of chips,

00:49:54   you would get a new process size.

00:49:57   So you'd have 32 nanometer chips,

00:49:59   another set of 32 nanometer chips,

00:50:01   and then you go to the next size smaller,

00:50:02   which I think was 28, or they skipped a 22.

00:50:04   Anyway, we've had 14 nanometer chips

00:50:09   for two generations now.

00:50:11   And so in theory, the next round of chips

00:50:13   should be on their next processor,

00:50:14   which is supposed to be 10.

00:50:16   But Intel has basically said the next round of chips

00:50:19   will not be 10 nanometer,

00:50:20   they'll be a third round of 14 nanometer chips.

00:50:24   Basically because their 10 nanometer manufacturing process

00:50:26   is not ready yet, so.

00:50:28   - Well and also it's worth pointing out

00:50:29   that the 14 nanometer process

00:50:31   is just barely coming online now.

00:50:34   Like it was also, whatever it was saying

00:50:37   was the broad belt delay, much of that,

00:50:39   or maybe all that, I don't know,

00:50:40   Much of that was the process,

00:50:43   the 14 nanometer process being so delayed.

00:50:45   And so they had a huge delay just trying to get this out.

00:50:48   It was about a year late or a year and a half late, right?

00:50:51   - Yeah, and so this is like getting out ahead

00:50:53   of the next one by saying,

00:50:54   we're not gonna do what we did,

00:50:55   but be like, oh, we're kind of on schedule,

00:50:57   but things are kind of late and they'll come out

00:50:59   or whatever, it's like, look,

00:51:00   it's just 10 nanometers is not going to be ready.

00:51:03   But we're also not gonna delay having a new line of chips.

00:51:08   we are going to make one more round of chips on 14 nanometers.

00:51:11   And they had to come up with a new name for them, apparently doing the names of these

00:51:15   things.

00:51:16   It was going to be Skylake, and then it was going to be followed by Cannonlake.

00:51:18   So in between them, the third line of 14 nanometer things is going to be KABY Lake.

00:51:25   Anyone want to take a swing at that?

00:51:27   >> KB?

00:51:28   It's pronounced hover.

00:51:29   >> KB, Kaby, I don't know.

00:51:32   I'm sure it's probably KB.

00:51:34   That's what they're going to call the third round of 14 nanometer chips, which is pretty weird as we've discussed on previous shows

00:51:40   Moore's law can't continue forever because they're as far as modern science knows there is a minimum size that things can be

00:51:48   We're always trying to smash them up and do smaller and smaller pieces to find out what can be smaller

00:51:52   But the rules of physics change in crazy ways when you get really small, so we were not going to

00:51:58   Double the number of things we can fit in a unit of space every 18 months

00:52:03   forever and ever. This does not mean the end of Moore's law, it just definitely needs to slow down.

00:52:10   I think I'm fine with that. I like the fact that Intel is owning up to this rather than just saying,

00:52:17   "Oh, it might be a little bit late," because then they're just, you know, they're gonna end up

00:52:20   shifting a whole year over, and I'm glad that they're gonna do another round of 14-nanometer

00:52:24   chips rather than just delaying forever and ever until 10 nanometers is ready.

00:52:29   Well, but I mean, they basically did that with Haswell.

00:52:33   - I know, and it was bad, wasn't it?

00:52:34   - Yeah, well, I'm using one now.

00:52:36   Right now, if you've bought a Mac in the last couple years,

00:52:41   you've gotten a Haswell, unless you bought

00:52:43   one of the 2015 revisions of only the Retina MacBook 13,

00:52:47   the MacBook Air, or the MacBook One.

00:52:53   The 15 still using Haswell, the iMac is still using Haswell,

00:52:57   the Mac Pro is using the Pentium Pro.

00:53:00   I mean, it's like everything is delayed and old

00:53:05   because this one was so, so delayed.

00:53:08   Like, beginning to 14 was so delayed.

00:53:11   And so what Intel did in the last year and a half,

00:53:14   while Broadwell was being delayed,

00:53:16   they released a kind of like 1.5 version of Haswell.

00:53:19   You know, it's the same core,

00:53:21   but they just like gave us like higher binned,

00:53:23   higher clocked parts just to kind of tide us over

00:53:26   and tide over their partners like Apple

00:53:28   who are trying to sell computers around these things

00:53:29   trying to make decisions around these things. That's why I'm using a 4 GHz computer right

00:53:34   now because like Haswell didn't start out at 4 GHz and it probably, I'm sure Apple's

00:53:38   margins on the 4 GHz chips are not that great because of the various binning and yield issues

00:53:43   that are typical of the business. But they had to give us some kind of minor increase

00:53:48   like all right, take what we have, raise the clocks as much as we can without shrinking

00:53:52   the process and we'll do something. We'll give people something new to tie them over

00:53:58   while we work out our next revisions delays.

00:54:00   So they already did a third revision, kind of, with Haswell.

00:54:05   So what they're basically saying here is that

00:54:08   this is going to happen again.

00:54:10   This time we're just warning you in advance.

00:54:13   - And they're giving it a new name.

00:54:14   And I don't think much is known about this new line of chips.

00:54:16   Like, will it just be, like you said,

00:54:18   just the higher binned versions of the other things

00:54:20   with maybe a couple of little errors fixing them or something

00:54:23   or will it be, you know, like, will they bump the GPU for it?

00:54:27   like, well, this will actually be a significant new line of chips.

00:54:31   It seems like that by giving it a new name, a new lake name, that it's not just going

00:54:36   to be like a 1.5, but it's certainly not going to be a new architecture, a tock, right guys?

00:54:41   It's not going to be a new architecture, I guess.

00:54:44   They're saving that for, I don't know.

00:54:47   It's confusing.

00:54:48   Although, like, the relationship between Intel's roadmap and what Apple releases, in some ways

00:54:53   it seems like it's tied tightly together where you're like, oh, Apple can't release the machines

00:54:56   because Intel doesn't have the chips ready,

00:54:59   but in other respects, they're totally unrelated.

00:55:01   See the Mac Pro.

00:55:02   - Well, no, that's not entirely true.

00:55:04   The Mac Pro is very related,

00:55:05   but it's related to the Xeon roadmap.

00:55:07   - I know, but aren't they already behind

00:55:09   an entire year on Xeon chips?

00:55:11   - They are, but the last revision of Xeons

00:55:13   was a little bit questionable.

00:55:14   The problem that I saw when looking at

00:55:17   what you would do with the Haswell chips,

00:55:19   on the Xeons, Intel sells most of these things in servers,

00:55:23   and the server business is very different

00:55:25   from the desktop workstation business that the Mac Pro sits in. And on servers, what

00:55:29   you really--what you generally want with the market demands is more and more cores, even

00:55:35   if each core isn't as fast. And so that's why you have most of the Xeon lineup, the

00:55:39   ones--at least the high ones that sell nicely and have nice profit margins and have all

00:55:44   the PCI express lanes and everything. Most of those optimize for core count and have

00:55:49   a clock speed of like 2 gigahertz, you know, so that you're getting like relatively low

00:55:53   clock speeds, very high core counts. The Mac Pro, it uses a very small number of these

00:56:00   chips. It offers three or four of them, and they're like the highest clocked versions

00:56:05   they can get at each core count basically without blowing thermals it seems. When they

00:56:10   moved from the previous ones, which was Sandy Bridge EP I think, into Haswell EP, well when

00:56:16   Haswell EP came out, sorry, this is the generation that they've skipped, there really was not

00:56:21   much of a gain to be had in the areas that the Mac Pro

00:56:24   actually sells chips from.

00:56:26   Like in those parts, it really was like,

00:56:29   the core counts went up slightly,

00:56:31   like you could get like a 10 core,

00:56:33   and like a 14 core, and stuff like that.

00:56:34   There was like a small core count upgrade,

00:56:36   but like all the clock speeds went down or stayed the same,

00:56:38   and power was a little bit weird.

00:56:40   So it was not a great update.

00:56:43   And so that might have been why Apple seemingly hasn't,

00:56:46   and seemingly won't use those.

00:56:49   I don't know.

00:56:51   But it also, you know, the Mac Pro also has

00:56:54   skipped generations before, this won't be the first time.

00:56:56   And so it might also just be partly that,

00:57:00   that it wasn't a major upgrade from Intel,

00:57:02   and partly Apple doesn't care that strongly

00:57:05   about upgrading these on a frequent basis.

00:57:08   I don't know.

00:57:09   - Back in the good old days, when a new chip came out,

00:57:11   everybody voted in their machines, because you had to,

00:57:13   because there was competitive pressure to.

00:57:15   And for the most part, that was a pretty good thing.

00:57:18   even if these were not a big upgrade over the old ones,

00:57:22   especially on the thing like the Mac Pro,

00:57:23   like, oh, more cores, who cares?

00:57:25   But like, that's, isn't that what people are buying this for?

00:57:27   Like the super parallel task where two extra cores

00:57:30   would be a significant difference, right?

00:57:32   Like of all the machines that are actually gonna care

00:57:34   that you have, you know, 12 versus 10 cores,

00:57:36   the Mac Pro is the one.

00:57:38   The tips are saying that they didn't have

00:57:40   any good GPU options, so they would have had to jump the,

00:57:42   bump the CPU without the GPU.

00:57:44   But anyway, Apple historically has not liked

00:57:46   to play that game, and they're like,

00:57:47   unless we have an impressive new machine,

00:57:49   we're not just gonna give you new CPUs,

00:57:51   except I guess in the laptops where they'll do that

00:57:54   with the Pazwell 1.5s, I don't know.

00:57:56   - Oh yeah, I mean they, I don't know if they've ever

00:58:00   skipped a major laptop CPU generation until now

00:58:04   where they appear to be skipping quad-core Broadwells

00:58:07   and going right to Skylake, but this,

00:58:10   and this is a weird situation with the timing

00:58:12   of those things, I don't think they'd ever skip one

00:58:14   before that.

00:58:15   So what does this mean if hypothetically you have two copies of, I think it's a late 2011

00:58:21   Hi-Res Antiglare 15-inch MacBook Pro, and there is not a singular, there is not one

00:58:26   retina Mac in your entire household, and you're thinking you're going to hold on for the next

00:58:31   major MacBook Pro revision, or, well, I shouldn't say major, but you know, the next big chipset.

00:58:37   Last I heard, that was theoretically coming this fall.

00:58:40   Does any of this change that?

00:58:41   - No, 'cause what they're saying is that

00:58:44   the generation of chips that we're moving into,

00:58:47   the 14 nanometer, this is what started with Broadwell

00:58:51   this past spring with the MacBook One,

00:58:52   and then the Airs, and then later on the 13.

00:58:55   That generation is the 14 nanometer one,

00:58:57   and that's, we are just in the infancy of that.

00:59:00   - And we're gonna be at 14 for a while,

00:59:01   is what they're saying. - Yeah, exactly.

00:59:02   - Just the 14, which is fine,

00:59:04   'cause Skylake is apparently a big deal,

00:59:07   get some nice power savings.

00:59:09   I don't think it's the end of the world,

00:59:10   and it's not, you know, if you're waiting

00:59:13   for a new line of Macs to buy,

00:59:15   when the Skylike Macs come out,

00:59:17   they oughta be pretty nice machines.

00:59:18   So that's not a big deal.

00:59:20   It's just, the next year and the year after,

00:59:24   I don't know what the strategy is for,

00:59:27   I mean, if it's just gonna be three years

00:59:30   of 14 nanometer chips, yeah.

00:59:33   - Well, I mean, we've had, what,

00:59:34   two and a half years of Haswells,

00:59:36   and, you know, it hasn't been great.

00:59:39   Like we've been kind of stalled at performance

00:59:41   for a long time, but they're still good chips.

00:59:45   I mean like this iMac, this four gigahertz,

00:59:47   four core iMac that I have is amazing.

00:59:50   Like it is by far the best computer I've ever owned.

00:59:54   And by far, not by far, it's one of the fastest.

00:59:57   I did briefly have the six core cylinder Mac Pro

01:00:01   and that was faster in parallel tasks,

01:00:04   but this actually beats it in single threaded,

01:00:06   which is another problem the Xeon line has

01:00:08   that the consumer desktop chips, because they are newer

01:00:12   in their core revision, usually beat the Mac Pro Xeons

01:00:17   in single-threaded tasks,

01:00:18   which is embarrassing for the Mac Pro.

01:00:20   Anyway, I don't think this is gonna be that bad of a thing.

01:00:23   I think it's gonna be really just a continuation

01:00:26   of what we've seen for the last few years,

01:00:28   just now again, now they're just making it official.

01:00:30   Now they're admitting up front and warning us up front,

01:00:33   okay, this is what's going to happen again,

01:00:36   and we'll see what we can do about it.

01:00:37   On a final note, I should point out, lest my father, the ex-IBM-er, give me hell about

01:00:43   it, that IBM is actually at 7 nanometers, is that right?

01:00:46   Not really, they're just demoing, it's like a tech demo, like "look what we can do in

01:00:50   the lab, my cross is no object!"

01:00:53   Try shipping millions of those things every quarter.

01:00:55   I agree, I agree, but I just wanted to point it out, because obviously other people are

01:01:00   making progress in this arena, but just like you both said, that's not producing millions

01:01:05   of them, that's making one in a lab.

01:01:07   great expense with very exotic materials. What element are they using? Expensive stuff.

01:01:14   In summary, it's going to be expensive. Well, no, it's the same, like, I remember back in

01:01:21   the PowerPC days, they had similar type of things. Companies would say, "We can make

01:01:24   a chip at amazing speeds." It was probably like 1 GHz or 2 GHz back then, but we were

01:01:29   at like 300 MHz. Like, "Wow, how can you make something that fast? We use these exotic materials."

01:01:33   you could do lots of things for enough money, right?

01:01:37   But you can't, if it's gonna cost $10,000 per chip

01:01:39   or whatever, that doesn't make a difference.

01:01:41   It's not viable, you have to make it into something

01:01:44   that could be manufactured cheaply

01:01:46   and then you've got something.

01:01:47   That's Intel's whole thing, like they're gonna do a shrink

01:01:49   and suddenly the cost of all their chips

01:01:51   aren't gonna triple.

01:01:52   - The first computer magazine issue I ever bought

01:01:55   from a newsstand had a cover that was advertising

01:01:58   100 megahertz.

01:01:59   'Cause this was a big deal in 1995.

01:02:02   3 or something and it was it was a DEC alpha chip that had been announced

01:02:08   that could hit 100 megahertz and I was very disappointed to learn that. I vividly

01:02:12   remember having a really intense argument with a I think he's a second

01:02:18   cousin of mine I know a relative of mine who was in college at the time and I was

01:02:23   in like middle school or something like that and he had if memory serves he had

01:02:29   had, I want to say it was like a gigabyte of storage across probably like six hard drives

01:02:35   because it was that long ago. And I remember arguing with him, there is no possible way

01:02:40   a human being could use a gigabyte of storage. It's just not possible. What could you possibly

01:02:45   put over a gigabyte of storage?

01:02:47   - It was back then you were storing like, you know, pages or like Word documents. It

01:02:51   wasn't like anything big. It was like...

01:02:53   - It didn't have like these podcast apps downloading as audio and leaving their temp files around

01:02:57   in your disks.

01:02:58   - Well done, well done.

01:03:00   But yeah, it's wild how things change.

01:03:02   In any case, why don't you tell us about one last thing

01:03:04   that's awesome and then we have one more thing

01:03:05   to talk about.

01:03:06   - Our final sponsor this week is MailRoute.

01:03:09   Once again, MailRoute.

01:03:10   Go to mailroute.net/atp.

01:03:12   You've heard from me over the years that I don't use Gmail.

01:03:16   I use regular IMAP hosts

01:03:18   because I don't like proprietary email systems

01:03:20   and I don't like webmail.

01:03:22   And I just want something that will work

01:03:23   with regular mail apps that I can get them anywhere

01:03:27   on any platform and I don't have to be tied to,

01:03:30   you know, somebody's web interface I don't like

01:03:32   or somebody's terrible IMAP gateway Gmail.

01:03:35   So anyway, and one of the problems when you try

01:03:38   to host mail somewhere other than Gmail, somewhere else,

01:03:41   or you try to host it yourself, if you're on your own server

01:03:44   or if you have a business and you have to administer

01:03:46   their email server, one of the big problems

01:03:49   with doing this is spam filtering.

01:03:51   Spam filtering is so complex these days, it's so advanced,

01:03:54   that really doing it right requires a dedicated service

01:03:58   that specializes in that, like a cloud service

01:04:01   that can combine wisdom from all sorts of things

01:04:03   and not just be like one installation,

01:04:04   like a simple Bayesian filter is not enough these days.

01:04:08   So MailRoute is a professional spam

01:04:12   and virus filtering service.

01:04:13   So what you do is you point your domain MX record at them.

01:04:18   So they become your official email host to the world.

01:04:21   Then you give them the address of your real MX record,

01:04:24   or your real server, and so they act as a proxy,

01:04:26   as a middleman between the internet and you,

01:04:28   and they shield you from all the crap

01:04:30   people try to send you that's spam and viruses.

01:04:32   I admit, when I first heard about it, I was skeptical.

01:04:35   I thought, you know, how well could it really work

01:04:37   compared to the customizable spam filtering and stuff

01:04:40   that Fazmail had that I'd be using for years before that?

01:04:43   And it worked so well that I,

01:04:45   when I was using other things, I would still get spam.

01:04:50   I would get maybe three to five spam messages a day,

01:04:53   and I thought that was about as good as it can get, really.

01:04:56   And then I switched to mail route, I don't know,

01:04:57   maybe six months ago, something like that.

01:04:59   It is so good that I had received one spam message

01:05:03   one day last week, and that was noteworthy,

01:05:06   because I just hadn't seen one in weeks or months.

01:05:09   Like, mail route is so good, you literally go from

01:05:13   whatever amount of spam you're getting,

01:05:14   you go to zero most of the time.

01:05:17   Like, it becomes a very rare occurrence

01:05:19   to see spam actually get delivered onto your inbox.

01:05:21   It is that good.

01:05:22   it isn't so aggressive that they're constantly

01:05:25   throwing away legitimate mail either.

01:05:27   Anything that's kind of on the border

01:05:29   where they're not quite sure if it's spam,

01:05:31   they put it in the quarantine section

01:05:32   for like a week or two or something,

01:05:34   and then they email you a summary

01:05:36   of just like the subject lines of those things.

01:05:39   And you can just give a quick little glance down that list

01:05:42   every day, every week, or whatever,

01:05:43   and you can see, all right, you know what,

01:05:44   that one, that was legitimate, let that through.

01:05:46   All the rest of these, forget about it.

01:05:48   Those are all spam.

01:05:49   So it's very, very easy to make sure

01:05:50   you don't have false positives for too long,

01:05:53   and to train it not to do that anymore

01:05:55   for whatever slightly spammy things

01:05:57   that you get emails for.

01:05:59   But the spam filtering is just shockingly good.

01:06:02   It is so good.

01:06:03   I don't think I will ever go without it now.

01:06:05   It is, I can very, very highly recommend it.

01:06:08   Over the years, I've recommended people go to IMAP posts.

01:06:11   I've recommended Fast Mail a lot specifically.

01:06:13   I will say now, I very, very highly recommend

01:06:17   putting MailRoute in front of your IMAP post.

01:06:19   It is really good.

01:06:20   From what I hear from Gmail people,

01:06:22   it sounds like it's actually better

01:06:23   than Gmail spam filtering.

01:06:25   And I know a lot of Gmail people think it can't get better,

01:06:27   but if you're ever seeing spam, it can get better.

01:06:31   (laughs)

01:06:32   That's basically it.

01:06:33   So mail route really is amazing.

01:06:34   I very highly recommend mail route.

01:06:37   If you have your own domain name, or if you host email,

01:06:40   or if you have your own post,

01:06:41   put mail route in front of your mail, trust me.

01:06:43   It is that good.

01:06:44   They're not paying me to say that.

01:06:45   They're paying me to say this other script

01:06:46   that I'm mostly ignoring

01:06:46   'cause I'm telling you something better,

01:06:47   which is I use it and I love it

01:06:50   and I can tell you first hand it works for me.

01:06:52   Nobody can buy me saying that.

01:06:54   I'm telling you that because it's true.

01:06:55   Anyway, back to their script.

01:06:57   Go to mailroute.net/atp for a free trial

01:07:01   and if you use that link, mailroute.net/atp,

01:07:04   you will also get 10% off for the lifetime of your account.

01:07:08   Not 10% off for your first purchase, not 10% off once,

01:07:11   10% off for the lifetime of your account.

01:07:13   That is awesome.

01:07:14   Thanks a lot to MailRoute, mailroute.net/atp.

01:07:17   So lately there's been a humongous kerfuffle with Reddit.

01:07:22   I have to admit I don't really ever use Reddit.

01:07:26   I think I may have mentioned in a show on the show in the past, I don't know if it ever

01:07:30   made it to the final version, but the only time I ever really used Reddit was when I

01:07:35   needed to acquire things after one of my favorite websites got shut down.

01:07:42   And I used Reddit here and there for that.

01:07:44   But I don't use it for discussions. I don't use it for really anything else. And I guess

01:07:50   there was a real big—there was a lot of agita with regard to their now ex-CEO, Ellen

01:07:57   Pao. Is that right? Did I get that right?

01:07:59   I believe so.

01:08:01   And I can't pass judgment as to whether or not she was good at her job because I don't

01:08:06   really use Reddit. But it seemed like a lot of people came out of the woodwork to try

01:08:12   to get her out of the CEO position. And it seemed like, from an outsider's perspective,

01:08:18   the whole of Reddit tried to kind of hoist her out of the chair, if you will.

01:08:25   And the only real concrete information I've gotten about this is from Hello Internet,

01:08:30   and I just listened—I think it was today, actually—to their latest episode, and we'll

01:08:34   have a link in the show notes, where they talk about this. And CGP Grey seems to be

01:08:38   a pretty big Reddit user and a pretty big Reddit fan, whereas Brady Hare and the other

01:08:43   host seems to have a much more similar perspective to myself, or I should say I have a more similar

01:08:50   perspective to him, in that it all seems really gross to me from the outside.

01:08:55   And it all seems like a kind of crummy corner of the internet, but I'm probably judging

01:09:01   unfairly because I'm only hearing about the crummiest parts of this corner of the internet.

01:09:07   So I don't know, Jon, do you use Reddit at all?

01:09:11   Do you have any thoughts on this?

01:09:12   I do.

01:09:14   I think it's worth explaining what Reddit is, broadly speaking, because if people don't

01:09:21   use it, like, I don't think the name tells you anything about it, and you could get confused

01:09:26   about what the heck goes on there.

01:09:29   It is in the grand scheme of things.

01:09:30   There's a website where people go to have little communities.

01:09:36   You make a subreddit about whatever topic you're interested in talking about and there

01:09:39   people can share links with each other and they can comment on the links and they can

01:09:43   also vote on each other's comments.

01:09:45   That's basically what it boils down to.

01:09:46   You've probably heard of lots of sites like this where you can make a little sub-community,

01:09:50   you have an account, you can share information with each other in links and comments and

01:09:56   then the things that you share other people can comment on and then everyone gets to vote

01:09:59   on the things that you say and it seems like all right well whatever but you know we've

01:10:02   had web bulletin boards we had you know dig and hacker news and like a million other sites

01:10:09   that are like this reddit is one of those sites it was one of the ones that it kind

01:10:14   of caught on it's very large very popular there are subreddits for everything you could

01:10:18   possibly imagine from technologies to cats to alligators to knitting to everything you

01:10:23   imagine. It seems like, you know, everything's like, whatever. It's, it's like, Content-Ask owns them

01:10:30   now. I'm not sure if they ever made money. They have difficulty trying to come up with a

01:10:33   monetization pattern, but they get a tremendous amount of traffic and it's a place where people go.

01:10:37   Everything else about it, like, is sort of problems of their own making, because if the,

01:10:44   setting aside, how do you make money? Like, we talk about that to death. Like, hey, you got a

01:10:47   lot of traffic, you got a lot of users, you got a lot of eyeballs or whatever, figured how to

01:10:50   monetize them, stuff like that. The decision Reddit is faced with is a decision that many

01:10:56   sites that are very popular are faced with, and almost every site I can think of off the

01:11:00   top of my head makes better decisions about this, which is, since anyone can come and

01:11:03   make an account, and anyone can come and make a little subgroup or whatever about whatever

01:11:06   topic they want, do you have some kind of policy, terms and conditions, code of conduct,

01:11:16   Something saying what are you allowed to do and say and what kind of groups are you allowed to have on this site?

01:11:22   And I like me I can't think of another

01:11:25   Website that doesn't have something like that certainly Twitter does and Twitter

01:11:29   You know ways complain about the harassment on Twitter and stuff like that at least they have a policy right

01:11:33   Reddit's big thing was like no man free speech anybody can come here and make a subreddit about anything and

01:11:40   Like we should be allowed to say what it like

01:11:43   Anyone tell me when someone brings up free speech on the internet. I just I don't think I've seen an example of it being correct

01:11:50   Maybe because I don't hang out on constitutional law boards

01:11:52   Maybe that's the place to work. What people really mean is I'm allowed to say whatever I want on someone else's website

01:11:58   Because free speech right so read it the people who run read it fancy themselves as like we're not just a website

01:12:04   we are the internet and we are completely neutral and

01:12:08   Even though we're a privately held company that pays money to someone to run our servers and pays employees to write the software for them

01:12:14   And does all this stuff really we are just a giant level playing field for everybody to come and and do it

01:12:22   And therefore we don't want to have pretty much any controls over what you're allowed to do what you're allowed to say

01:12:29   What kind of groups you're allowed to have all?

01:12:32   everything is welcome, all people, all ideas, all behavior, as long as I guess you don't hack us and

01:12:38   destroy our servers or something like that because that would be bad.

01:12:40   But anything beyond that, like if you're just using our site, if you are typing words into a box and clicking buttons,

01:12:44   anything you do is a-okay with us. And a lot of the controversy with the new CEO came in because she

01:12:51   decided that some things are not okay, that we're gonna have some sort of barriers and rules about, "Oh, okay,

01:12:57   well, you can't just type any words you want into this box.

01:12:59   You can't have communities surrounding anything you want because we don't want we don't want you those type of people using our website

01:13:07   We don't want them to be part of our quote-unquote community and you were saying case like oh

01:13:11   It seems like an icky place or whatever

01:13:12   There is no reddit community ready this gigantic reddit has people who are just obsessed with cute cat pictures, right?

01:13:17   And then reddit has people who were just you know

01:13:20   Showing was it pictures of fat people to make fun of them or whatever or like racist bulletin boards or you know

01:13:29   like every possible while ID you can imagine the whole idea is like oh this is this is a place where we can go and

01:13:34   we're allowed to use these people's website and put our information in their database servers and

01:13:38   Send requests to their web servers and it's all free speech man. We can do whatever you want and

01:13:43   Reddit trying to add some kind of controls like okay well certain things

01:13:47   We don't want to be on something that reddit.com like the certain things

01:13:51   We don't want so if you want to do that go someplace else and do it

01:13:54   The reddit community such as it is like the people who believe in free speech

01:13:58   Like you can't stop us from putting words into your servers. That's our

01:14:03   That's our birthright reddit is about free speech

01:14:07   And it's just so depressing to me to see and by the way, they're backsliding on that now

01:14:13   They're like, oh that that CEO is gone and the new person in here

01:14:16   I don't know all the backstory this but the new person is actually one of the early people in reddit is like actually we want

01:14:21   We want you to allow it to be do anything you want

01:14:24   except for these very few things that we narrowly define as being bad.

01:14:29   I don't understand what the sort of the mental barrier here is to the people who are running

01:14:35   the site, putting their foot down and saying, "This is okay, and this is not." They somehow

01:14:40   feel like by doing that they're crushing freedom under their jackboot. Like you just run reddit.com,

01:14:45   it's a website on the internet where people type words into boxes and put them into database

01:14:49   servers that you run. That doesn't define free speech. So I find it incredibly frustrating to see

01:14:55   all this posturing and all this anger all so misdirected. And the only thing I put in the

01:15:02   show that's about this is two tweets from Laurie Voss, who I think sum up this entire thing.

01:15:07   She says, "The tragedy of all online community spaces is that the goals of inclusive and safe

01:15:13   are at the extreme mutually exclusive goals." So what she's saying is like, "Do you want to

01:15:18   Want to have a place where everybody is welcome?

01:15:20   Oh, we are an inclusive site.

01:15:21   Anyone can come to Reddit, anyone can make a subreddit.

01:15:23   Like it's good to be inclusive.

01:15:24   Everyone likes inclusive, right?

01:15:25   We don't tell you you can't come here

01:15:27   because your eyes are the wrong color

01:15:29   or because you speak the wrong language

01:15:31   or you're into the wrong, you like dogs instead of cats.

01:15:34   We're an inclusive site, right?

01:15:36   But also safe, what does safe mean, right?

01:15:39   Our next tweet, at some point you have to exclude someone.

01:15:41   You get to pick if it's the people feeling unsafe

01:15:44   or the people making them feel unsafe.

01:15:46   So at some point you have to exclude somebody

01:15:47   if you let everybody come in, people come in and do terrible things. So you have to

01:15:51   at a certain point exclude somebody. Who do you want to exclude? Do you want to exclude

01:15:54   the people who don't feel safe because there's a bulletin board about how to rape women and

01:15:59   that kind of makes the women on Reddit not feel very good, that they are somehow associated

01:16:03   with the same website where this board exists? Or do you want the people who are making the

01:16:08   "let me tell you how to rape women" subreddit, do you want to make them leave? You have to

01:16:12   exclude somebody because if you let everybody in, some people are not going to feel safe

01:16:16   or comfortable or welcome in this community. These two tweets basically sum it up and incredibly,

01:16:22   every time I read anything from these Reddit people, they make the decision of like,

01:16:25   what we really want is for the people to be able to make the rape boards. Like, you know,

01:16:30   not the specific examples, I don't know if they've banned them or whatever, but like,

01:16:32   some people are made to feel uncomfortable by what pretty much anyone would say is vile, terrible

01:16:40   ideas and behavior. And it's like, well, they need to be allowed to do that because we are

01:16:45   the government of the United States, and we must allow them the freedom to put their words into

01:16:49   our database servers. Therefore, free speech. Sorry, I just got to let them do that." It boggles

01:16:56   my mind. It makes me angry and frustrated, mostly because there is a lot of good stuff on Reddit,

01:17:00   because again, the people who are making the cute cat pictures, those cat pictures are really cute.

01:17:04   They're not bothering anybody. And the thing is, I think I saw a graph somewhere trying to show the

01:17:11   overlap between the various boards. There is some surprising overlap, but it's sometimes not so surprising overlap.

01:17:17   Like, you know, how many people are member of the white supremacy board, but also member of the cute cat pictures board?

01:17:23   Okay, how about members of the white supremacy board and members of the men's rights board?

01:17:26   All right, and you start to see a little bit more. Members of Gamergate and members of the young Nazis of America board. Like, you know,

01:17:33   it's not 100% overlap,

01:17:35   But I don't I would love to see a graph of like are the cat people really just cat people or

01:17:40   So many people love cats that an equal number of neo-nazis are in the cat boards as are in every other board. I don't know

01:17:47   Anyway, I think that that it is a ridiculous thing that the people running reddit can't seem to get a handle on the idea that

01:17:55   They get to choose what kind of people get to put words into their database servers and they seem

01:18:02   incapable of making any sort of common sense decision about what should be allowed to happen on reddit or

01:18:08   Maybe they're just totally for like we there maybe they're anarchists

01:18:11   Maybe they say we want to run a bunch of servers and we want to let everyone say anything

01:18:14   Go for it and like there you go

01:18:16   that's that's what you're gonna do but that's why people keep talking like the downfall of reddit and everything because I think

01:18:20   Most people who use reddit had an idea that the people who ran reddit had

01:18:26   Similar standards them about what is and isn't sort of decent and acceptable and the people who run reddit want to let everybody do everything

01:18:33   The rest of the people have to think are just gonna slowly fade away because no one wants to be associated with a site like that

01:18:38   Like a very small number of you're just gonna turn into you know, 4chan or 8chan, right?

01:18:42   Like if if you let everybody in eventually everybody leaves except for the most terrible people and congratulations, you know reddit

01:18:49   You are running a site filled with the most terrible people

01:18:51   Feel free to do that. Feel free to let them all type in text boxes to each other

01:18:55   I hope you feel good about the new site you've built

01:18:57   because that's what you're effectively creating

01:19:00   by allowing anybody to do everything.

01:19:03   - I don't have much to say on Reddit

01:19:04   because I don't know anything about Reddit.

01:19:07   And every time I have visited it for something,

01:19:10   either trying to find information on something

01:19:11   or following a link there,

01:19:13   every visit to it has made me never want to visit it again.

01:19:16   (laughing)

01:19:17   Literally, I'm not exaggerating.

01:19:19   Every time I'm like, oh, this is terrible, why do people,

01:19:21   and I know, academically,

01:19:24   I know there are good parts of it.

01:19:26   I've never stumbled upon those really.

01:19:28   - I never read a good Ask Me Anything or I Am A Thing.

01:19:32   - Those are usually really hard to navigate.

01:19:35   First of all, I hate nested comments.

01:19:36   Like, and I was an old forum nerd back in the day,

01:19:40   so I know the whole, I know the problems

01:19:43   with both nested comments and also flat threads.

01:19:47   They both have dysfunction, just different dysfunction.

01:19:51   so you know I I'm very familiar with the challenges in both the community

01:19:57   monitoring and regulation and also the technical side of how community software

01:20:03   online had like how how this community style thing is built and and how things

01:20:09   are enforced it's it's a very hard problem not to mention my time at

01:20:12   tumblr we saw a lot of things not quite the same level as what reddit is dealing

01:20:18   with, but a lot of kind of related problems of things like, you know, like

01:20:23   every every couple days we get a phone call from like a high school principal

01:20:28   or like some kind of like, you know, middle of nowhere police department or

01:20:33   something complaining about something that some kids had about some other kid

01:20:36   on tumblr and wanting us to take it down and like when you you have to deal with

01:20:40   stuff like that when you're a community site it's dealing with community stuff

01:20:43   is just really messy.

01:20:46   It's very difficult and it's very, very messy.

01:20:49   For all the reasons you just said, Jon,

01:20:50   like there's a lot of these decisions

01:20:53   that there is no good option.

01:20:55   You just have to choose between two bad options

01:20:57   and a lot of things, like what I was saying

01:20:59   with my cell download stuff earlier,

01:21:01   like a lot of these things, you just have to pick

01:21:04   the lesser of two evils and no matter what you pick,

01:21:06   it's gonna have big downsides for someone

01:21:08   or it's gonna disincentivize good stuff

01:21:10   or incentivize bad stuff.

01:21:12   - But we're not even at the hard questions.

01:21:15   There are the type of things like,

01:21:16   am I allowed to threaten to rape somebody on Reddit?

01:21:18   And the Reddit CEO is like, well, I'll allow it.

01:21:20   Technically, it's not a threat of violence.

01:21:22   - Oh, God.

01:21:23   - I mean, we're not talking about the people

01:21:25   getting into a heated argument

01:21:26   and should they be allowed to call each other jerks

01:21:28   and sort of ad hominem attacks are not allowed.

01:21:32   That's in the gray area.

01:21:33   It's difficult to figure out or whatever.

01:21:34   We're talking about just straight up racism,

01:21:39   sexism, real threats of violence,

01:21:42   and seriously, the new CEO is on the board going,

01:21:46   responding to people, "I'm not sure if that,"

01:21:48   you're not sure?

01:21:50   I guess that's the kind of community you want.

01:21:52   Again, I'm not saying he shouldn't be allowed to do this.

01:21:54   Feel free to make this kind of community,

01:21:55   but if he's waffling, like, "These are the tough decisions.

01:21:57   "Boy, should we allow rape threats,

01:21:59   "or is that against the rule?"

01:22:00   I guess we have to allow it.

01:22:01   I don't want that guy's speech to be impinged

01:22:03   because I am the government of the United States of America.

01:22:06   - Well, and I think, first of all,

01:22:08   So we should definitely say that the last episode

01:22:11   of Hello Internet and the last two episodes of Rocket

01:22:14   covered this probably way better than we can

01:22:16   'cause those are people who know a lot more

01:22:17   about Reddit than we do.

01:22:18   And so they really did a very good job of that.

01:22:21   So listen to those if you want more of the inside stuff

01:22:23   about what specific challenges that the CEO of Reddit faces

01:22:27   and why Ellen Pao might have been good or bad for that.

01:22:33   Anyway, but it seems like one of the big pressures

01:22:37   they have now is business.

01:22:39   They make some money, and they have this corporate owner,

01:22:44   and they have a board of directors,

01:22:46   and they want more money because the money they make

01:22:49   is not really commensurate with the amount of traffic

01:22:51   they get, they should be making a lot more.

01:22:53   - But is this what you would do to make more money?

01:22:56   Like wouldn't you try to go mainstream?

01:22:58   Like don't you wanna--

01:22:59   - Well, so that's the thing.

01:23:00   So let me tell you another story from my past

01:23:02   that's long and boring.

01:23:03   The reason I'm telling it is because it's actually

01:23:05   about the same coffee shop as the last Long and Boring story.

01:23:08   - Oh God, here we go.

01:23:09   - So this coffee shop in my hometown,

01:23:10   Cuppa Joe in Bexley, Ohio,

01:23:13   while I was there, they had a big problem

01:23:16   where middle schoolers discovered it.

01:23:19   And so it was always packed full of like 12 year olds

01:23:23   who were just kind of loitering around,

01:23:27   hardly buying anything but just taking up all the seats

01:23:30   and hanging around outside and kind of standing around

01:23:32   smoking and all it's like it was just not a great scene and and they it kind

01:23:37   of drove other customers away because who wants to hang out around a bunch of

01:23:42   delinquent middle schoolers who are bored and loitering and not doing

01:23:45   anything and smoking and so one summer they banned them I don't know how

01:23:50   doesn't really matter how they banned all the middle schoolers they implemented

01:23:54   some kind of like minimum age thing or you had to be with a parent or whatever

01:23:56   it doesn't matter somehow they banned them and for a while the place was empty

01:24:01   and I knew the manager at the time, and I asked him,

01:24:03   like, you know, how was this going?

01:24:04   And he said, they were making more money than ever.

01:24:07   (laughs)

01:24:08   Because, you know, even though it wasn't getting

01:24:12   the volume it got before, it was getting more

01:24:15   of the profitable customers, the older people,

01:24:18   like the college students who were there studying

01:24:19   and buying drinks all day, the people who would come

01:24:22   in the morning and come in the evenings,

01:24:24   who, you know, were just buying stuff,

01:24:26   drinking it and leaving, and making, you know,

01:24:28   cycling the tables more often, not putting off other people

01:24:32   like when they when they come in there, and I don't know how

01:24:35   the result that I left, but

01:24:37   they I was very surprised at the time and looking back on it. I

01:24:42   shouldn't be that that worked so well for them and you know

01:24:45   read it. I think faces not that different of a dilemma here

01:24:50   where it's like if they want to become more mainstream, if they

01:24:52   want to become more attractive to advertisers and therefore

01:24:55   more profitable, if they want their the value of their ads to

01:24:57   go up, if they want the number of advertisers to go up, I think they have to get rid of

01:25:02   all this garbage. They have to get rid of all this, like, you know, anarchist/libertarian,

01:25:07   free speech, crazy people who are really just aggressive, hateful people for a lot of them.

01:25:13   Like, you know, not all of them, obviously, but many of them are just these very aggressive,

01:25:16   hateful people who feel entitled to have their words all over the place. I think you can't

01:25:21   have both ways. You can't have that garbage on your site that is highly offensive, legally

01:25:26   questionable, no advertiser wants to be associated with that, and also make more money with advertising.

01:25:33   You have to pick one or the other.

01:25:34   And it seems like they're trying to hide the most objectionable stuff. Like, "Okay,

01:25:39   a lot of people are offended by this, but we won't run ads against it, so we're

01:25:42   technically not making money off of it." And everyone's like, "Well, you're just

01:25:45   subsidizing it with the other stuff, and we'll put it off in the corner and they'll

01:25:47   stay by themselves." And it's like, why are you bending over backwards to make sure

01:25:52   that these people have a place to share hate with each other? Never mind that it's not

01:25:56   not just, "Oh, we're just letting them put their words into our database."

01:25:59   If you want to go on to the sort of secondary effects of allowing a place for people to

01:26:05   reinforce each other's hateful ideas and recruit new people and like, "Oh, no incitement to

01:26:10   violence that's against the Reddit guidelines."

01:26:12   Yeah, I'm sure that will work exactly as intended when you get a bunch of hateful people talking

01:26:15   to each other constantly over and over again and sharing hateful pictures and videos and

01:26:19   all sorts of terrible things.

01:26:21   I'm sure nothing bad will come out of that.

01:26:22   It will just be positive.

01:26:23   Like as long as they stay within the terms and conditions, no.

01:26:25   It's so ridiculous.

01:26:26   It's so asinine.

01:26:28   And again, if that's the site they want to run, go for it.

01:26:30   But if their job is to make money for their parent company,

01:26:34   this is not the way to do it.

01:26:35   And using your middle school coffee shop example,

01:26:37   it's as if at a certain point,

01:26:39   all the non-middle schoolers left,

01:26:41   and all that was left was middle school students,

01:26:44   and you realize, well, now we have to cater to them

01:26:46   because they're our only customers.

01:26:47   Like they're, not that Reddit is at this point,

01:26:49   but it's like, if you suddenly turn into,

01:26:51   well, I guess we have to become a dance club or something.

01:26:53   Because everyone else is dumb, right?

01:26:57   Reddit is not at that point, not even close to it.

01:26:59   But if you just keep going down this path, I feel like the regular people will leave

01:27:04   and you'll just be left with, "We are a community of hate speech bulletin boards because these

01:27:09   people have no place else to go."

01:27:10   It's like 8chan or 4chan to a lesser extent.

01:27:14   That's not the kind of business you want to run.

01:27:16   That's not a growth market.

01:27:17   That's not a lot of people with a lot of spending power.

01:27:20   If you want money, you want to go mainstream.

01:27:21   And it just seems like the Reddit CEO is trying to preserve this ideal, like we provide a

01:27:29   forum for people to openly share ideas.

01:27:32   That's sort of a high-minded ideal of Reddit, but the consequences of it are not going to

01:27:37   be good for Reddit's bottom line, or for the popularity of Reddit, or for Reddit's reputation,

01:27:44   or for his reputation, or just anything really.

01:27:46   It just seems like all downside.

01:27:47   And again, go ahead, you know,

01:27:49   if that's what the company wants to do,

01:27:52   feel free to pursue that path,

01:27:53   but I don't see it turning into anything good

01:27:56   for anyone involved.

01:27:57   - So this probably doesn't mean anything

01:28:00   for the purpose of our discussion,

01:28:01   and I probably shouldn't take as a sign

01:28:03   of me being right or wrong about this,

01:28:05   but that coffee shop closed last fall.

01:28:07   - Coffee shops close all the time.

01:28:09   - I know, it sounded like the building they were in

01:28:12   sold to get knocked down and built something bigger.

01:28:15   - Yeah, but I don't know.

01:28:17   That's part of the whole ideal of the sort of the,

01:28:21   someone in the chat room said

01:28:22   there really is a Reddit community.

01:28:23   I think there's a stereotype of a typical Reddit user

01:28:27   that's perhaps somewhat accurate,

01:28:29   but I think there are a lot of people,

01:28:31   like all three of us, who don't go to Reddit

01:28:33   unless someone links to it.

01:28:34   And when we link to it, we look at the funny cat GIF

01:28:36   and then we move on with our lives, right?

01:28:39   But are we Reddit users?

01:28:40   Are we part of the community?

01:28:41   We're not part of the community, but we follow,

01:28:44   It's a popular site, people link you to it, you end up there, you look at something and

01:28:48   then you go someplace else.

01:28:49   I think a lot of that traffic is from things like that.

01:28:52   I don't know, maybe they know the breakdowns.

01:28:55   I don't know if most of their traffic comes from people just posting hate to each other

01:29:01   and over and over again or getting in big arguments and 17 level indented comment threads

01:29:05   attached to things.

01:29:06   Maybe that is where their stuff comes from.

01:29:08   It just seems to me it's a bad decision.

01:29:11   I think less of the people who run Reddit every time I hear one of their decisions or

01:29:14   see them in this sort of like laurely nuanced discussions as they try to parse what should

01:29:20   and shouldn't be acceptable behavior on Reddit and just reveal themselves to have no understanding

01:29:24   of other people's experiences.

01:29:28   I'm sure that all the specific examples I cited in this podcast are not accurate.

01:29:32   Feel free to wander Reddit yourself and try to find out what these people's opinions are.

01:29:36   But every time I see anything from them, every time I go to Reddit and see these discussions

01:29:39   with the new leadership behind Reddit, I just think.

01:29:44   I was excited when Ellen Pao came in and started making changes, like really straightforward,

01:29:50   sensible changes.

01:29:51   Get rid of the hate groups.

01:29:52   And it was like, "Oh, you can't get rid of the hate groups.

01:29:54   What do you mean?"

01:29:56   Again, I don't know the details.

01:29:59   I don't know why she was pushed out.

01:30:01   I'm sure she was subject to the same harassment that every woman gets who opens her mouth

01:30:05   and says anything ever, let alone is the CEO of a company, is in charge of a bunch of little

01:30:10   man-children who are just angry about everything.

01:30:12   Especially that company.

01:30:15   I can't imagine a more hostile, well maybe there are some, but that was definitely on

01:30:20   the more hostile side.

01:30:21   Yeah, and I don't know if she, you know, I have no idea what hers is, but all I know

01:30:26   is that the few things I heard announced come out of her leadership seemed like good ideas,

01:30:30   and then there was backlash against them.

01:30:32   I felt depressed about it, and then when she was kicked out, I got even more depressed,

01:30:35   and then seeing the views of the new CEO who is apparently one of the founders, it's just

01:30:38   like, nope, they're not getting it.

01:30:40   So anyway, Reddit is really not for me.

01:30:44   And furthermore, I think if Reddit's goals are to be a place that continues to have a

01:30:50   lot of traffic and then makes a lot of money, that the current strategy they're pursuing

01:30:54   doesn't make any sense to me.

01:30:55   I think the other thing that I find fascinating about all this, which I kind of hinted at

01:30:59   earlier is how a group of users of a website can seemingly have so much influence that

01:31:06   they can cause the chief executive officer of a company to be fired, dismissed, quit,

01:31:16   whatever. That's just weird to me that even a couple hundred thousand people, I think

01:31:23   I'd heard that there were a couple hundred thousand signatures on this petition, this

01:31:27   online petition, which isn't even really signatures but whatever, that that many people came together

01:31:33   to decide to try to oust a CEO, man, woman, it doesn't matter.

01:31:40   That's just insane to me that the users of a website, none of whom I think are paying

01:31:47   the website any money, seem to think that they can have that kind of influence and in

01:31:50   numbers seem to have that kind of influence.

01:31:54   And I think a lot of this started,

01:31:55   if I understand things correctly,

01:31:57   because a really beloved moderator was dismissed

01:32:01   or something like that.

01:32:02   I guess the person who did the--

01:32:03   - Well, this is kind of like,

01:32:06   this is really about ethics in gaming journalism.

01:32:07   It's like that was not really the reason.

01:32:10   That was what ignited a lot of the argument,

01:32:13   but it was really a lot of other pressures, I think,

01:32:16   is what it sounds like.

01:32:17   But, you know, and especially 'cause it was,

01:32:20   even that is just like the game journalism BS line.

01:32:23   Even that official storyline is questionable

01:32:25   because she didn't fire that moderator.

01:32:28   The male co-founder did.

01:32:30   - I didn't realize that.

01:32:31   - Yeah, this whole thing's a mess.

01:32:32   Anyway, getting into details of it is just a rathole.

01:32:36   There's no good to come of diving into the details of it

01:32:40   because that's not really what the argument was about.

01:32:42   And what it comes down to is if a whole bunch of users

01:32:46   were calling for someone's head

01:32:48   and that resulted in her getting what I call quitfired,

01:32:52   says, you know, she was, you know, she was forced to quit.

01:32:54   (laughing)

01:32:55   That was not, you know, oh, I think I'm just done now,

01:32:58   coincidentally this week.

01:33:00   No, she was quit fired.

01:33:01   It's cool, happens to the best of us.

01:33:03   But that wouldn't have happened if just some angry users

01:33:07   who are, you know, hateful, horrible people,

01:33:10   if just they had a problem with her.

01:33:12   The reason that happened, I think you can read into

01:33:14   who they appointed instead.

01:33:16   This other co-founder who's been gone for a while

01:33:19   who has these very much like, you know,

01:33:21   libertarian, kind of anything goes viewpoints.

01:33:24   If the board appointed him as the new CEO,

01:33:28   that means that the members of the board

01:33:31   and the people, you know, who have control

01:33:33   and equity and influence of the company,

01:33:35   a good portion of them want this anything goes attitude

01:33:39   to stay.

01:33:40   It's not enough that a whole bunch of users complain.

01:33:43   If only these users were getting angry

01:33:45   and these users are people who are not

01:33:47   particularly credible, who you don't really want

01:33:48   to cater to if you can help it.

01:33:50   That wouldn't have been enough to get her out.

01:33:52   This only gave the people in power

01:33:55   a motivating reason to get her out now,

01:33:58   because she was pushing in a direction

01:34:00   that not that the users didn't want,

01:34:01   that they didn't want, that the power holders did not want.

01:34:06   And they put in a guy who is the complete opposite

01:34:08   in this area, in this area, you know,

01:34:11   like trying to keep things tame and under control

01:34:13   or being where everything goes.

01:34:14   So obviously, the people who own and control the company

01:34:18   want it to be the other way.

01:34:20   - Yeah, I think it's, I mean, the CEO situation, again,

01:34:22   I don't think the details are important.

01:34:24   It's just basically like a vote of no confidence.

01:34:26   It was a vote of no confidence from the most vocal users,

01:34:28   with the whole petition thing,

01:34:29   and apparently a vote of no confidence from the board.

01:34:31   It just seems like, you know, you came in,

01:34:34   you started to make some bold changes,

01:34:36   but in the end, we are not behind you.

01:34:38   This is the board saying this.

01:34:39   In the end, we do not support your decisions

01:34:41   enough to let you see this through,

01:34:43   and we're changing course and you're out, right?

01:34:46   So that is what it comes down to,

01:34:47   It was like, it's lots of company.

01:34:49   If you're gonna be the leader of a company,

01:34:50   you either actually have to have absolute power,

01:34:52   in which case you can weather any storm

01:34:54   and do what you wanna do,

01:34:55   or if there are people who are bosses of you,

01:34:58   and from their perspective,

01:35:00   they don't have confidence

01:35:01   that the changes you're gonna make are good,

01:35:03   and they're hearing a lot of noise from the users,

01:35:05   and the users are, the most vocal users

01:35:07   are saying with their actions and their words

01:35:11   that they don't like what you're doing either.

01:35:13   You can imagine a board that's nervous about a company,

01:35:15   like, oh, we have this thing, we think it can be good,

01:35:18   this person comes in and makes a bunch of changes,

01:35:20   and then all I hear is a bunch of noise and bad things,

01:35:22   change course, change course, right?

01:35:24   And then the new course, like some people in the chat room

01:35:26   are saying, this new CEO is going to, you know,

01:35:29   get rid of the bad even more than she could.

01:35:31   That's possible because it's kind of like, you know,

01:35:33   only Nixon can go to China, like in Reddit,

01:35:35   only a man can make changes.

01:35:37   Because people will only accept,

01:35:39   like the same exact words come out of this guy's mouth

01:35:41   and everyone's like, oh, that's interesting,

01:35:42   I'd like to hear more of you guys,

01:35:43   And she says it is like you are destroying our freedom like so it's I

01:35:47   Just go back to this thing like what kind of community does

01:35:51   These people want reddit to be who do they want to be in this community?

01:35:56   And and what do they want it to be like what do they want the experience of ready to be like same thing for the Twitter?

01:36:01   People what do you want the experience of Twitter to be like I mean?

01:36:03   The Twitter CEO at least came out and said we are terrible about dealing with harassment

01:36:07   Do you want Twitter to be a place where you go and if you are?

01:36:11   Any type of group that's typically harassed you're going to get harassed and you have no recourse

01:36:14   That Twitter is trying to make that better by having the tools to report people not require

01:36:19   You know

01:36:20   I think I've originally way back in the day on Twitter if you reported someone for harassment

01:36:23   You had to inform the person that you were informing on them like it was part of the process

01:36:27   Like they would send them an email and say just so you know

01:36:30   This person reported you for harassment or something like that was just a crazy setup made by people who didn't understand

01:36:35   the dynamics of harassment anyway

01:36:38   Like I said, Reddit, they can do what they want, but when I hear, everything I hear coming

01:36:43   out of this company so far is making me, like now I'm at the point where I'm almost like

01:36:48   questioning whether I want to tap links to what I know is probably good content that

01:36:52   I'll enjoy, like the interviews or funny cat pictures and stuff like that.

01:36:55   And now I feel like by tapping this link, am I supporting an organization that is trying

01:37:01   to make a home for people who I think should not have a home.

01:37:08   I'll put it this way.

01:37:09   I wouldn't give these people a home on a website that I ran.

01:37:12   These people give them a home on the website that they run, which is fine, but do I want

01:37:15   to support that with my clicks and whatever ads I'm going to load by tapping that?

01:37:18   That's the point where I am with Reddit, where I'm thinking that.

01:37:20   And I've never thought that about Twitter.

01:37:22   Maybe it's because I don't feel like I'm viewing ads in Twitter because I'm using a third-party

01:37:26   client or whatever, but at this point with the Reddit links, I don't even know, even

01:37:31   into my casual usage if I'm going to continue to do it.

01:37:34   You can just install an ad blocker. Thanks a lot to our three sponsors this week, Lynda.com,

01:37:39   Hover, and MailRoute. We will see you next week.

01:37:41   week.

01:37:48   Cause it was accidental, oh it was accidental John didn't do any research, Marco and Casey

01:37:58   wouldn't let him Cause it was accidental, oh it was accidental

01:38:05   And you can find the show notes at ATP.fm And if you're into Twitter, you can follow

01:38:14   Follow them @CASEYLISS

01:38:19   So that's Kasey Liss, M-A-R-C-O-A-R-M

01:38:24   Auntie Marco Arment, S-I-R-A-C

01:38:29   USA, Syracuse

01:38:31   It's accidental

01:38:35   They didn't mean to

01:38:37   Accidental

01:38:39   ♪ Check my cast so long ♪

01:38:43   - Boy, I hope no one who is a big in Reddit

01:38:46   listens to our show 'cause it's just going to be

01:38:48   a long series of corrections about everything

01:38:50   we got wrong about Reddit.

01:38:51   That's why I keep trying to pull back to the big picture.

01:38:52   Big picture, we are casual users.

01:38:54   It seems like a place I don't wanna hang out.

01:38:56   - Yeah, as I said, like every time,

01:38:58   literally every time I've gone there,

01:38:59   I've been turned off to it.

01:39:01   - I don't see, I don't have that problem.

01:39:02   I've gone to good content, but just like,

01:39:04   boy, everything I read about it

01:39:05   and every time I dive into these threads,

01:39:07   I'm like that screenshot can't be real.

01:39:08   This is obviously Photoshopped.

01:39:09   Go to the actual site and find the actual comment. I'm like this guy. This is the guy who's in charge now

01:39:14   There's not like someone impersonating him

01:39:16   The site hasn't been hacked like these are the real words out of these people's mouths like not you know

01:39:22   It's basically saying not a place. I want to

01:39:24   Hang out or be associated with

01:39:27   Yeah, it's probably gonna sound weird, but I get really turned off by reddit because I know a few people

01:39:36   I know certainly plenty people online that use Reddit. I know a handful of people in real life that use Reddit and it's kind of

01:39:43   creepy to me

01:39:46   how people who take Reddit really seriously seem to think that, and one of you said this earlier, the

01:39:52   entirety of the internet happens because of Reddit. Like anything that's good on the internet

01:39:58   it's because of Reddit according to these people and that's just, I don't know, it's almost like a cult and I don't I don't mean that

01:40:06   I don't know. I'm sure someone's gonna be offended by that. I don't mean it to be offensive.

01:40:09   It's just it's the that's the closest analogy I can think of is that oh man, you don't understand.

01:40:14   Reddit is where the internet happens, man. It's cuz of us.

01:40:18   That's why you know all this cool stuff and you know every gif that's ever been on the internet. That's cuz of us, man.

01:40:23   There's no place else to get gifs.

01:40:25   And it's just I don't know.

01:40:27   It's just it's creepy to me and and that I find that to be a real turn-off and and admittedly

01:40:32   I'm very ignorant when it comes to Reddit. Like you guys have said, you know,

01:40:34   I'll browse over to something, typically like an IMA or an AMA, and then I'll leave.

01:40:42   And that's that.

01:40:43   But God, it just seems so weird and creepy and cultish to me that I've never wanted

01:40:49   to invest any real time in it.

01:40:51   And now, what with their tacit, if not explicit support of all these hate groups, I just,

01:40:58   "No, thank you.

01:40:59   Not for me."

01:41:00   There is something to that, like the whole of, you know, you get enough people together

01:41:03   who are enthusiastic about sharing things with each other and they make these little

01:41:06   communities like that is what makes the internet great.

01:41:10   But you have to realize that when you reach a certain level of popularity, percentage-wise,

01:41:16   you will inevitably get a certain subset of people who exhibit what most of us would consider

01:41:21   to be bad behavior.

01:41:23   They're abusive to people, they do obnoxious things like griefing and trolling.

01:41:28   They make your site worse.

01:41:30   And so most sites have at least an attempt to come up, especially if this is the whole

01:41:35   thing, this is just a big community, come up with a set of rules to try to contain that.

01:41:39   And the perverse thing about Reddit is like, no, our philosophy is we don't do that.

01:41:43   That's the whole thing, man.

01:41:44   Just anything goes, you know, again, anything goes except you can't hack us.

01:41:49   We don't like that.

01:41:50   That is anti-social.

01:41:51   But anything else, anything goes.

01:41:54   Type anywhere.

01:41:55   It's like that's how every community dies.

01:41:58   Back in the Usenet days, you'd see entire newsgroups go south because a set of bad people

01:42:02   would come in, they'd make everyone miserable because you can't stop them from posting in

01:42:06   the newsgroup, you'd add them to your kill files, but it would just likely be, you know,

01:42:10   there's eventually people, "Okay, fine, this group is yours.

01:42:13   We'll leave and we'll go start a different group.

01:42:16   This belongs to you now.

01:42:17   You can talk about whatever you want here and we'll go elsewhere."

01:42:20   And that, I think, is the dynamic.

01:42:22   If you have a very popular community, you have to have some kind of rules to control

01:42:27   Otherwise you will inevitably only be left with the worst of the worst because there are some set of people who are going to be

01:42:34   Obnoxious and other people don't want to be around obnoxious people and eventually all you're left with are obnoxious people

01:42:39   yeah, I mean this this whole this whole mess of

01:42:43   What it takes to to maintain a community site and all the garbage you have to deal with as the platform owner

01:42:51   Like I just you know, I saw I saw a lot of the tumblr. I've seen a lot in the past with sites

01:42:56   I've been a part of, it has made me never want to make something that includes hosting

01:43:03   user published content ever again. Someday I might forget that I'm saying this and do

01:43:07   it anyway, but there are some features of Overcast that I have considered doing or some

01:43:13   related things like, "Oh, maybe it'd be cool to make a podcast CMS and host a podcast for

01:43:19   people like the Tumblr for podcast kind of thing." Then I think, "Oh, then I have to

01:43:24   I have to deal with all this crap.

01:43:25   I have to deal with, oh well, someone's gonna post,

01:43:27   you know, a hate podcast, and I gotta hear

01:43:30   from some police officers and then take it down,

01:43:32   deal with them, you know, arguing with me,

01:43:33   and it's like, I don't wanna have to deal with any of that.

01:43:36   It's so, and even simple things like user reviews

01:43:39   of podcasts, I never wanna do that for so many reasons,

01:43:43   and this is one of them.

01:43:44   Like, there's so many things like,

01:43:46   I don't wanna have to deal with, it's like,

01:43:49   the position I take is much like some of the design decisions

01:43:52   Nintendo has made with online gameplay.

01:43:54   It's like you don't really give people a way to be horrible

01:43:56   so you don't have to deal with it.

01:43:58   That's generally my goal is like just make products

01:44:01   that don't involve people publishing stuff

01:44:04   on something you own and you having to deal

01:44:06   with the ramifications.

01:44:08   - It's hard work and volume wise you've got a problem

01:44:10   but I just, I'm boggled by the easy decisions.

01:44:13   I feel like if you did like a podcast hosting thing

01:44:15   and someone put up like KKK Weekly,

01:44:19   I would feel like no, boom, out.

01:44:21   You know, and why do you get to cut off?

01:44:22   It's my company, I get to pick what I want.

01:44:24   And it's like, oh, Slippery Slope,

01:44:26   don't try to put your podcast on Marco

01:44:27   because he censors you.

01:44:28   I honestly think that nobody would be like,

01:44:31   because Marco got rid of the KKK podcast,

01:44:33   we should not put our podcast on his site

01:44:36   because he's obviously,

01:44:37   like the whole Slippery Slope central thing,

01:44:38   like most people will be like, all right, well, duh.

01:44:41   Like, why would he want that there?

01:44:42   Well, you know, it's like, well, he was gonna ban that.

01:44:45   What else is he gonna ban?

01:44:46   What about if he doesn't like vegetarians?

01:44:47   I can't have my vegetarian podcast.

01:44:48   Really?

01:44:49   Are you really afraid of that?

01:44:50   Like, I feel like that is your job

01:44:52   as someone who starts or runs a company to say,

01:44:54   yes, I am imposing my values on the private company

01:44:57   that I own, like that's, yes, that's what I'm doing.

01:45:00   I am not the US government.

01:45:02   If I don't want vegetarians, I can ban them.

01:45:04   And then they can say, don't go to Marco's podcast site

01:45:07   'cause he banned that vegetarian podcast and he's a censor.

01:45:10   All right, well, fine, then that is up to you, Marco,

01:45:12   to decide what are you going to ban and what are you gonna,

01:45:14   but I honestly think if you ban the KKK podcast,

01:45:17   nobody who you care about is going to say,

01:45:19   well, I'm not gonna go there

01:45:19   'cause you banned that KKK podcast.

01:45:21   Well, you know, then go someplace else.

01:45:24   I'm not the only private internet in the world

01:45:26   that will store your text in my database, right?

01:45:29   Make your own site, go somewhere else.

01:45:31   Like obviously you get down to the hard decisions, right?

01:45:34   Well, what if you're a staunch Democrat?

01:45:35   Do you want Republican podcasts or whatever?

01:45:37   Oh, don't go there, 'cause that's not like,

01:45:39   you build the community through your decisions.

01:45:41   If you didn't allow Republican podcasts

01:45:44   and only a lot of Democrat ones, guess what?

01:45:45   You'd have an entire site filled with Democrat

01:45:48   and left-leaning sites.

01:45:49   Is that the thing you wanted to build?

01:45:50   If it's not, then you made a bad decision.

01:45:51   If it is, congratulations, you are the world's biggest collector of left-leaning political

01:45:57   podcasts in the United States.

01:46:00   You design your own community by deciding what you want to let in and what you don't

01:46:04   want to let in.

01:46:06   Not making any decision is a very big decision in and of itself.

01:46:11   I don't know.

01:46:12   I just don't know.

01:46:13   I find it depressing.

01:46:14   Well, I mean, the biggest problem is that Reddit is this big public online community

01:46:21   that seems, from the little I know about it, it seems to have, by design and intentionally,

01:46:28   incredibly light moderation. And I've never seen an online community that had very light

01:46:34   to no moderation where that worked out well. It just doesn't work.

01:46:38   I think they have heavy moderation, but just within the groups. And if you don't like it,

01:46:43   your own subreddit and then you can be the moderator and then you get to decide the rules.

01:46:47   Like that is that's the whole system. It's like you can make your own little cabal and

01:46:51   have whatever rules you want. Well, but right. But then then but then so the site as a whole

01:46:55   doesn't practice heavy heavy moderation because you can just create your own hate area over

01:46:59   here. That's why that's why people think of it as the internet like oh it's not it's not

01:47:03   no reddit is the internet like yeah within your subreddit you can have rules and you

01:47:07   can have mods and you can say in this post you can only in this subreddit you can only

01:47:11   "post things in Pig Latin, if you post anything

01:47:13   "not in Pig Latin, you'll be banned and we'll vote you down."

01:47:15   Like fine, whatever you want, but that's like,

01:47:17   yeah, they have perfect control, but the whole of Reddit,

01:47:21   you're telling me I can't start my own subreddit?

01:47:22   Censorship, Reddit is the internet, no it's not.

01:47:25   (laughing)

01:47:27   - Yeah, that's what I'm talking about, like, you know,

01:47:28   in general, there is seemingly little to no moderation

01:47:32   of the entire Reddit community.

01:47:34   It is, you know, yes, within subreddits you can moderate,

01:47:37   but there, you know, so the community at large

01:47:40   which has very little and like that, like you know we've seen I've saw so

01:47:44   many times in the past community like one of my best online community

01:47:50   experiences was the time I spent on the something awful forums and I don't

01:47:56   sound stupid now to a lot of people, but that like when I was there, I don't know

01:48:01   what it's like now, but when I was there in the early two thousands very heavily,

01:48:06   I basically lived there, that was my internet.

01:48:09   It was insanely well run, 'cause it was very tightly

01:48:12   moderated and there was a paywall to get in.

01:48:15   And so you didn't have problems of spam,

01:48:17   you didn't have people being total jerks,

01:48:20   'cause if they were, they'd get banned

01:48:21   and they'd lose their 10 bucks.

01:48:22   And if they wanted to cut back in,

01:48:23   they'd have to pay 10 bucks again.

01:48:25   So it was incredibly healthy.

01:48:27   I know that sounds crazy to people who know

01:48:31   of something awful on a surface level,

01:48:34   But it was an incredibly well-run community

01:48:37   because it had very distinct rules

01:48:39   and they were enforced very, very well most of the time.

01:48:43   And so you didn't have this kind of rush of craziness

01:48:47   from the public and just hate everywhere constantly.

01:48:50   It was very well-run.

01:48:53   And I was a member of other communities before and after that

01:48:55   that were way, way less well-run.

01:48:58   Twitter being a good example.

01:49:00   I mean, Twitter is not quite the same thing,

01:49:02   but it has many of the same challenges.

01:49:04   And there are many, many problems.

01:49:07   Once you start removing layers of moderation

01:49:11   and involvement by moderators,

01:49:15   and you start permitting more and more things

01:49:16   and having fewer and fewer filters,

01:49:17   it just, it deteriorates very, very quickly

01:49:21   into all these things that people hate about Reddit.

01:49:24   - I think the Ars Technica comments are a good example

01:49:26   because they are also kind of like,

01:49:28   we don't wanna stop anybody from saying

01:49:31   whatever they wanna say, so on and so forth,

01:49:32   but they have guidelines that seem unenforceable.

01:49:35   They basically have like, don't insult other people,

01:49:37   like no ad hominem attacks, right?

01:49:38   That's like, oh, that's totally unenforceable.

01:49:40   How can you have a board where people argue

01:49:42   over Mac versus PC or argue about Scientology,

01:49:46   argue about global warming and enforce that?

01:49:48   It's impossible to enforce.

01:49:49   People are gonna call each other jerks, right?

01:49:51   They do it with human beings.

01:49:54   Human beings look at things and say,

01:49:56   is this person just saying something

01:49:58   that the only purpose is to put this other person down

01:50:01   doesn't add anything to the argument, like, then you can, I

01:50:04   mean, this is even before they had downvoted, like the

01:50:06   moderators will come in and moderate a particular comment by

01:50:08   a particular person. And then they would complain about it,

01:50:10   and they would fight over it. And it's just a tremendous

01:50:12   amount of work. But having those people in there constantly

01:50:15   trying to make those decisions about our Are you violating our

01:50:18   posting guidelines, because the posting guidelines were just so

01:50:22   like, it's seemingly unenforceable, certainly not

01:50:25   machine enforceable. And people can just make new accounts and

01:50:27   so on and so forth. But they, you know, they had a few things

01:50:29   over there. One, having an account with a lot of posts and a long registration date

01:50:33   was seen as something valuable, like you had reputation based on how long you'd been there

01:50:37   and how many posts you'd made and so on and so forth. Again, this is even before voting.

01:50:41   So people didn't want to just abandon their account and start a new one. And that meant

01:50:44   that essentially the equivalent of Twitter eggs were an indicator that maybe someone

01:50:51   is just here to cause problems if they have one post and they registered their account

01:50:55   today or whatever, even though registering the account was free and stuff like that.

01:50:58   But just having those moderators in there trying to do it, trying and basically failing,

01:51:02   because you can't place a community this large, but just trying to do it.

01:51:06   Knowing that if you say something too obnoxious, it's kind of like speeding.

01:51:11   If you go 70 miles an hour, maybe fine.

01:51:13   But if you go 300 miles an hour, someone's going to find you and stop you eventually.

01:51:19   And so people get moderated, people get banned, people make new accounts, those new accounts

01:51:22   get banned.

01:51:23   It's just a constant battle.

01:51:24   I just feel like having people in there trying to do the right thing, the right thing according

01:51:30   to you know, it lets you know what does Ars Technica think is the right thing to do?

01:51:34   Ars Technica thinks the right thing to do is to not be insulting to each other.

01:51:39   That's I mean, it's not, again, not a high bar.

01:51:42   People can still say some pretty terrible things within the limits of the guidelines,

01:51:46   but it lets you know, is this a site where I'm going to come on and people are going

01:51:49   to tell me to kill myself immediately?

01:51:50   No, because that person eventually will get, you know, their post will disappear and now

01:51:55   they have downvoting and that account will get banned.

01:51:57   And like, there are people fighting on the side of the users, so to speak.

01:52:01   There are people there trying to moderate and as unsuccessful as it may be, it sends

01:52:06   a signal about the site.

01:52:08   Like the site sort of has a personality.

01:52:10   And the personality is this bad thing that most people think is bad, this site also believes

01:52:17   is bad and is trying to do something about even though it's really hard.

01:52:19   [BEEP]