534: On the Cheerleader Bus

45:26But Apple hasn't said anything one way or the other, so it makes me think that this is Final Cut Pro Jr. Some of that I think is unavoidable, just because when you look at a lot of the stuff you were just mentioning, when you look at the actual needs of Final Cut Pro users in the field today, they scale up dramatically past what the iPad could do or what you could easily get into or onto or off of the iPad. So they can't really do everything that Final Cut Pro does on the Mac on the iPad. So there had to be some omissions and cuts and everything. The file format stuff, you know, that's, that could be for other reasons.

46:01Um, you know, it could be that this is the future format of final cut. It could be, you know, that, that the old format or that the, the desktop format, um, just has stuff that the iPad can't do well or won't handle well. Um, you know, so there's all sorts of, of plausible reasons for that. It would be easier to make technical arguments if Max didn't literally have M1 SOC and like, this is not an iPad SOC. It is a Mac, the m does not stand for ipad it's like it's like so you know it was like like again the only thing that it really has against it is just lower ram and i don't even know off the top of my head some of the the big ipads might even have eight and you can find final cut in eight which is you know not great but anyway apple didn't explain um so this this is an area to watch in the coming years assuming this application continues to be a thing ipads will get more powerful and eventually will be at the point where like the cheapest iPad Pro you can get is more powerful than, you know, the Macs that are running, you know, the minimum spec Mac that you can run real Final Cut on today.

47:01So in the interest of being completionists, we should probably have a little bit about Logic Pro specifically. Creators can make precision edits and draw detailed track automation with the Apple Pencil and connect a smart keyboard folio or magic keyboard to utilize key commands that speed up production. That's also true of Final Cut Pro. Logic Pro for iPad supports round-trip capabilities. Hey-oh! Making it easy to move projects between Logic Pro for Mac and the iPad. But then Jason Snell clarifies, except there's just one thing. Many Logic users also use third-party audio plugins. You may not know it, but iPadOS supports Apple's audio unit plugin format and has for a while now.

47:38But the only catch is that the maker of the plugins you rely on must make iPad versions available, or your quote-unquote round-trip logic project really won't be. some pro filter makers like fab filter support the ipad others like isotope seem to have not even heard of the ipad sad trombone sorry marco uh your mileage may vary so uh full full marks for supporting round trip in principle but a bit of a bummer and i guess it's at least partially beyond apple's control but a bit of a bummer that the plug-in story is lacking at this moment well and and this i mean again like this is that this particular thing you know i actually don't use isotope's plugins i use the isotope editor app because i find the use of plugins adds a lot of latency and and burden to the actual logic editing process so i just i pre-process the files in isotope's own app and then i import them into logic but like again trying to figure out how to do that kind of stuff on an ipad think about the all the limitations the ipad os will fight me on this so first of all my podcasting workflow i use audio hijack pro to record and do a bunch of other stuff it broadcasts the live stream it records a different backup format you know it splits the different parts i'm also i'm recording the contents of our zoom call this is all stuff you literally cannot do on ipad os and then i take those files i upload one by dragging and dropping to our cms um you can do that on ipad os it's clunky but you can do it um and then i i run a command line script that i wrote a bash script well what's that what's that i don't know i don't understand oh because meanwhile in the background my dropbox app which is actually maestral but my my dropbox client was running and got the files that the two of you put in the shared folder that we have i run a shell script that i wrote in bash that takes those files puts them in the right directory that i actually like you know run some stuff on runs ffmpeg to convert a couple things like between formats and sample rates and whatever else then runs a custom a custom app i wrote to line up the tracks uh that i've never released and i'm probably not going to so please don't ask it's it's sloppy you don't want it but anyway um so it automatically syncs up the tracks runs them through my voice boost command line utility that i also haven't released that applies eq and volume normalization to each of our voices based on a profile i've created for each of us then it gives me those wave those wave files i can either import into izotope if i need to do any kind of d reverb or in most cases i don't so i just import them directly into logic those files by the way are about a gig each so a gig per person per atp episode now imagine the process of trying to do that kind of workflow on an ipad because on my mac it's running one shell script that i wrote you know eight years ago and i've been reusing this entire time you can do some of that with shortcuts.

50:23Some of those apps exist on the iPad, but it's a way, way smaller ecosystem. And the process of dealing with all these different files and different processes is much more clunky and much more cumbersome on an iPad OS. Some of these things aren't possible at all, like the application that records the output of another application that doesn't cooperate with it normally. So again, I think these apps will be good for people who have, first of all, very different needs than me obviously logic is not made for podcasting again it never stops reminding me of that when i'm using it um but but you know these are apps that like people who use these have complicated workflows sometimes and complicated needs that ipad os is going to keep getting in their way so on one hand i'm really glad that apple did this i was not expecting this this came out of nowhere for me it shows that they you know this is a large amount of software effort that they put in here and apple does not put in large amounts of software effort for their applications these days like when's when's the last time apple made a really great first party app not os not not platform app it's few they're few and far between these days and so to do this kind of effort is unexpected and very very good to see that being said it also does highlight for a lot of people and will highlight for a lot of people who try to use it, the limitations of iPadOS for doing this kind of work where you're having to integrate stuff from a lot of different apps, you're having to pull in different files and different formats and use different tools to operate on them.

52:00That's not the strength of iPadOS at all. It fights you at every turn. So hopefully this is one part of a larger effort to maybe make iPadOS better at dealing with those things. And I know that's a tall order because, you know, We've seen over time it's very difficult to add power to iPadOS without making it confusing for the people who want it to be simple. And so that's going to be something that hopefully we'll see more of at WBDC. Maybe we'll see some kind of direction for iPadOS that is more than what it seemed like over the last year or so, which is basically half abandoned. I would rephrase that, though.

52:39You said we've seen that it is difficult to add pro level more functionality without making it harder for the casual user. I think Apple has been unable to do that. I don't actually think it's difficult. What Apple has always been trying to do is to sort of strike a balance, to have their cake and eat it, too. We want to add pro level features, but we want to sort of make them accessible to casual users as well, which is an admirable goal. But they've been failing at that goal. if you want to add pro level features and not trip up casual people it's real easy to hide them it's really easy to have them all turned off by default all completely invisible you know like that that's not ideal you don't want to do that you don't want there to be this hidden underbelly or whatever but you can do that an example of that is the unix stuff on mac os it's there but if the casual user never even seems to know the terminal application exists it's in the utilities folders not even an application so they never even need to see it but so much functionality is under there for the advanced users who know where it is but it does not shove its head in you know shove you know it doesn't rear its head it's not in your face you don't have to know it exists at all to be a casual mac user apple could have done stuff like that on ipad os they could have you know made it possible to have multiple audio streams and more generic file complete file system access and all sorts of other things like you know a terminal a command line like all that stuff is under there and could be there in a way that does not impair the casual user it's just that every time apple has tried to do that they've tried to not bury it away they've tried to integrate it to give an example of when they've succeeded cursor support on ipad os that is a more advanced feature that they surfaced in a way that does not interfere with casual users but also provides casual users with additional functionality.

54:30It meets them where they're at. You don't have to know every single detail of all the fancy stuff or whatever. It is very iPad-like, it's friendly, and you can sort of gradually work your way up to it. An example where they've not done that is multitasking, where they keep trying to have new paradigms and things to do, and they have impaired casual use, and they've also been non-satisfactory. So, yeah, the way I would phrase it is Apple has thus far proven unwilling to just do the simple thing, which is to hide all the advanced features away. And I, you know, like I said, that's admirable. I admire them for not taking the lousy way out, which is, ah, we'll just hide it all under a terminal and you got to do all this stuff from the command line.

55:09But practically speaking, it means that the stuff you were talking about just plain isn't possible for anyone. Like every time I see a 700 step shortcut, it's like it's impressive in the way that a house of cards is impressive. but it's like come on you like it as as clunky as your shell script might be for someone who knows programming it's so much more tractable than a shortcut that scrolls for 17 pages yeah yeah also it's a really short show like it sounds like it's really complicated oh my god i run my files through a shell script it's really not that it's like it's like 20 lines it's not much if you did that in shortcuts it would not be 20 lines no definitely not nope it wouldn't be possible first of all as you pointed out like literally not possible due to sandboxing and restrictions and stuff like that but even if it was possible it'd be so much more complicated yeah so anyway i i think this is this is overall a very promising story like again this shows a level of investment in the ipad platform that frankly i don't think anybody was expecting apple to be doing right now like i think everybody who uses the ipad to get pro stuff done if there's any of those people left they had kind of felt like apple's you know apple has one foot at the door with this platform like for pro use i don't i don't know a lot of people who are happily using ipads for pro use in this past year i know i think stage manager really threw everyone for a loop and the the kind of built up frustrations over time and apple's lack of progress in this area i think turned a lot of people off and and sent a lot of people back to back to or to the mac to do pro level work and i don't know that this is enough to change that obviously for people with certain needs maybe it will be uh but it's a really good first step to show like no they're not giving up on the idea of a pro ipad workflow uh platform and and you know i i don't know if it's going to work but i'm really happy that they're doing it i i think this is healthy for them um as a user of logic every week for my work i'm a little concerned that they're going to you know axe the mac version and make this the new version via catalyst and maybe it will be even less suited to podcast work than it currently is for some reason i don't know yet time will tell but as just an apple watcher in general this is a this is a cool surprise again it shows investment that i wasn't expecting in the platforms and in their applications frankly so i'm tentatively optimistic i think it's a pretty cool thing overall hopefully it doesn't bite anybody in the but in the future in terms of what these apps are on the Mac.

57:37And we'll see what happens with the iPad. I think I would like a little bit more bottom up on this. This is top-down stuff. Let's make the pro applications for the iPad a similar top-down type of thing. We were always wondering about, asking for, speculating about Xcode for the iPad. Instead, we got Playgrounds. That's top-down stuff. It's like here is the stuff that the user interacts with. or bringing pro-level functionality to the iPad. You need bottom-up stuff as well. And they've added a little bit of that. They added virtual memory recently to the iPads that are capable of it. That's the type of bottom-up stuff that you need to support the top-down things.

58:16But there's a lot of bottom-up stuff that is lagging behind the top-down stuff. Logic and Final Cut could surely benefit from a more sophisticated audio-video underlying system in terms of interaction between other applications. You know, like Final Cut has the plugin system that it uses with audio units. Not Final Cut, Logic does. Final Cut, I think, has some integration with if not third-party stuff, then like Apple's compressor app. Is that still a thing? Apple's compressor app is a separate app? It is, and I own it. I don't know if it will continue to be a thing. It's basically Apple's version of FFmpeg.

58:53Yeah, but that type of coordination is more difficult on iPadOS. You know, or even just like, you know, we can't do recording the the the audio subsystem the fact that audio hijacked nothing like audio hijack exists on the ipad and can't exist and ipad os itself seems to have limitations in terms of controlling multiple audio streams and multiple applications let alone hijacking them or coordinating between them that's the lower level stuff that lets the higher level applications actual actually achieve parity with the max ones right because the mac can do all those things and now you know apple can cheat like apple's applications can use private apis apple can put things in the os only for them again without an apple presentation about the technical underpinnings we don't know how much of that stuff they did how much bottom-up stuff did they have to do in 16.4 to make these two applications possible and can that stuff benefit third-party applications on ipad os they feel like now they're a little bit top heavy um and on that front it's like it's great that they did these for the ipad but it's all about the follow-through right you know are they going to continue to update them um is there is there some kind of target they need to hit in terms of usage to keep these teams together are these teams going to go off and do something else right are they really dedicated to this in those ways it would actually benefit the longevity of this software on both platforms for these to be the secret future mac versions because then at least they have just you know everybody's working on the same project and they can try to improve it again kind of like they did when they made the new version of iMovie which had uh which did such a bad job of covering the feature side of the previous version that they can apple continued to ship the previous version i forget what the numbers were it was like iMovie 8 or something it was you know it was kind of embarrassing but like but apple believed that the underpinnings of the new version of iMovie were important enough to get out there and eventually the underpinnings of the new version of iMovie became the only version of iMovie and i think they use a lot of that technology and foundation for the future version of final cut which also was a disaster in its own way because it took away a lot of features that the old version used and you know, hopefully Apple learns from this, but I do worry about them spending what must have been years or, you know, however long to make these seemingly very good versions of Mac applications on the iPad.

1:01:08And then just kind of like forgetting about them going, Oh, we tried that and it wasn't that great. So nevermind. Like it's like, we talked about this with the pro max, you know, are they really, uh, you know, dedicated to the Mac pro? Are they going, are they behind this machine are they going to update it pro software is the same thing they make these things they seem like they're really great they seem to support them and then some of them just kind of fade away or just like you know live on an undead form or get canceled unceremoniously that's the life of the quote-unquote pro uh user of apple stuff both on the hardware and the software front so fingers crossed yeah and i think ultimately though like the the odds of these becoming the new mac apps also i think is so high in part because of that factor that apple does not multitask well um and and apple does not like to repeat work between platforms i think we've seen over the last several years whenever apple has a chance to unify something between mac os and ios they'll take it and they they don't they don't want to keep making a mac app if they have an ipad an ios app that they can just port over to the mac and make it one unified app and and frankly i mean while i wish this was different i also can't blame them i mean i don't like writing a mac app either like i'm currently like trying to do everything in swift ui i'm like you know having to port everything over between you know all the all the ui kit stuff and all the ns everything stuff is terrible and it's it feels like wasted effort um and so anyway i i think what we've seen from apple is that they will unify when possible and ultimately as much as the transition kind of kind of hurts in the meantime you know look at messages of the mac it was pretty rough at the beginning when it when it became a catalyst app but now it is a better app than the old you know app kit native version of messages because now it has feature parity with the ios version which we were all using anyway And this is a little bit different in the sense that these apps were, you know, these pro apps are Mac apps first.

1:03:13And so, you know, we're not crying for feature parity on the Mac. However, this is going to allow Apple to, you know, after maybe a bumpy intermediate period, to update them more often. If they can justify the effort to say we're going to have one version of these apps that runs on all of our major platforms, that will help them justify having engineering time on them. And while it would be great if Apple could use their vast resources to work on both, like in parallel tracks, we know by this point, we know Apple will have to know they won't do that. Like they will they will half ass one of them at least.

1:03:51And if you're lucky, it'll be just one, if not both. And so if they have less of a surface area that they have to work on, odds are better they will keep it updated and keep moving forward. So... this is probably going to be the future of these apps on the Mac at some point. I don't know when they'll make the changeover, probably soon. And I think at that point, we will have round tripping between them. And the pros will all complain about all the stuff they're losing in the transition. And it'll be bumpy for a while and we'll get through it. And at the end of the day, we'll have apps that are eventually good and updated.

1:04:27Can't wait for Final Cut Pro for watchOS. We are sponsored this week by Backblaze, unlimited computer backup for Macs, PCs, and businesses for just$7 a month. Backblaze is a wonderful cloud backup service. I have been using them for many, many years now. I so strongly recommend that everyone should have cloud backup. It's automatic. It runs in the background all the time. You don't even notice it running. You don't have to think about running it. You don't have to tell it to run. You don't have to have a certain drive connected to your computer or be in a certain location. It's just always running.

1:05:01And it protects you from so many different hazards because it's not just like, you know, something on your desk sitting there in your office. So you don't have to worry about like fire, flood, theft, power surges, all that stuff. It's just cloud backup. It's automatic. It is wonderful. Backblaze has restored over 55 billion files for their customers. And they have so many ways you can restore files if you need to. So first of all, even if you have a data loss, suppose you're just like away from your home computer, you can access your files on their phone app or using a web browser. You can get your files right there wherever you are.

1:05:30So that's great, even if you haven't even had a loss. But suppose you've had a loss. You want to do a big restore. You can do it on the web interface. you can do it in the app you can even have them send you a hard drive if it's like a lot of data you don't want to download over the internet they can mail you a hard drive you send it back within 30 days you get a full refund on the cost of the drive they also have now you know if you have concerns about maybe accidental deletion for an extra two dollars a month you can increase your retention history up to a year so they have great features at backblaze i love cloud backup and backblaze is by far the best one i've ever used so don't be that person who forgot to back up their important files sign up for a 15-day free trial today with no credit card required at backblaze.com slash atp start right now free trial once again no credit card required go start backup your files backblaze.com slash atp by far the best cloud backup service i've used and i so strongly recommend it please go use backblaze save yourself backblaze.com slash atp thank you so much to backblaze for sponsoring our show

1:06:34All right, so in the show notes for two or three weeks now, we've had a heading, Mastodon, Blue Sky, Twitter, etc. And I guess this was cued by an Ask ATP question from Stephen Wood. How are you guys finding Mastodon engagement compared to Twitter? I've stayed on Twitter but missed some of the people that have moved away, including yourselves. Just curious if you're enjoying it more. And it seemed to me that, John, you had thoughts that you wanted to share. would you like to share some thoughts yeah um so blue sky i believe we talked about it on this show ages and ages ago when i was talking about it as a uh i think maybe when we first talked about mastodon i said well there's this project blue sky that twitter uh started that is trying to make a decentralized uh you know twitter-like service and they're working on their own protocol and supposed to be all open and so on and so forth and i don't know if that's going to come to anything but in the meantime we've got masted on right um i had signed up for the blue sky whatever it was like hey sign up here if you want to hear when we actually do anything i signed up for that you know months and months ago maybe years ago whenever the project first came out eventually my number did come up and they said hey blue sky is ready for you to check it out so i signed up for it this was months ago and uh got onto it uh and used the app on ios and i was like well this you know here it is that's kind of like twitter uh and it's on my phone and the app is not great and whatever and i just ignored it for a while um and mass and of course we've talked about many past episodes that we were all on it we're all doing sorts of all sorts of stuff we talked about the various applications so on twitter continues to in its uh slow decline in functionality and enjoyment most of us are mostly off of it some people are still there um but recently in recent weeks blue sky has seen uh you know a surge in awareness let's say i'm not say a surge in popularity because blue sky is still invite only um and so the population of blue sky is being kept artificially limited by the number of invites which is very small in the tens of thousands i believe as compared to the millions that are on mastonaut and the many many more millions that on twitter um but it is in the public consciousness uh and so i think it is worth talking about how these different platforms are shaping up and my main thought on this topic uh is that the way these platform services protocols so on and so forth are you know are developing has almost nothing to do with any of the technology on either one of them the technology itself is interesting we can talk about activity pub versus at protocol and versus twitter thing or whatever but i really feel like that is not the story and i think it is apt for social networks to not be driven by technology it is entirely driven by social aspects who is going to each of these services why and you know what forces are are moving them there because kind of like any empty building like the services are defined by who shows up who shows up who is there what do they want to do uh what do they do when they're there uh and that really shapes the services so despite the fact that the all the activity pub people in the blue sky people are arguing back and forth about their technology stacks and stuff like that.

1:10:01That is not the story here. The story is entirely about the people that arrive. And in particular, Blue Sky, you know, having been spawned off from Twitter, I believe it was started by Jack Dorsey, Twitter. It was maybe his baby or maybe the CTO who then became CEO's baby. And it got spun off of Twitter. So now it is not owned by them, I believe. But anyway, Blue Sky was backed by money from rich people. let's put it that way right i don't know if it's a venture capital thing or whatever but like jack dorsey's rich person they've got not a lot of money but millions of dollars right behind them to develop this protocol to hire a bunch of people who worked for a while with nothing to show for it who are still working on this protocol and have shipped this sort of prototype application there is no business model there is no money coming into this this is entirely a cost center that millions of dollars are being funneled into at this point right mastodon is an open source project started by one person so the the sort of pedigree and origin of mastodon versus blue sky could not be more different and i think that difference in origin actually does have a lot to do with who has shown up um blue sky being the thing that was started by millionaires with millions of dollars poured into it has ended up attracting people who did not end up uh on mastodon because now not because there's anything wrong with this this is not a value judgment but i'm just going to say this is as what i believe to be a fact rich people know other rich people like the circles that rich and powerful people travel and they touch each other right you are more likely to know a senator if you are a rich person this is just the way it is and if you're a senator you will know way more rich people than you will if you're not a senator let's say same thing with celebrities rich people powerful people not that they're all one big happy family and they all know each other but you will have occasion to be fewer degrees separated from them the same way in a tech nerd circle that i may not know people who are like rust programmers for android or something but they are way closer to me in my apple mac circle than they are to someone who is into rock climbing because that just just the social networking like who knows who i know this person knows how many degrees away are you from kevin bacon or whatever like literal kevin bacon is way closer to the rich people who funded blue sky than he is to most of the people are on mastodon and i think not that that is the one only explanation for why celebrities are on blue sky not on mastodon but it has a surprising it's surprisingly large amount to do with it because and i say that because having been used used both the services for you know they're almost their entire existence practically speaking mastodon has more people has better applications has more features uh and despite the fact that they're arguing about oh but mastodon makes you pick a server or whatever a that's not actually true anymore mastodon defaults to one application and B, Blue Sky gives you the exact same option.

1:13:10It says, do you want Blue Sky or some other thing? And there is no other thing you can pick right now. But that is not the difference. The difference isn't that your favorite senator tried to sign up for Mastodon and got scared away by the thing where it asks you to pick a server. That did not happen. Your senator did not ever try Mastodon. Your senator had never heard of Mastodon. Or if they did, they heard it once and ignored it. But when enough of their rich, powerful friends mentioned Blue Sky, that made them, the same way you're going to go try it if you hear something from a friend of a friend than if you just see it randomly on the internet.

1:13:40Like, oh, maybe that's something I should check out. If you're thinking about checking out Blue Sky, think about where you heard it from. How many degrees away from yourself did you hear about Blue Sky? And what made you check it out? It's because you trust those two or three connections to say, maybe that's something that I might be interested in. Because a friend of a friend of a friend says, hey, Blue Sky, whatever. Maybe you want to check it out because you heard your favorite celebrities on Blue Sky. That is the degrees of separation. I am a fan of Tom Cruise. Tom Cruise is on Blue Sky. I don't know if he actually is or whatever.

1:14:10Nothing to do with the actual services. Blue Sky is worse than Mastodon right now because it's younger. It is younger. They're just working on it. The protocol is not done. The client is bad. The client's not done. Like it is very young compared to Mastodon. Mastodon has been around for literal years. Even though it's only had one person working on it, they've got a huge head start. But there's that, you know, the X factor. It's not an X factor. Celebrities are on Blue Sky. Celebrities are on Twitter. fewer not none but fewer celebrities are on mastodon uh and right now in the very very early days of mastodon that is defining the big difference between these platforms and i've seen a lot of people say well blue sky is more fun and blue sky is just better in this way or that's like that may well be true but i think it's because of the people that are on blue sky it has nothing to do with the app which i hate with a fiery passion has nothing to do with the protocol which actually is very interesting and cool it has some advantages over the mastodon protocol and it's all to do with the people that are there that's the game in social networking it's network effects it's who's there it's it's you know you could call it marketing or whatever but honestly i don't think either mastodon or blue sky have some sort of master marketing plan that is causing this to happen it's happening more or less organically because of the seed of where these things came from what their pedigree is and who knows who and whatever so that doesn't mean it's set in stone forever things can change real fast it's early days in both of these services now but if you're wondering what the difference is technically speaking that's not the answer practically speaking is who's there and do i want to be there do i my my favorite celebrity and or government person and or whatever is on this service and i want to be in there because i want to see their tweets the big fun threads reverends posting pictures of their butts and the hell thread and all you're hearing about that because powerful people people with big audiences celebrities they're all there when aoc is posting funny things off the cuff on blue sky it's because there's only a you know 50 000 people there and she feels more comfortable and why is she even there because you know connection to connection to connection i really what i'm rooting for right now is i really hope the blue sky apps will get better there are third-party blue sky apps but they're not any better in my experience than the first party one none of them preserve my timeline position which is a deadly sin for me but boy uh if you're currently on maston on enjoying the wide selection of really good applications, including the one really good one that I use, Ivory.

1:16:33Appreciate what you have, but you may be sad that your favorite celebrity is not there. I heard sometime in the last week or so, somebody describe the two services approximately thusly. They said, if you were a Twitter person and then you decided to leave Twitter, but you really like Twitter, you still enjoy Twitter, then you're going to probably end up on Blue Sky because Blue Sky looks like Twitter, feels like Twitter. I just got an account in the last two or three days and I'm casually lurking and occasionally replying to things. Blue Sky looks, feels, smells, walks, quacks like Twitter or Twitter of like a year ago.

1:17:16If you were on Twitter and wanted to leave Twitter but you're kind of sick of Twitter, then Mastodon is for you. And I think that obviously there are exceptions and that's not a universal truth, but I think by and large, that rings true for me. And I've been enjoying Mastodon quite a bit. I've been using Ivory and it's excellent. I cannot begin to tell you how much I dislike the apps, just like John had said. The web app for Blue Sky is not great. The mobile app, well, I mean, the web app and mobile app are both deeply impressive given that it hasn't been that long to my knowledge that they've been in development, but they are not great.

1:17:55The whole vibe on Blue Sky is very different than Mastodon. Mastodon's a bunch of nerds. That's good. And it's bad. It's a little of both, but it's a bunch of nerds. Blue Sky is not. Blue Sky is, it seems to me anyway, to be a lot of people just kind of kicking the tires and kind of s***posting for the fun of it. not necessarily in an angry, evil way like he used to be on Twitter, although I'm sure that's happening too, but more in a ha-ha kind of way, which that probably doesn't seem like a strong distinction, but in my eyes it is. It's not a bad vibe over at Blue Sky. It's just very, very, very different.

1:18:36And I agree with you, John, that I think it will continue to be very different because regular people seem to be embracing Blue Sky to the speed that they can, given that it's all invite only, whereas the nerds are embracing Mastodon. And I do agree that up until very, very recently, the onboarding process for Mastodon was gross. And I understand why it was the way it was. I get the decisions that were made to land there, but it was tough. It was gross. And Blue Skies was - I wouldn't say it was gross, but part of it, it was derived from the fact that Mastodon is many years old. because they were mass on is already decentralized and so when you have a service that is already decentralized you have to present that decentralization in some way otherwise it's not really decentralized blue sky doesn't have that problem it's not actually yet decentralized right in the beginning mass and i wasn't decentralized either when that guy put up his first server there was only one because he was the guy who just made it and there was no you know and so but in you know they're like look mass on is many years old we have a a rich tapestry of instances for you to choose from because we're many years old we're literally decentralized of course that has to be in the applications it's like a competitive advantage for blue sky to say we only have one thing for you to choose from but like i said blue sky still puts up the screen that says please choose your instance and the choices are the one that exists and please enter like other underscore underscore and like it's just like and there is no other as far as i know there's literally nothing else but the ui is still there for it if we could fast forward seven years and blue sky still exists and it actually is decentralized they're going to have the exact same problem that mastodon has which is oh blue sky is decentralized and there are 27 instances which one do you want to pick will default to this one but you get to pick from other ones and so i you know i don't i don't buy that argument at all even in the sort of like hurdle type thing and because i feel like if those same people who who came to who came to blue sky and are all having fun in the big threads if those people all showed up on mastodon they would be having more fun because they'd be using ivory which is a way better client and they'd be posting all the same things and because it's so few people now it's not like all these people these are the people who could serve could don't but could survive without the algorithmic timeline you know mastodon doesn't have a big algorithmic timeline thing blue sky has one bad algorithmic timeline called what's hot that keeps changing all the time but the numbers are so small that if all these people showed up on mastodon they'd be having all the same amounts of fun the only thing they'd be they might get derailed complaining about the like quote tweets or whatever but like it's so new on blue sky it's like hey everybody check out this new thing and the reason i know this is because i remember hive do you remember hive this is the thing from like a few months ago that was done by like two kids or something yeah two million people signed up for it in like two days again blue sky has like 50 60 000 or something two million people signed up for hive you know what hive was like it was like the blue sky health threat it was it was a bunch of people from a specific community and they were the celebrities from the gaming world were there and they were all posting and having fun like oh isn't this a fun service and then of course it all collapsed because it was run by two people or whatever right and that was centralized you know setting it aside people can have this kind of fun if they all show up and it's like a flash mob everyone shows up at a particular location and they have all the fun it's it's a question of whether they'll be able to stay there i don't think blue sky is obviously not probably not going to collapse like hive did uh but then again hive did have way more users for a brief moment before it went down the tubes uh mass on is just a more mature platform that these people could have come and have fun on.

1:22:10But then you get into the things that are actually different, as in, what does MassDom want to be? What does Blue Sky want to be? They do have different aims there in terms of, you know, Blue Sky is big on, they do want to have different pluggable algorithms and pluggable moderation systems, things that are not sort of in the goal set of what MassDom is doing. Blue Sky wants to be decentralized, but isn't, blue sky hasn't figured out their protocol to the degree that active activity pub has activity pub is an open documented protocol supported by you know w3c or whatever blue sky doesn't even have that protocol done yet right so these are two things that are in different stages of development and the vibe is different on them because they're they're just doing different things they're they're different ages they're different stages they're different populations of people they're very very different but in the end i think both of them do want to be like the best of twitter like margot was talking about when he finally came on board mastodon once he started using ivory which is essentially tweet bot for mastodon it was like oh this is essentially twitter it because twitter to me looks like tweet bot and now i have essentially tweet bot for mastodon and i just slide right in like a you know i love if you love twitter you're gonna love ivory on mastodon i think that is a totally valid statement you can will eventually be able to make the same statement of blue sky i think they both want to be like the best twitter ever was they're just at different stages along the path to getting there and they've taken a few different turns here and there i mean i would say though the the crowd being different like what casey said earlier that's a massive difference in in what it's like to use these two services i mean you're right the apps are very different The MasterDot apps are great.

1:23:52The BlueSky app is terrible. But if you'll forgive a quick story, in high school, I was in the drum section of the marching band. And there came a time when a few of the drummers were asked to ride with the equipment that would go to the away games. But the equipment wouldn't fit on the bus with the rest of the band. It would fit in an empty section in the little half bus that the cheerleaders took. So for about a year, whenever there was an away game, like three of us drummers would ride in the back of the cheerleader bus. That would be so terrifying for me as a high schooler. Oh, my word. Yeah.

1:24:34So this sounds like it would be amazing. Like if you just hear it, like, oh, my God, that must that. What a dream. and what it actually is is kind of terrifying because you're these like three dudes who are completely invisible in this environment and you think you think the oh maybe they'll talk to us no you gotta talk to them marco you gotta go you gotta go make some friends yeah well they spent the time talking to each other and we spent the time embarrassed to say a word and it was very clear oh this is the cool kids up in front of us here and then there's us and we are probably better off not saying anything so we don't make you know asses of ourselves um and so we just didn't that's kind of how it felt when i got to blue sky because i only got there recently like in like a few days ago and that's kind of how it felt like oh god this is okay first of all this is where all the cool people are not the nerds and believe me i love i love us nerds but it's like oh this is where the cool people from twitter went they didn't go to mastodon i'm gonna say the popular people not the cool people well okay i'm gonna draw that distinction okay this is where the popular people went all right and it was very clear and it was also extremely clear right from the get-go oh this is where all the women and people of color went because that's something you don't see a lot of on mastodon maybe that's just the circles that i'm in you know the thing is with social networks it's all about who you follow and your experience your picture of the network is defined by who you have chosen to follow and so you might have a picture of something like you might think the entire social network is person type x or is missing person type y but in reality you just don't follow them or you do or you follow too many of this person or whatever so it's hard for me to say with any certainty that this is this is the case and obviously the numbers are very different but blue sky seems at the moment to be way less white nerdy than a white male nerdy especially than who i have found on mastodon now again part of that's my fault i'm sure it's who i've chosen to follow i did notice though you know so for like the first couple days when it was only the popular kids i was afraid to say anything it's like i was it was just like being in that cheerleader bus like i'm gonna keep my mouth shut because i don't fit in here like i am not qualified to speak in the presence of all these cool people excuse me popular people and it was exactly that same feeling again um and then what happened was i found i found a couple of people i knew then i started looking through their follower lists and their follower lists and their follower lists and i found first of all a bunch of people who i actually know and i followed them i also found a bunch of new people of much greater diversity than what i what i've been finding in my crowds on mastodon and i kind of found oh here's here's here's where the nerds are the cool nerds still excuse me the popular nerds um but but you know the cool people the nerds nonetheless like okay i'm starting to find my people they're starting to be there that's a good sign and so now i'm kind of making it my own and the app does suck and there's not a not a lot of people there but the people who are there it it seems to have a level of of appeal and broad appeal that is more broad than what mastodon has attracted so far yeah and part of that is the is you know the people on twitter who are getting it part of it is the influence of the celebrities and the rick sheeple as john was saying part of that is is just that this is where people are going.

1:28:20And it's hard to nail down what gives something traction, what makes something popular or cool socially in different groups of people. It's a hard thing to pin down or control or predict or explain, but there is traction there. And so what I think we are heading towards is, hopefully, a future where both of these networks will continue to exist and be healthy. um blue sky again it's super early days they're gonna have a lot of challenges ahead of them especially once things get messy you know right now they're really small and it's invite only well at some point they're gonna either die or they're gonna get bigger and when they get bigger they're gonna have bigger problems to deal with just like we were saying with mastodon a few months ago they're gonna have to deal with abuse and nazis and harassment and and all and you know racism and all these threats like they're gonna have to deal with all the stuff that makes it hard to run social networks and hard to moderate jack dorsey it does not have a great track record in that area well i mean but the protocol has some ideas in that area which are different than the maston ideas but in theory like it people might say oh they're ignoring the issue they're not ignoring it they're just way way farther behind on what they think their path is to dealing with will their path be successful there's reason to doubt again based on what they've done in the past but maybe they learn from their mistakes they have some interesting ideas that have not been tried in the same way so that's part of what's interesting but when you're invite only 50 000 people you can get away with stuff that you cannot get away with when you're you know multi-instance 10 million people so we'll we'll see how that turns out but i i do think a lot of the things you're experiencing are more explicable than just like oh that's just the people you follow because mastodon is more mature because there are multiple instances and because it is decentralized it has allowed by design and i would say it's a good thing that it has allowed communities to form uh with uh you know surrounding certain sets of values that's what decentralization provides you can have an instance with a set of rules that you find comfortable and acceptable this is where you want to be these are the people you want to be with uh there's lots of different instances with lots of different sets of rules people disagree about the rules but that's what's great about being decentralized because mastodon was inherently a nerdy thing and was founded many many years ago the seed the sort of the seed of the communities was a bunch of weird nerds that was it seems to have been the seed of many mastodon communities but the mastodon communities have had years to grow and develop their own cultures and their own sort of values and their own moderation rules and their own moderation systems and communities have already had time to grow and then fizzle out and die on mastodon multiple times over like stars coming into existence and then fading away or supernovae right that's happened many times over mastodon what that meant was that when new people show up to mastodon it's like you showing up on the the cheerleader bus again there is an established community there and you're not a part of it and There are multiple established communities.

1:31:33And I think a lot of what has happened on Mastodon is people would show up, especially people who did not fit into the same mold as the seed communities, and they felt like this is not a place for me. And granted, Mastodon has multiple places, but you can't try them all. Maybe you try one place and it doesn't feel like the place to you, and maybe you're like, that's it. Like, that's your impression of Mastodon. Like, you don't know if there are other short buses with different groups of people who are talking about computer stuff. where you're like the short buses with the cheerleaders that is not the place for me and so you just bail on that so i have heard from many people of color who have come to mastodon and said it's not welcoming me because they would come into a community and they would say my concerns as a black person seem to fall on deaf ears in this community because what they tell me is oh in this community here's our set of rules and and here's the way we deal with things and I'm telling them as a black person, those rules don't work for me because reasons X, Y, and Z, and they're not hearing me.

1:32:31And that person bounces off Mastodon. It was like, well, you didn't try all of so many other communities. Like, what am I going to do? Try every single community and figure out if I can find the one that fits with me. So that is the strengths and the weaknesses of decentralization. It allows sub-communities to form, but the weakness is sub-communities form and they're their own rules and there's no way you can have you know a universal view of every mastodon community you know exactly where you should go and where you'll become maybe there isn't a mastodon community that works the way you would like it to be oh you should form your own that is way harder than just signing up for one simple thing blue sky doesn't have that problem invite only seated by you know fancy people who invite other fancy people like again the invite system you know makes it hard-coded dictates essentially you will be it's not just open to everyone the the seed round of people invite other people who invite other people and it's people who you know or whatever and it's small enough that it doesn't matter and so then when when somebody shows up there and says i felt unwelcome in the one mastodon community that i tried when i show up here it's a bunch of celebrities and a bunch of people you know posting and i don't i don't know if this is my community but i feel like i have just as much right to be here as anyone else because it's just a bunch of invite-only people like it's early days and that is an advantage of being young and not having established communities of not being decentralized if but if blue sky wants to travel that road seven years from now if it really is decentralized and if the decentralization really does have like algorithmic uh you know plug-in algorithms and plug-in moderation systems in different instances it'll be in exactly the same situation as a mass it on just with different seed groups so then maybe at that point maybe a nerdy person shows up to blue sky and they feel like oh i don't i can't find my community here because all these communities are talking about you know things that i'm not interested in or whatever what we want to see as we like in the whole people who promote the idea of no more centralized social networks because we've tried that many times they all suck eventually um what we want is for there to be lots of different communities and everyone to be able to find a place where they feel comfortable we're so far from that it's not even funny right we don't even know if that's a feasible thing that could even happen because the closest things we have like people keep comparing it to email but there's not really any culture in email your inbox doesn't really affect other people's in fact they shouldn't really cross over with each other right uh so email is federated and it is decentralized and you can choose your provider but honestly it does not really change your experience that much whereas social networking totally does um and i think any kind of like bridging between blue sky versus you know activity pub versus at protocol versus whatever is going to be extremely tricky and very unlikely to happen and so currently i'm in the terrible position that i do not like of you know previously i was still checking twitter but spending most of my time and energy in mastodon now i'm still checking twitter and and still spending most of my time and energy in mastodon but now i'm also checking blue sky and it's like this is this is untenable i'm checking three places i never left twitter and now i'm not leaving mastodon and now i'm blue skies too and three places to check it's just too many but the problem is the reason i didn't leave twitter was there was people still on twitter who weren't on mastodon and i was still getting value from them and i you know i i try to put most of my energy into mastodon but i'm like but i still want to read what people say on twitter now there's blue sky and on blue sky unfortunately it's not just like i want to read people who are over there i also want to participate a little bit over there so it's like oh this this cannot i cannot scale like this i cannot be continuing to check three places so i think there will be some sort of i'm hoping there will eventually be some sort of consolidation and resolution um i'm excited by the fact that people are trying lots of different things i'm excited by different protocols and different apps i'm not excited by checking three different places so i will i will uh live in this world for as long as i can but I would it's not that I want one thing to win but that's the beauty of decentralized things and open protocols is if an actual open protocol and decentralized thing wins you just need the one we don't need seven competing versions of web technology we've got the web we've got HTTP we've got HTML we've got CSS we don't need oh well you use HTTP well I use Gopher like no we don't need that just the web is fine we just need the one web I think that will also be true of if this type of decentralized social network can work at all, which remains to be seen, we only need one.

1:37:08It just needs to be at least as good as the web in terms of being open and the platform that nobody owns. And despite the fact that W3C is populated by Apple, Microsoft, Google, like, oh, the web is corporate controlled or whatever. It's not ideal, but it's the best we have ever had. I love the web if tweet-like social networking could get to the same level of openness and decentralization as the web. I would call that a big win. But you don't have seven webs despite the web three idiocy. You just have the one. So I would like us eventually to come together. Now, what was the web? It was a thing started by one person.

1:37:50I'm not saying that means that Mastodon's going to win because it was started by one person and so is the web. It doesn't mean that at all. All I'm saying is that you don't have to have millions of dollars funding you, although technically that one person did kind of have millions of dollars behind them. I mean, he was using a next to do it for crying out loud. Nobody no person in their bedroom has a ten thousand dollar next cube to develop their funny idea of how to publish scientific papers so that other people can read them. But I want there to be one of these things. I and I want the one thing to be good and I want there to be a lot of third party applications for it.

1:38:23right now I have three things and one of them is hopefully going down the tubes which is Twitter and the other two are intriguing but at very different places in their life and I'm checking all three of them yeah and I think when you look at the protocols and everything yeah they're different I don't anticipate the servers like you know Twitter is Twitter has isolated itself because of its leader and and his decision making they have isolated themselves so that they're off on an island they have nowhere to go but down whatever you think of their current state the age of Twitter getting better or bigger is done.

1:38:56Twitter is only ever going to stay the same or go down from where it is now. That problem will slowly solve itself. Then you have Mastodon and whatever else comes along. Mastodon, I think, has enough mass and enough usage. It's not going anywhere. It's not going away. For the foreseeable future, it's here. Let's take a look at Blue Sky now. I think if you look at these two, the amount of traction Blue Sky has is strong it's very very strong so it is probably going to be here for a while too hopefully i i don't i don't see i don't know enough about the protocols to know on the server side how complicated and possible it would be for following to work across servers say you know between at protocol and activity pub i don't i don't think that's going to really become a thing at the server side, but that can probably become a thing on the client side.

1:39:55So what I expect is that in the near future, or maybe even already, I don't even know, like whatever people are doing in test flights and stuff, I think it is probably pretty reasonable to expect there to be unified client apps where you can log into your Masson account and log into your BlueSky account and follow using both protocols, see the tweets and post the tweets in one app that will probably come soon if it isn't here yet and and i think that will solve a lot of these problems for us because it's all of them i mean i think sure you can make that happen it's great that you can do that because twitter explicitly by the way didn't allow that at all because they're a centralized company and they just said no uh but it doesn't solve the problem for me because yeah that can be implemented from a technical perspective but you that's not you can't like reply to one like you can't have a unified conversation in that context because Because threads can only take place on one platform.

1:40:49You can't have a thread with mixing people from two different platforms. It's insanity. You would never be able to keep track of who can see what and who's replying to what and I'm replying there. Even if you try to duplicate all your replies, it just doesn't work. Although, in all fairness, that's a problem on Mastodon 2. No, but in my experience in Mastodon, it works pretty well because Mastodon is designed for everyone to have the same conversation and the same thread. I see threads all the time on Mastodon. that's the whole point of it that people from different instances they're all participating in the same thread because they can all see the same things in the same thread it's not true when you're crossing when you're when you're crossing between mastodon and blue sky well in in practice there's a lot of things that subtly break on mastodon when you cross instance boundaries um it shouldn't but it does there are rough edges there but like the basic model of like every thread i've ever participated in mastodon has people from different instances and they're all more or less seeing the same thing yes there are rough edges yes there's annoyances and that some of that is client stuff some of that is protocol stuff blue sky hasn't even come close to dealing with any of those issues they're so they don't even have their basic single instance functionality down yet let alone their multi-instance because they're gonna have to deal with all the same stuff right but across blue sky and mastodon a unified client can exist and would save me swapping apps but the combining them into a single chronological timeline for instance like a unified timeline across services, I don't think is tenable.

1:42:10I don't think it will ever be tenable. And honestly, who would want to do that? The only people would be like super nerds who are like, I'm across all social networks or whatever. Even the most, as they say, extremely online people, they just want to be extremely online in one thing. So I don't think it's kind of like saying I have a single application that lets me browse gopher sites and websites. And I can combine all the comments on this YouTube article with the people who are commenting on gopher and the people who are commenting on the web no like it's yes you could do that technically speaking there's nothing stopping you gopher is an open protocol hdb is open protocol you could make it work but the people who are only using gopher would only see the gopher comments the people only see in the web would only see the web comments and it would make no sense to either one of them because they don't know some people can see both of them i am not optimistic about that kind of scenario i would i prefer there to be for this to be you know the platform nobody owns is not 17 platforms nobody owns we just need one platform nobody owns that covers the problem space and that has a viable path forward for continuing to advance and the web it's been a bumpy road if you look at the how has the web advanced as the platform that nobody owns but nevertheless is massively influenced by trillion dollar corporations it has not been smooth but it has mostly worked and i can say it has worked way better than twitter has worked so that's all i'm saying is like you know if if we could get half as good as the web i think that would be great i just don't i wish i could think of a third one go for the web uh i guess use net i mean we have like the phone system yeah that's a social network in a way kind of but like but yeah like that's that's more that's more peer to peer even though we have party lines and stuff anyway i i'm not i don't i don't see a future in which uh blue sky and rastodon both thrive as in like a decade from now and if we are in that situation i mean it's kind of like you know we were in that place where i was like oh well there's facebook and there's twitter and there's the web and it's like yeah kind of i guess but these these centralized social networks have burned us all so many times they've burned the people invested in them they burned the people who worked with them they burned all the users of them and everyone who thinks like facebook will live forever uh it won't i hope uh but it's lasted much longer than the previous one so there is kind of a lengthening of the timelines we are still in the friendster phase of non-twitter you know twitter-like services and in the friendster phase a lot of names came and went uh no one remembers them so we're not even at myspace yet like mastodon is not myspace and blue sky is not myspace one of them might become myspace but i don't think either one of them is Facebook either.

1:44:53And again, that's the analogy with the centralized systems. I want the web. I want the web for Twitter-like social networking, because I think Twitter-like social networking is a valuable form. Email is valuable. The web itself is valuable. You know, instant messaging in the form of WhatsApp and iMessage, all that is valuable. This form, small message, like the way Twitter, that thing that Twitter established, that is useful and valuable and apparently very hard to do well i hope we i hope we come up with a better way to do that that is not owned and controlled by a small group of people who make terrible decisions i mean and and part of you know so just before i move on like the the unified client app thing you know maybe we won't see unified timeline maybe we will start with you know the way that like multiple accounts show up in an app maybe maybe you can have multiple account types and you can like all you can have the same interface but you have to switch between the two accounts you know fine that's probably how we start and i don't think that's necessarily a bad thing because there's a lot of feature overlap and i think that's that's probably pretty straightforward but ultimately my biggest concern with blue sky is the same concern i have with mastodon i don't know how these things are going to scale in either the technological sense or the financial sense or the moderation sense.

1:46:16Those are the three big areas. And Mastodon is way ahead of Blue Sky in its scale, and they've had a long time to work out some of those things. I don't, like, looking at the realities of how ActivityPub works, the realities of running large instances or having people with large followings, ActivityPub does not seem to scale very well. it seems like it's it's holding up okay under the existing scale of mastodon most of the time that's a lot of qualifiers mastodon does not feel like something that has a lot of headroom in its design in terms of technical scaling blue sky obviously a way smaller scale currently but because there's so many twitter people designing it and running it twitter for all of its faults and even though it had a rough start scaled really well to a scale that none None of these other services can even dream of.

1:47:09Eventually. This is what we talked about when we talked about Maston. Early Twitter was very much like Maston and absolutely did not scale. And they basically redid it. Yeah. And that, you know, granted, like, yeah, it had a rough early time. But that's ancient history now. Modern Twitter scaled really well, really impressively. Everything was extremely fast, extremely reliable, hardly ever had any downtime, hardly ever had delays and stuff showing up. that is not how mastodon is right now and it's at a fraction of the scale and so i i'm concerned about mastodons and activity pubs ability to scale and in some of the protocol arguments they've actually like the people who are attacking the at protocol on tentacle means like it seems like the at protocol again i don't know much about it yet but it seems like they have they have taken scaling in a different way that is probably more likely to scale better and i and i would trust the people who are building blue sky to know that stuff better at the scale that we're talking about with you know possibly becoming a twitter-like replacement i would trust those people to know how to do that better than activity pub does it well so the thing about well mastodon interestingly uses very similar technology stack to early twitter which is kind of why we know it's not going to scale well but but because it's decentralized there's always the out of like well an individual instance can never get that big or whatever but anyway i will point out that maston is only one order of magnitude away from twitter scale which granted order of magnitude is a lot they're at 10 million twitter is a couple hundred million right this is one order of magnitude away and that may be sufficient you know with decentralization they could potentially get there blue sky looks like they have a better handle on scaling at least a better idea of how scaling will work but you don't know until you actually try it right and they're they're like because we have to see them actually do it um the other thing about twitter twitter started off you know like a ruby application that didn't scale with a you know a bad database back in and then they fixed it all and redid it all with the addition of millions and millions and millions of dollars right don't forget that it's not like they just it's not like the person who wrote the original implementation of twitter gave it a second try and it worked out no it was they have many employees tons of money a lot of that was vc but eventually they found a business model that did actually produce many millions of dollars in revenue not enough to make them you know a gang buster successful corporation which is why they sold to that idiot but still that is an ingredient that is in you know that is absent from both mastodon and blue sky because blue sky yeah they got millions of dollars poured into them but nothing like the amount of money twitter had had poured into it and also twitter eventually found a way to have revenue and blue sky is not there's not a glimmer in their eye so uh you know the the decentralization as your ultimate out to scaling is not entirely valid but if you just want to scale a little bit more and you can be like hand wave hand wave as long as we have more instances instead of bigger instances and as long as our bigger instance scales up like when i look at activity bub i think of someone looking at the web in the 90s and saying oh this will never scale and they were right http 1.0 with apache or whatever ncsa web server ncsa httpd that wouldn't have scaled to the current scale of the web but somehow through the starting off with that janky technology through a very bumpy path we got to where we are today where the web is as they say web scale and it's not because the people who originally made it had the forethought to make something that was scalable because they didn't They just made something that they thought worked or whatever.

1:50:50And it had to be revised and it had to be improved. And lots of other third parties had to come up with ideas. And, you know, a company funded by something else entirely, which is Google figured out how to search the web and made bazillions of dollars, has now contributed to the advancement of the web by, you know, coming up with the new versions of HTTP that are in turn more efficient and making browsers that are efficient. And then like activity pub looks way more like the early web than blue sky does. Now blue sky is like, well, why, why do we repeat the same mistakes? The people making blue sky understand how to scale.

1:51:24And they have some interesting ideas about how this could be done well. But the thing about interesting ideas is they're interesting because like, Oh, I don't think anyone's tried that before. And that's kind of a danger zone where it's like, you know, it's the worst is better thing. We've done the thing where you start off with something simple that an unscalable and figure out how to make it bigger. uh the ones where you start out with something that is novel and interesting uh and scale better than anything came before it it happens it can be done but even in the best case scenarios which i'm thinking of this because part of what they're doing with the the data model and everything is a little bit like git not really but like they use some of the same terminology as in terms of having a repository the approvably correct repository with transactions committed to it you squint it might look like a database might look like git or whatever git is a good example it is a very interesting model for dealing with change management and fantastically successful partially because of its pedigree kind of like perhaps blue sky maybe one day but git is not universally better than all the centralized services that came before it git does not scale as easily as some of the centralized ones due to the way it's designed if you have a single repository with millions and millions of files doing anything and git all of a sudden starts slowing down and the answer to that from git perspective is don't do that don't have a repository with billions and millions of files have fewer smaller repositories use git modules do this do that do the other thing it's not to say that git is bad it's just that like as novel and interesting as git was and it brought all this value with with the with you know again the fact that it was novel and it did have features other things didn't have it's not universally better so i don't know how blue sky is going to turn out there again they're not done making blue sky they're still working on the basics of the protocol they just getting around to figure out how blocking works for crying out loud like they're in very early days and i do like a lot of the ideas they have in particular the fact that when you change instances you bring your posts with you although the way they do that seems a little janky but anyway uh i applaud the efforts in these areas but i do right now i kind of feel like activity pub for all its warts is following a path that looks familiar to me that where you start out primitive and not great and having scaling problems and you eventually figure them out and blue sky i the reason i signed up for it like when you know when they first came out it was like these people have thought hard about the problems these people are not just coming in blind and saying i'll just get something simple working and i'm sure it'll all work out they're not doing that they're saying we know here are the hard problems to solve and we have some ideas about how to solve them and we're going to come in and we're going to start off on third base and i i was like yes these guys get it they're gonna do it i'm rooting for them i just don't actually know if they're gonna do it and so it seems it seems like a higher risk factor where he seems unfamiliar and the fact that jack dorsey's involved doesn't really make me that excited anyway i do appreciate his millions of dollars i don't appreciate his personal involvement in any way the good thing is he's been distracted and has gone off it to deal with nostr noster he's involved with that too yeah what is that oh yeah he's he's given up on not he's given up on blue sky but he's that's his new darling is that oh god that i know even less about and it seems like their goals are different when i read the at protocol thing i felt like yeah they're thinking about the hard problems when i read about the the noster stuff i was like you're thinking about different problems and i'm not really interested in i'm not really interested in what you're doing even if you succeed it doesn't sound cool whereas i read the at protocol thing i said if you succeed yeah that'll be a thing i'd want to use so we'll see how this shakes out i don't need another app yeah yeah i'm gonna check four places i did sign up for nostr twice oh my word accidentally because that's how it works it's confusing anyway uh more updates on nostr in 2045 stay tuned yeah i just to me like the those scaling questions that's the biggest thing it's like who knows how this is going to end up but like what what blue sky is going through now this is the easy fun phase this is like you have only the cool people you have you have like you know a small number of people you don't have a lot of scaling challenges you're spending someone else's money you're not thinking about how to make money later like that's that's the stage there and now it gets so much worse after this like it's you know you tell people like oh it gets better like with some different kind of things when it comes to running a social network it gets worse everything just gets worse over time running a social network it's what do you even know about that marco so yeah they're gonna have to start facing these tough problems and the like the the the new cool darling phase we're in now is going to end and they're gonna face how how do you pay for this what does that do to the openness and the protocol and the client and everything else they're going to face how do we deal with nazis and blocking and everything else if they're just starting to deal with that now they're going that's going to quickly ramp up the fun part will be how do you deal with technical scaling because that at least is a little less controversial but the the harder problems of how do you pay for this and how do you handle moderation i don't see any evidence that you know as smart as blue sky is on their technical side I don't see any evidence that they're going to be that they're going to have an easier time solving those other problems than any other big social network has in the past I do think it is it is where a lot of people are and it's where a lot of people are going and it's pulling people off of Twitter who never made it to Mastodon it was delightful as I was going through and following people on Blue Sky rating other people's follower lists I was finding people that I've missed these last few months that I haven't been on Twitter because they never came to Mastodon like they they just they've just been on twitter and so you know macedon didn't bring everyone over who i wanted to follow so i'm i'm happy that this that this other place is attracting them so that i can see their stuff and i can keep contact with them and keep following them and have them keep following me without having to go back to twitter where i really don't want to go so i like that part of it but again like the the challenges are have yet to come and they're you know the the technical side is one thing but the the messy problems are not technical and so far we have no idea where blue sky is going to land on that mastodon does an okay job of that stuff not even a great job blue sky is a question mark so we'll see Anyway, thanks to our sponsors this week, Steam Clock and Backblaze.

1:58:07And thanks to our members who support us directly, you can join us at atp.fm slash join. And we will talk to you next week.

1:58:38And you can find the show notes at atp.fm And if you're into Twitter, you can follow them At C-A-S-E-Y-L-I-S-S So that's Casey Liss, M-A-R-C-O, A-R-M, N-T, Marco, Armin, S-I-R-A-C, USA, Syracuse. It's accidental. Accidental. They didn't mean to. Accidental. Accidental. Tech Podcast, so long. I've been fighting SwiftUI so much. It's been good, though. it's been like i talked about a little bit under the radar this week like my i was listening to that earlier today like my my main pattern has been i'll run into a problem i will explode my solution in complexity trying to solve the problem you then i will find a really simple way around the problem and i'll delete all the complexity and so i'm left with like a one-liner so you know some modifier or some some weird little technique or oh this state should have been a binding you know like something like that and and i'll fix the problem it'll be glorious i ran into this earlier this week um i was working on tapping on like a person's picture or a movie poster what have you and zooming in and taking up basically the whole screen and i was like oh god i'm gonna have to do like geometry readers up and down and inside now which is it might there's probably one or two geometry readers in call sheet right now, but there's very, very few.

2:00:19There's not a lot of them. Hey, for the record, I know there's a lot of people who are like geometry readers, make it hacky or whatever. I think geometry readers are fine. There are ways to use them very easily, very well, very cleanly. They are totally fine. I do not hesitate to use a geometry reader if I have to. I would hesitate a little bit. It's not because they're bad. It's just that very often when you when you use a geometry reader there's a cleaner way to do it without one sometimes there's not sometimes you got to use it that's why it's there it's there for you to use it right so don't be afraid to use it when you have to use it but not i'm going to say in every instance but in many many of the instances that i started out using geometry reader i came back later and realized oh there's actually a way to do this without geometry readers that is cleaner sure and that that feels too good to get rid of them and the problem is because well this actually gets into something that i've been thinking about when i hear you talking about swift ui and you casey as well uh with the complaints about it geometry reader can be thought of as a lower level of abstraction than the rest of swift ui because it's like look when you got to drop down a level uh here you can get at the raw geometry uh and it's kind of a weird clunky way to do it where this thing exists and its whole point is like i just i'm not a thing i just exist to extract these numbers for you that you're gonna find really handy so use them for something and you do and it may solve your problem right i what i found is that a lot of the difficulty in swift ui when people say oh if it does what you want you're fine but if you want to do something fancier you have a problem it's because swift ui doesn't have enough layers like it's not built on everything that's cool in swift ui is not built on slightly less cool pieces that are built on slightly less cool like you know the sort of the layer of abstraction a good api or a mature API will say, here you can do the easy thing.

2:02:09Drop down one level. Here is another whole world that is also really nice, has public APIs, is well thought out, but the pieces are just a little bit lower level. And guess what? Every one of those pieces, it was built on other pieces that are also really nice, also with a really nice public API. In fact, you can write your whole application on this level if you wanted. And it's all beautiful and it's just documented and it works in a sensible way and like that all the way down. That makes it so easy to do whatever you want because if you're doing everything at the very top level it's beautiful and if you want to do one custom thing peek underneath the covers and there is another layer underneath that that is also beautiful it's not underneath there is pipes and wires and like sparks are flying everywhere and everything's made of oatmeal that's not what you want like oh i gotta start swizzling symbols and i'm and i'm like i'm you know doing stuff with uh with unsafe pointers and like like that's not what we're talking about we're talking about when you open up one box inside is another beautifully wrapped box with documentation and tests and everything right and swift ui is not that and it's not because swift ui is the model is wrong or it's inherently wrong it's just that that's not like sometimes there are levels of pieces you can see piece x of swift ui is built on pieces y and z but very often it's like this is the top level thing and underneath it is no user serviceable parts as the the saying used to go on the stairs do not do not look and and it's and i feel that so much when i'm using it especially since i spend a lot of time at app kit as well which does have user serviceable parts underneath lots of stuff you know if you go all the way down there's like a hierarchy ns view is a class you can use and understand and was designed and is documented for humans to use and it underlies so much stuff so it's like i hope i never have to drop down to ns view i hope i can use it at this level you know i wanted to use nstable view and that's whatever that i don't know what the collection view is called in ns land but there's like five layers and every single one of those layers is a public api that is supported that you can use and you just need to drop down until you can solve your problem whereas swift ui is so often one layer and there's no reason it has to be swift ui camp and geometry reader is an example of that it's not that everything is built on geometry reader but it's kind of like they made a fake layer it's like well underneath the scenes geometry reader isn't what we're building on but you know what geometry reader is built on that's underneath all these other pieces but you can't see that part and so like i started off saying you should feel kind of bad for using geometry reader because there may be a simpler way to do it which is true but the main reason i'm saying that is because geometry reader is not really the next layer down in abstraction geometry reader is just like them poking up a little thing as like an apology of like no user serviceable parts but if you just want the dimensions this thing we'll give them to you good luck and then you run away so i i i don't know how best to phrase that i think i used to have a better term for this i used to talk about all the time when i would make frameworks in my jobby jobs is that you have to have you have to be built on layers and layers and each one of those layers is a perfectly serviceable top layer like that's the beauty of it whether you start from top down or bottom up it can't just be one amorphous blob and you can't say this is the interface and everything else is internals great frameworks great apis are you can use them at whatever level you want to use them and every level is designed every level is thought out every level is documented and that makes for an experience that everybody loves and i feel like swift ui can get there as long as the people making it have that philosophy because right now it seems like people making swift ui have the philosophy of like oh if you're having that problem just tell me and i'll add another parameter to the top level that is not a scalable solution we appreciate it sometimes you should sometimes the top level api is is too poor it is too it doesn't have enough features it should have another parameter or there should be another method but sometimes the solution is stop adding stuff to the top level show me what's underneath it oh there's nothing underneath it you need to build something underneath it like whatever the the next it's still swift ui it's just like just like you know ns view is part of app kit and so is nstableview they are part of the same family but they are different levels in the hierarchy swift ui needs that as well yeah it's just funny because you know the thing i was running into is you know this zooming stuff that we got sidetracked from and and i thought okay well we need to use geometry readers and figure out where i am in the screen and then i have to like do some crazy math in order to you know move the frame of that view and blah blah blah and then by sheer circumstance, I stumbled on a matched geometry effect, which is a view modifier that basically you put this on two different views and then say, okay, I want to go from one view to the other and magic happens.

2:06:53And it's not perfect, but it's good enough. And the conversation that you guys were having on Under the Radar was very, very relevant because I know, and if you look closely, and I probably shouldn't say this out loud because other people notice now, but I know that the animation for the zooming of the images is not 100 % correct. But you know what? It's close enough. And I have other things I should be working on than making that pixel perfect. I want it to be pixel perfect. I will hopefully one day make it pixel perfect. It's been in the back of my head, you know, underscore talks about this, you know, it's just like my subconscious brain has been chewing on it a lot over the last 48 hours trying to figure out how to make it pixel perfect.

2:07:35But ultimately, now is not the time to make it pixel perfect. It is serviceable as it is, and that's better than nothing. And that's what I plan to ship, at least for now. But yeah, this week's Under the Radar, they're all good. But this week's certainly struck home for me as I'm trying to figure out, okay, what juice is worth the squeeze and what isn't. So how close are you, I think, to shipping 1.0? So I was saying to you privately before the show, and I think it's very true, I need underscores, sailboat emoji, no new features, sailboat emoji, you know, his little sheet that he'll put up from time to time.

2:08:12He'll probably send you a PDF if you ask him. He probably will. In fact, knowing him, it'll probably be in my inbox by the time I wake up. But no, so I have 13 issues in GitHub at the moment, but I have labeled all but two of them as not right now. You know, here's a list of things I want to do eventually, but not right now. And there's only two things left for right now. One of them is to see about improving the way I do layouts such that there's some very, very small shifts as images are loaded and things of that nature. And the UI will shift around a little bit. It's a little bit annoying, but not devastating.

2:08:54And I'd like to at least spend a few minutes to see if I can improve that. and then the other thing I need to do is properly implement my paywall, which is to say I'm going to do the whole dance of, you know, you get some number of searches for free. I haven't figured out what that number is yet. And then subsequent to that, you're going to have to pay in order to do anything else. And so I was looking into, what is it? Oh, NS Ubiquitous Key Value Store, which I've never used before. I'm using CloudKit for stuff in this app, and I've used a teeny bit of CloudKit in Masquerade. And I'm using a fair bit of CloudKit in this app, but I've never used the ubiquitous key value store.

2:09:33And I was thinking about using that to store basically how many searches you've done. It's so much better than any other iCloud thing. I actually quite like the CloudKit fake database thing. I've written a pretty good facade in front of that that I'm pretty happy with. But I've heard horror stories about some of the stuff that preceded it, Like core data for iCloud, I've heard, was a nightmare from the moment it launched and is still a nightmare today. Yeah. The key value store was one of the original three iCloud methods. It was core data, that, and I believe the document-based. You mean CloudKit, not iCloud, right?

2:10:09No, I mean iCloud. When iCloud first launched as an API, it was three things. It was iCloud core data, which was a disaster and is, I believe, totally discontinued. there was also this key value store and i believe a document uh thing this is the only thing that has survived right but but to to john's point like the thing i'm talking about the thing i just said i have a facade in front of that is definitely that's cloud kit yep yep yep um but but i've never used the key value store before and so i need to take a quick look at that it doesn't look to be particularly complicated it looks to be pretty straightforward knock on wood it's really simple it it's very similar to just using ns user defaults or user defaults like it's it's it's very very similar to that and you'll see it's that you have to do like you have to like watch for like some notification to when you when you get new data that's about it it's you know it's very simple very old and i have found that's what i use for overcast's uh login yeah account thing and it's been fine like i've i've never had a problem attributable to that yeah and i don't i could be wrong i don't think that requires the user to have icloud drive enabled like remember i i hit that problem with when i was looking at cloud kit where cloud kit cloud kit's database and everything does not work for users who have icloud drive disabled that's where it stores data as far as i know the key value store does not have that restriction i don't think and i should probably test that but i think anybody with an icloud account at all signed into the device i think can do that um the the downside is that you are limited in terms of like there's it only allows a certain number of records and they have to each be a certain size or smaller but for what you're talking about you it'd be totally fine yep and so that so i need to do that so the two items are you know take away the shiftiness that's with an f take away the shiftiness and uh set a total limit of free lookups and lookups would you say yeah that's right yeah is it a total is it a total per day per week or total all time uh honestly i hadn't thought too much about it but my current thinking is all time so you'll get and again i don't know what the number would be but you'll get 10 searches or something like that maybe it'll be five maybe it'll be 50 i don't know i'm kind of in favor of per unit time but we'll see have you decided uh what price you're gonna charge for the premium uh no i don't i don't know it's all over the place right because the feedback i've gotten which has generally been very very good uh but some people are like oh my god take my money, I'll give you infinite money for this, which of course, you know, is being, they're being facetious.

2:12:38But the point is, you know, they would potentially pay like$10 a month, which I think is too much. And I've heard people say, you know, I'll pay one or$2 a month. And for a month, I think that's completely reasonable. For a year, that's not enough. So I'm thinking, as we discussed when we first had the conversation about what is now Call Sheet, I'm currently thinking somewhere to the order of$8 a year. And the other thing I am having really complicated thoughts about is whether or not I should offer a lifetime unlock for something to the order of like 40 or 50 bucks. And no, don't, well, don't do anything lifetime.

2:13:18Trust me. You will regret it. That's the thing is I don't know. And I was talking to our friend, Ryan Jones, and he had extremely strong feelings about how I should do the lifetime thing. And he made some pretty compelling arguments, which I'm not in a position to try to verbalize here in the heat of the moment. But suffice to say, his arguments were very compelling. So I'm not sure what I'm going to do. But here's what happens. Your lifetime customers will be the fewest. You'll make the least money overall in total from them. And they will be your biggest pains in the butt because the entitlement they will have makes it not worth it.

2:13:54Trust me, you do not want those customers and you won't get many anyway so it doesn't matter focus on your subscription price that everybody else will pay that's that's your product i think a dollar a month or up to ten dollars a year i think you could you could easily do yeah so i mean again this is all current thinking but knock on wood if i can't you know i'm only gonna i'm gonna time box which is a very corporate thing to say but i'm gonna time box the taking away the shiftiness uh to to like you know a day or something like that you know just give myself a little time to see if I can improve it, but if I can't, oh well.

2:14:26And then I think the task for the remainder of this week, what little is left, and next week is, you know, getting the paywall squared away. I need to rejigger the actual paywall screen itself because it looks like garbage right now. I need to rejigger that. I need to actually, you know, track, you know, how many searches you've done and so on and then put up a paywall when you've hit your limit. But hypothetically, you know as as early as about you know a week week and a half from now i might be pushing the first build to to you know app store review hopefully maybe a famous last words you got a factor in time to get rejected because your paywall screen doesn't comply with whatever that's true and this is my first time doing subscriptions so surely i'll get some sort of issue with that um who knows but one way or another that that is that is the hope so given that I told you, I think it'll be a week to week and a half.

2:15:20That means it'll be about three to four weeks before I push it to the app store. Because you always multiply by two is my lesson I learned over my many years in corporate America. Whatever I think it'll be, it'll be twice as long.