#dev 2022-08-15
2022-08-15 UTC
jacky, geoffo, angelo, willnorris, Loqi, jan6, jonnybarnes, thomas1, petermolnar, tetov-irc, mro, kloenk, nertzy, AramZS and chenghiz_ joined the channel
jacky joined the channel
# 
AramZ-S[m] I dunno I think they just don't know the framing
# AramZ-S[m] FlipBoard and other apps that are basically feed readers have immense popularity on the app marketplace
# AramZ-S[m] It's more of the same problem with social media which is that the platform inserting itself between the tech and the user has turned the underlying tech obscure.
# AramZ-S[m] I liked P2 a lot but it never really was picked up outside of hardcore WP devs. It would have been interesting to see an interation of WordPress where the authoring UI was impacted by the ideas P2 brought forward.
# AramZ-S[m] I mean, SmartNews is a huge biz
# AramZ-S[m] I say this as someone who doesn't particularly like blocks, but I think blocks made a lot of sense to the folks who are actively contributing to WordPress which are people who started out as indie self publishers but generally now work for enterprise publishers. Blocks is a solution that makes a lot of sense for large scale mainstream publishers and website owners who have the time, money, and interest in investing in the big cool
# AramZ-S[m] solutions one can build with them. But they are basically an annoyance for anyone using Wordpress as an individual blogger.
# AramZ-S[m] WordPress or not, almost every mainstream publisher engineer I have worked with does a lot of internal work around blocks or blocks-like functionality to satisfy particular goals around how more complex orgs want their pages to flow.
# AramZ-S[m] But like... if you just want to put words on a website it's really an annoyance.
# 
jeremycherfas AramZ-S[m]++
jacky joined the channel
# AramZ-S[m] The other thing to keep in mind is that WordPress Core took a hard turn into requiring a lot of really nuonced JS expertise for blocks, I was deep inside the community when it kicked off, and then suddenly realized that they didn't really have that among the regular contributers and it made the initial versions of the editing tool really rough while they skilled up, which didn't help with its popularity or getting buy in or feedback
# AramZ-S[m] from the larger community.
# AramZ-S[m] Like... I remember that when they started on Blocks they were just getting around to the idea that their JS needed unit tests. I learned QUnit inside WordPress core as part of contributing improvements to the bookmarklet and like... there was one other person who knew it.
# AramZ-S[m] Long way around of saying, I don't have a problem with Gutenberg, I just look at it as something that really isn't for me, but does have its uses for a lot of WordPress users.
# AramZ-S[m] And I think it's launch was prob not handled great
jacky, mlncn and jacky_ joined the channel
# 
@Honesty_Init ↩️ They do have some neat protocols. Webmention might be the better alternative to the Fediverse. And IndieAuth (OpenID substitute) gets way too little attention. But it's indeed a chicken and egg problem; apart from default integration. (twitter.com/_/status/1559205666245550080)
[schmarty] joined the channel
# 
[schmarty] goodwill is surprisingly flammable
[aciccarello] joined the channel
# AramZ-S[m] Yeah, I mean, the Core team did not message it well, shunted initial development off into a plugin group instead of integrating it well with the Core development flow and the accessability problem flows out of the JS one. Core team's whole approach to the new heavier and more advanced use of JS felt like ship it and we'll fix it later. And yeah, that and a variaty of reasons made it very badly messaged to the community. I think that
# AramZ-S[m] Gutenberg is still prob the worst handled feature WP has done, especially when it had the potential to have gone a lot different. Def not defending it. Personally I still don't like Gutenberg, but I do understand where it came from, which is the shift of the Core team and closest surrounding community becoming paid contributers through their employers, which happens with a lot of open source projects as they mature.
# AramZ-S[m] I think the big difference is that WordPress's shift towards sponsored contributers also was a much larger shift in focus and priorities than most open source projects experiance.
# 
Loqi A wish is a post indicating an acquisition or purchase either you wish to make or which you hope someone might purchase on your behalf for a birthday, anniversary, baby shower, wedding, or other event https://indieweb.org/wishlist
[bneil] joined the channel
[benatwork] joined the channel
# 
[benatwork] It didn't and I now think it's easier to build a new static file centric platform instead of retrofitting Known
# Loqi It looks like we don't have a page for "database connection lost" yet. Would you like to create it? (Or just say "database connection lost is ____", a sentence describing the term)
# 
[bneil] yeah, went down that rabbithole after https://0data.app/ - ended up looking at fusion. Its really neat but yeah, the specs look odd
# 
[tantek] database connection lost is /database-antipattern#DB_connection_loss
# [tantek] [bneil] lol when I scroll down that page I see an embedded Iframe with a security warning inside: "⚠ Did Not Connect: Potential Security Issue / Firefox detected a potential security threat and did not continue to play[.]rosano[.]ca because this website requires a secure connection. / Learn more…" [brackets added]
# [tantek] "Learn more..." links to https://support.mozilla.org/en-US/kb/error-codes-secure-websites?as=u&utm_source=inproduct
# [bneil] yeah, the apps page itself is a rabbithole like i was saying. A few apps lead to https://5apps.com/storage which was that reference to remoteStorage and that line of tech. Honestly I found it intriguing, i thought the concept of the semantic web had died out in 2000
# AramZ-S[m] Solid is still going strong, though I haven't played with it myself
# AramZ-S[m] I agree, the idea of using databases for blogging sites seems bad now. I wouldn't go down that route if I was creating a CMS intended for individual users now.
# AramZ-S[m] It's been interesting to see how the costs work at scale tho. Sometime flatfile dbs, nodbs etc... are less expensive, sometimes more so, depending on what you're doing with them, but it should never matter at the scale of an indieweb site maintained by an individual person.
# AramZ-S[m] [tantek] Yeah, that does make sense. I just understand that the place it was coming from for a lot of the participants who wanted it was basically that they were now working for corporate entities that desperetly needed that functionality. I do think it should have stayed in Plugin mode for significantly longer and the whole 'if you don't like it just install the Classic plugin' message was disengenious.
# AramZ-S[m] I dunno, I wonder a lot about how the alternative routes could have gone. Would some big players have dropped WordPress for their site? Would there have eventually been buy-in and feedback to shape it into something better?
# AramZ-S[m] I dunno about strong as a product, just still being developed sorry if that was unclear.
# AramZ-S[m] It's weird to think about how some of the blowback to Gutenberg came out of anti-JS sentiment, which was still strongly a thing at the time but is totally alien now.
jacky joined the channel
# AramZ-S[m] It's very weird to me to remember that there was a time when the idea of using JS as a developer in any way was considered a bad thing, to be avoided as much as possible, by a lot of people.
# AramZ-S[m] I generally agree, but I do see on the other side there's the people doing the biggest amount of contributions to WordPress Core basically having to decide between prioritizing a feature that allows them to continue to have their contributions funded and having to functionally abandon contributing for months on a side project to serve the needs of their employers. Capitalism ruins everything! :/
# AramZ-S[m] *to work on a
jacky and mro joined the channel
# AramZ-S[m] Yeah, I agree that JS was supposed to be for progreessive enhancement. But there def was a significant portion of web developers who were just generally anti-JS at one point.
angelo joined the channel
# 
[aciccarello] I work on enterprise JS web apps for work but for my personal website I try to only use it for progressive enhancement
# [aciccarello] That sometimes gets a little tricky though with static hosting.
# AramZ-S[m] I still don't like React so, pretty clear where I land lol
# [schmarty] i actually really like the RemoteStorage model. it's a way simiplified version of the "pods" idea. sign in is just OAuth and the reference PHP app is pretty easy to understand.
# [schmarty] we might be able to learn some things from it as we keep extending Micropub. for example queries for venues and /nicknames-cache
# [schmarty] it's like they took the "Solid" pods idea and really stripped it down. no webid, only oauth. no RDF, just loose JSON schemas.
mro joined the channel
# [schmarty] i have some really loose ideas about where quotation marks go, i guess 😂
jacky, mlncn, mro, bterry, nertzy and mro_ joined the channel
[campegg] and jacky joined the channel
jacky joined the channel
# AramZ-S[m] Yeah, Hyper has some similar concepts around this type of structure as well.
# 
jacky hyper like https://hypercore-protocol.org/? yeah
# [schmarty] i'm passingly familiar with both hyper and solid pods. i don't immediately see what concepts they share in this context.
jacky joined the channel
# AramZ-S[m] In the sense that they both build on 'local' file structures that connect with each other and other systems. The idea of own your own 'data' and have it useful in other contexts is at the core of both projects, even if they take pretty different approaches in terms of what 'own', 'data' and 'share' means. I think of them as addressing similar philosohpic questions.
# [schmarty] that's an interesting look at it. i see them as very different in that pods are meant to centralize data into the hands of the user and hyper (or at least the older dat ecosystem) are more built around sharing verifiably published data.
jacky joined the channel
# AramZ-S[m] I mean, whatever you use it for, I do think the intent of Hyper is to make it possible to centralize data into the hands of the user. One of the cooler use cases is having a common folder structure that you can use to make specific data accessible to a site that requests it, things like username, login, etc... or the way that you can build social networks basically byp peering addresses of personal versions of that system. If you
# AramZ-S[m] have a database of tweets on your system, peering with others gives you Twitter and in that respect I do see it a lot like Pods. I have used Hyper and I've only read about Pods so I might be wrong there. I think on the wider scope the eventual ends to which Pods can be put vs Hyper is pretty different, but I do think they're trying to give very similar answers on the question of how the future of the web might work.
nertzy joined the channel
# IWDiscordGateway <capjamesg> [snarfed] have you decided yet about doing another IndieMap crawl?
gxt, bterry, AramZS, jacky and tetov-irc joined the channel
# [tantek] owning a blob of opaque JSON (or XML) data is pretty much useless in practice, which is why the Solid approach is standardizing perhaps one of the least interesting layers (far too low) to be useful to anyone except for devs that have time to tinker with (reverse-engineer) proprietary JSON/XML formats and write translation / import / export code - which y'all do right?
# [tantek] the point of having a standard protocol is that implementations can implement it, pass a test suite, without ever having to talk with any other implementers, and things *just work* between implementations. no need to collaborate, or engage in MxN conversations with every other implementer
jacky and tbbrown joined the channel
# [tantek] Markdown << Criticism: the fact that Markdown actually has its own file extension .md rather than "just" using plain text (and .txt) is a strong clue that it has in practice readily diverged from its original "key design goal" of "_readability_, that the language be readable as-is, without looking like it has been marked up with tags or formatting instructions" (from https://en.wikipedia.org/wiki/Markdown#History) so much so that
# Loqi ok, I added "Criticism: the fact that Markdown actually has its own file extension .md rather than "just" using plain text (and .txt) is a strong clue that it has in practice readily diverged from its original "key design goal" of "_readability_, that the language be readable as-is, without looking like it has been marked up with tags or formatting instructions" (from https://en.wikipedia.org/wiki/Markdown#History) so much so that" to the "See Also" section of /markdown https://indieweb.org/wiki/index.php?diff=82956&oldid=75462
jacky joined the channel
# [schmarty] Markdown++
# [schmarty] i like it at a niiiiiice flat 0.
# [schmarty] (i use the heck out of some markdown but agree it gets pretty wack.)
# [schmarty] it's like the poster child for worse-is-better-and-also-sometimes-worse-is-worse.