#dev 2021-12-16
2021-12-16 UTC
gerben, Seb[d], [tantek], Kaja, corenominal[d], kloenk, capjamesg, doosboox, willnorris, [chrisaldrich], Seirdy, IWSlackGateway, [fluffy], [schmarty], darkkirb, lanodan, rrix, srushe, EvanBoehs[m], binyamin[m], HarryCoburn[m], Charlotteracracs, KartikPrabhu, alex11 and BinarySavior joined the channel
# 
micahrl[m] Jumping off what ya'll are talking about with "plumbing", I was wondering recently if the deployment and hosting of sites is something that is part of the indieweb discussion, or if that is kind of left intentionally out of scope... like is it considered more desirable to have something that runs on a commoditized linux server than something that runs on a proprietary cloud API like say Netlify Functions?
# 
aaronpk looots of notes here https://indieweb.org/web_hosting
# micahrl[m] Not a particular site, but indieweb connectivity software in general... one of my projects I hope to be useful to others at some point, and currently it's written for a Linux server because I already have one. But I am wondering how to balance the competing interests of being independent of a corporation which might die or change its product offering, and of not having to maintain your own Linux server.
# micahrl[m] I'll read that page, ty
# micahrl[m] Yeah, that's one way to describe it. Like if something goes wrong with your current host, how easily can you host elsewhere.
# micahrl[m] I guess one way to tackle it is to write something that could be deployed in either scenario, but of course that is more work :) and more testing etc
# 
Loqi migration in the context of the indieweb refers to the process of moving your indieweb site from any one or more of one CMS / web host / DNS provider / URL design / domain name to another https://indieweb.org/migration
# Loqi ok, I added "[[migration]]" to the "See Also" section of /web_hosting https://indieweb.org/wiki/index.php?diff=78400&oldid=77189
edburns[d], KartikPrabhu, akevinhuang2, Eddy04[d], [jeremycherfas]1, [jeremyfelt]1, [tantek]1, [aciccarello]1, [tw2113_Slack_]1, ben_thatmustbeme, PapaWonka[d] and [Sam_Butler] joined the channel
# [Sam_Butler] This is quite a bit different from what I first asked about a couple days ago, and I'm not entirely sure how useful it is. In any case, if it interests you, glad to collaborate and explore possibilities together! Also for a contextual note — this is not suggesting any kind of replacement to a Signal/Matrix. Those very well may be more useful for day-to-day chats, and I don't think there's any need to try and replicate
# [Sam_Butler] Signal/Matrix on E2EE and security. The specific use case here, is when you'd like to have a chat that is public and discoverable and comes up in search results — so specifically, encryption/security/etc may be out-of-scope. The relevant place where security/privacy could come up, is in terms of preventing people from scraping chat data for sentiment analysis and to build models off of — and in that case, perhaps the solut
# [Sam_Butler] could just be mechanisms to increase the friction for automated behaviors. On the other hand, there could be times when people could want to scrape their own chat data and analyze it — for those contexts, maybe it could be a matter of informed consent and transparency? Gets into data sovereignty territory as well ... who owns the data of public facing chats, and what should be the permissions of what can be done with it? If
# [Sam_Butler] people have any references for other projects exploring those questions, would be glad to read up!
jjuran, benji and Seirdy joined the channel
# 
[KevinMarks] Have you looked at the https://chat.indieweb.org/ implementation of this conversation (you're in the slack version of it). See how that is marked up with microformats, and that it is logged to github for persistence too.
tetov-irc, kogepan and [jgmac1106] joined the channel
# IntriguedWow[d] Nice!
KartikPrabhu joined the channel
# 
capjamesg[d] An interesting article on web standards.
# capjamesg[d] "The total word count of the W3C specification catalogue is 114 million words at the time of writing. If you added the combined word counts of the C11, C++17, UEFI, USB 3.2, and POSIX specifications, all 8,754 published RFCs, and the combined word counts of everything on Wikipedia’s list of longest novels, you would be 12 million words short of the W3C specifications."
# capjamesg[d] (I don't agree with the article. It's just an interesting take.)
# capjamesg[d] (especially because engines like chromium exist upon which one could reasonably build something with the right resources
akevinhuang, KartikPrabhu, lanodan and kogepan joined the channel
# [tantek]1 yeah, W3C specs could be better written, no argument there
# [tantek]1 (and tbh a lot of them really shouldn't be specs)
# [KevinMarks] If you applied mf2/indieweb interop rules to them, how many would remain?
# [tantek]1 almost none of the non-browser specs
# [tantek]1 maybe only the ones from the Social Web WG 😇
jjuran and astralbijection[ joined the channel
# [tantek]1 [benatwork] if you ever find a desire to make a webmention handling service of any kind, you should call it Werdmention
# [tantek]1 and a site for hosting webmention based groups: Werdburg
# [tantek]1 on a more (semi-)serious topic, has anyone had a need for more "coarse" maps for location-based archives? (e.g. checkins or posts with location info) ?
# [tantek]1 I feel like GWG was asking about that the other day and I have some ideas now
# [Sam_Butler] [KevinMarks] thank you for the input! Can you point to where the IRC logs are backed up on GitHub? Has there ever been an idea to make them indexable/discoverable via search engine? Also, what is the difficulty of setting up an IRC channel like that with microformat markup and logging to github?
# [snarfed] [Sam_Butler] details on https://indieweb.org/discuss, eg web UI code https://indieweb.org/discuss#Logs, search https://indieweb.org/discuss#Search
# [tantek]1 "fun" 😂
# [snarfed] aaronpk really? I haven't noticed that. eg https://chat.indieweb.org/dev/2021-12-15#t1639583007026800 , https://github.com/snarfed/bridgy/issues/609
KartikPrabhu joined the channel
# Loqi [[snarfed]] petermolnar it's been a while 😐 https://github.com/snarfed/bridgy/issues/609
# aaronpk cause it does include the wiki edit message as the git commit message https://github.com/indieweb/wiki/commits/main
# [tantek]1 I think I've triggered that before
# [tantek]1 This is an example of multiple scales of a hex map that could be used for various archive page display purposes, e.g. with the "zoomed in state" being the one that you were in the most during a particular month: https://indieweb.org/maps#US_states_and_CA_counties
# [tantek]1 curious if that sort of summary 2D presentation/aggregation/linking to posts is appealing to anyone else for their own archive pages
# [tantek]1 GWG ^
# [Sam_Butler] Is there any documentation of how to configure an IRC client in the IndieWeb way — bridged to Slack/Discord, and with the automatic backups to GitHub?
# [Sam_Butler] Got it, makes sense. Thanks for the input as always.
# [tantek]1 modularity++
# capjamesg[d] modularity++
barryf[d] joined the channel
# 
petermolnar [snarfed]: my problem was not with brid.gy; it was webmention.io. Because the source only contains https://href.li/TARGETURL the parser on webmention.io says the TARGETURL is not found on the source page, and, unfortunately, it's correct.
# petermolnar so far I couldn't find any working solution; I tried altering the themes on tumblr, to no avail.
# petermolnar source username.tumblr.com/post/id, which contains href.li/ORIGINALPOSTURL, and the webmention is sent source=username.tumblr.com/post/id,target=ORIGINALPOSTURL to the webmention endpoint of ORIGINALPOSTURL
# petermolnar me. I'm sending that. Because I wasn't aware tumblr rewrites outgoing urls in the source html of the post.
# petermolnar so you insert the url anywhere in the theme, including the HTML description, and tumblr nicely rewrites it at HTML level.
# petermolnar so it made me think if a webmention receiver, when looking for the source url in the code, should or should not take the bloody shorteners and silo rewrites into account when looking for the original url
# petermolnar > send the target URL of the href.li link - that's the source url.
# petermolnar let me prepare an example
# petermolnar the tumblr post: https://pmlnr.tumblr.com/post/670768358907658240/lanzarote-view
# petermolnar I want the tumblr post to ping my post, so I can fill it in as syndication link. The problem is that if you look for 'https://petermolnar.net/photo/lanzarote-view/' in the tumblr post's source, all you'll find is 'https://href.li/?https://petermolnar.net/photo/lanzarote-view/ <https://href.li/?https://petermolnar.net/photo/lanzarote-view/>'
# petermolnar (ignore the horrible tumblr theme, please)
# Loqi [Peter Molnar] A view of Lanzarote https://petermolnar.net/photo/lanzarote-view/lanzarote-view_huge.jpg
# Loqi [Peter Molnar] A view of Lanzarote https://petermolnar.net/photo/lanzarote-view/lanzarote-view_huge.jpg
# petermolnar oh
# petermolnar the only trouble is that while it shows up in my dashboard, I need to go to https://webmention.io/api/mentions?domain=href.li&token=TOKEN instead of https://webmention.io/api/mentions?domain=petermolnar.net&token=TOKEN to fetch it, is that expected?
# petermolnar I see.
KartikPrabhu and tetov-irc joined the channel