#@VicVijayakumar1995: PHP is dead, learn ColdFusion 2002: PHP is dead, learn ASP.net 2003: PHP is dead, learn Django 2004: PHP is dead, learn Ruby on Rails 2010: PHP is dead, learn Flask 2011: PHP is dead, learn AngularJS 2016: PHP is dead, learn Next.js 2022: okay this is awkward (twitter.com/_/status/1587462991955591168)
[KevinMarks]1, mro_ and pharalia joined the channel
#Xe[schmarty]: I understand that it was a stressful time but like holy crap as a user that did not go well in my eyes
mro, saptaks, omz13, gxt, chenghiz_ and [campegg] joined the channel
#[schmarty]Xe: i very much agree and very much appreciate you putting it into words. not to mention your extremely good post with concrete steps for getting prepped for the release! thank you for both!!
#Xe[schmarty]: no problem! I've been thinking about doing a "SRE 101" online class and most of it would have things like that
geoffo joined the channel
#Xehow to prepare for these kinds of vulnerabilities, writing down instructions on how to deploy custom software, etc
#Xekeeping a work log and a weekly/monthly summary of what you got done so you can use that when review time comes around
#Loqi[Xe] I want to make 2023 be the year that I work on and publish that online course about #sre/#devops essentials. Going to have a lot of philosophy, useful advice for navigating corporate structures, how to avoid "I told you so"s, handle catastrophic vuln...
#barnabysounds like an interesting course, I’d definitely be interested in reading it. Never published anthing similar myself but video+blog posts on your site after some period of being only available to patrons sounds like a good model to me
mro and pharalia joined the channel
#[schmarty]Xe: haha, o no. exciting, though! i have recently enjoyed Chris Ferdinandi's courses at gomakethings.com. He rolled his own platform that is a blend of video (private vimeo in his case) and sort-of-like-transcripts versions in HTML, ebook formats, etc.
#[schmarty]i forget where i read this but if i recall correctly the site structured so all the public stuff is static HTML and all the requires-payment stuff is fetched through an auth wall and cached via serviceworker.
#[Murray]Heh, weird community crossovers 😄 Chris was talking about that earlier today on the Frontend Horse Discord, as to whether it might be better using fully static (SSG) HTML and redirects, but yeah that's how I understood it to work currently 🙂
#[manton][tantek] Delayed reply… Thanks for that 404 heads-up, I looked into it and Hover reorganized their help pages… I found one that seems pretty much the same and updated the link to it.
[manton]1, Zagura[m] and jacky joined the channel
#barnabyZagura[m]: regarding webmentions vs activitypub posts, they’re technically quite different even if they can be used to create similar UX and interactions (replying, liking etc)
#barnabythe main difference is that (afaik) activitypub posts contain a copy of the activity or “object” they’re about, whereas a webmention is just a ping which tells one page that another page has linked to it
#barnabyso with webmentions, it’s up to the receiver to fetch the sending page and do something with it — for example parse it, see if there’s a valid reply there, and if so, display it under the original post
#jackyb/c even things like signing messages, embedding verbiage are all done out of band of the actual sending of the webmention (whereas you gotta do that on each message sent)
#barnabythe main trade-off is that webmentions are easier to send but a little more complicated to parse, whereas provided you can verify an activitypub message’s signature then you can trust that its data is valid
#barnabyhope that’s useful, feel free to ask if you have further questions!
#Zagura[m]idk how i found it, but how webmentions.io work then?
#Loqiwebmention.io is an open-source project and hosted service for receiving webmentions and pingbacks on behalf of your indieweb site https://indieweb.org/webmention.io
#barnaby^ that’s the tl;dr, it’s a hosted service which can do all the hard parts of receiving webmentions for you
#Zagura[m]hosting a dynamic website seems like a lot of work (& not free), so i was hoping it could work with a static site
#barnabyyou direct potential webmention senders there instead of to an endpoint on your site, and it parses+verifies the sender, and gives you a simple JSON API you can fetch to show responses
#barnabyyep, webmention.io is a good method for handling responses on a static site
#rubenwardyOk, so the docs led me to believe that I can add a h-card to my homepage and then reference that as rel=author from other pages. But the IndieWebify.me tool doesn't seem to be picking that up.
#@voxpelli↩️ Even more important: Do not use the same code base.
That was what killed StatusNet / Identica – the precursors to Mastodon which defined the OStatus protocol which Mastodon started out on.
Compare itto eg Webmention which has heaps of implementations: https://indieweb.org/Webmention (twitter.com/_/status/1587911238331248640)
#barnabyafaik indiewebify.me doesn’t always follow and parse external pages, it works mostly with the microformats markup on one particular page
#rubenwardyOk. Is it fine then, or should I duplicate the info? The getting started guide says:
#rubenwardy> Why? When you publish content, you can link back to your home page using rel-author and your authorship information can be retrieved from the h-card.
#[KevinMarks]1In general, mf2 parsers don't follow links to other pages (unless they're explicitly doing xfn style rel=me link graphs)
#barnabyyour homepage h-card looks fine to me, whether you want to duplicate the info is up to you — it mostly depends on what authorship info you want to make visible on the blog subdomain
#barnabya lot of people have a “full” profile h-card on their domain, and additionally have simple name+photo+url authorship h-cards on individual post or feed pages
#barnabybecause very often it makes sense to have that data human-readable anyway
#barnabyand that way, parsers which are looking at an individual post get basic author information for free, and can follow the link to parse a more detailed h-card if they want
#barnabyand if you don’t want to have a small author profile on indvidual post pages, then a rel=author will be sufficient for any implementations which follow the authorship algorithm
#barnabyjust be aware that it might not show up on indiewebify.me
#gRegoru- for URL properties, p- for plaintext, etc.
#barnabythe difference between u-author and p-author is extremely minimal if its nesting an embedded h-card, which should almost always be the case for the author property
#barnabyit changes how the value of the “value” key in the embedded h-card is parsed (as plain text or as a URL)
#barnabyyou can play around with some markup and see the parsed results in real time here to get a sense for exactly what happens https://waterpigs.co.uk/php-mf2/
#@hamishmckenzieThe problem isn’t that Elon Musk owns Twitter – it’s that you don’t. We shouldn’t have to freak out about a dictator taking over a vast digital empire. People should control platforms rather than the other way round. (twitter.com/_/status/1587849734181425152)
#[tantek][benatwork] why consider self-hosting a Mastodon instance instead of Withknown + Bridgy Fed?
#[tantek]and yes we're having this conversation in #indieweb-dev because no non-dev should be even remotely considering self-hosting a Mastodon instance
#[tantek]or even bothered with the noise about discussing the possibility thereof
#[tantek]I mean it's your discretionary hobby time if you feel like spending it on Mastodon /admintax
[manton] joined the channel
#AramZ-S[m]I think also... it's not really what Mastodon is for? I've been talking to a bunch of folks and generally what I'm starting to get a sense of is that the best reason to start up a Masto instance is because you have a community that can have a unified set of rules and build a useful and mutrually engaging Local. I don't think it makes sense from a project perspective to self-host Masto unless you've got a plan for it. Solo
#barnabyyeah, something like microblog.pub seems like the better alternative for single-user instances who want indieweb and AP support built in
#barnabymastodon doesn’t send webmentions natively, right? did anyone make a bridge to send them already?
#barnabyseems like all you’d need would be to set up an AP identity which people can follow, and whenever it receives a new post, it goes and sends the relevant webmentions for it