#tantekrel-subscribe is an experimental rel value for linking from your home page to your subscription endpoint, and is currently prototyped by {{aaronpk}} on aaronparecki.com; try the Follow button in the right side of the footer of his home page or any permalink.
#aaronpkhmm, is it worth making a little web interface that provides the follow button and the follow form? to make it easier for people to do that part
#aaronpkoh, one argument for having the rel value be the base URL and hardcoding this parameter name is to be able to add additional parameters for extensions. things like a redirect_uri, or if we need a CSRF "state" parameter or something
#KartikPrabhucan't "follow" be handled like a webmention POST request?
#KartikPrabhudon't we do that with webmention though?
#aaronpkthe redirect here is important because it gives you the user a chance to preview the follow request, and you get to use your own authentication mechanism at your reader
#aaronpkwebmention has a dialback verification step where the source URL is checked
#KartikPrabhuso does this where the software at reader.example.com asks you to verify that you want to subscribe
#aaronpkright, which is why you can't do it with a post request
#Loqifollow is a common button in silo UIs (like Twitter) that adds updates from that profile (typically a person) to the stream shown in an integrated reader, and sometimes creates a follow post either in the follower's stream ("… followed …" or "… is following …") thus visible to their followers, and/or in the notifications of the user being followed ("… followed you") https://indieweb.org/follow
#aaronpkthe rel value they use is "http://ostatus.org/schema/1.0/subscribe" but that URL has never had content on it
#aaronpki still don't understand why people use URLs for rel values and then don't put docs there!
#tantekindeed, I think we discussed that yesterday and I pointed out that was a good example of the fragility inherent in using URLs for propery names (i.e. the primary assumption that RDF is built on)
#[chrisaldrich]I like this idea, particularly the ability to micropub a potential follow post. How does/could it work alongside of something like https://www.subtome.com?
#aaronpkit's a very different approach from SubToMe, but accomplishes the same goal
#[chrisaldrich]I think in an indieweb world of a dozen or more post types that one should help the reader find/subscribe to one or more of one's available feeds. (i.e. photos, locations, notes, articles, everything......). I don't presume that everyone will want my firehose.
#tantekaaronpk: I agree a publisher could offer different feeds, however I'm not sure that a follower knows enough at that moment in time to make that decision
#tantekit's additional cognitive load to consider such options
#aaronpkso if you're on my checkins page, the options are "checkins", "home page", "all posts"
#aaronpkbut that's also why i like this approach because there's room for UI experiments without protocol needs
#[chrisaldrich]I remember publishers could add links in their headers for explicit feeds <link rel="alternate" type="application/rss+xml" title="Blog Feed" href="http://example.com/feed/" /> that most RSS readers respected (or at least looked for), but the idea of post discovery is certainly much simpler.
#tantekchrisaldrich and that publisher option never resulted in a decent RSS reader UI for picking
#[chrisaldrich]It was one of my favorite parts of woodwind....
#tantekno big surprise frankly, the folks doing RSS format/protocol design back in that day didn't iterate with user tests etc. on this stuff with actual prototypes
#tantekit's less work for the publisher to be lazy about that up front also
#aaronpkthis is all a very theoretical problem until there are more compatible readers anyway
#tantekso why do extra work as a publisher that may never get used, and worse, may just confuse / cognitively load the user ?
#[chrisaldrich]tantek: it's one of my favorite things about people with wordpress blogs -- as a reader, I know how it's built and how to make my own custom RSS feeds from their tags, categories, and formats to get exactly what I want from those sites.
#aaronpki already had these different "channel" on my site because I *want* them on my site as you're browsing around. regardless of feed readers.
#[chrisaldrich]reading back over it, I think my note on taking work away was poorly worded. I'm simultaneously thinking about things from both the publisher and consumer sides. Too often I find it painfully difficult to extract the content I want from many sites. No publisher or reader does remotely a good job at this presently.
#[eddie]like Inception level odd. However, I don’t see anything in the JSON of webmention.io to filter those posts out
#[eddie]I notice that webmention.io and bridgy is giving me a lot of “on a post that linked to http://eddiehinkle.com” with eddiehinkle.com as the target. However this is like 3rd level reference, rather than 2nd level reference. (Like, Marty replied to a post on my site. The post on my site appears to have a link to eddiehinkle.com). so I have to webmentions for Marty’s reply. One is his webmention to my post, and the second is that my post has a
#LoqiSalmentions are a protocol extension to Webmention to propagate comments and other interactions upstream by sending a webmention from a response to the original post when the response itself receives a response (comment, like, etc.) https://indieweb.org/salmention
#aaronpkbridgy adds more links in the page than strictly required. it often means you end up seeing comment threads on your posts without doing any extra work which is neat, but i think this is a side effect
#[eddie]Ahhh, gotcha. It’s strange that there isn’t some attribute to signify that. lol
#gRegorLoveYour first message cut off at "my post has a" btw
#mblaneynice work aaronpk, though I wonder if you considered web actions and indie config when looking at how to implement follows?
#mblaneyI feel like we had the follow problem sorted, but I guess that depends on how many other pieces you're prepare to implement.
#mblaneyFor instance, I've got a follow button on my home page at https://unicyclic.com/mal but you need a follow web action set up in indie config for that to work.
#mblaneyI've also implemented a status option for indie config, so if you have that too the button gets updated to "following mal" via CORS.
[jeremycherfas] and [eddie] joined the channel
#[eddie]gRegorLove: I was extremely happy when I wrote that article and realized that I had reactions set up and would be able to poop emoji myself. In my work slack we’ll do things like that all the time, where we add a random emoji reaction that emphasizes a point.
tantek, leg, cweiske, barpthewire, gRegorLove, loicm_ and arush joined the channel
#aaronpkmblaney: I did look at some other options before adding the new follow button to my site. I think there's room for experimenting with both options still, since this is still a relatively under-developed area. one thing about the web action approach I don't like is it's tied to the browser, so it won't work on my phone for instance. also despite having written some of the docs on
#voxpellimiklb: I haven't read the post yet, but imaging such a button could implement many different techniques to achieve its goal, indie-config and eg the new Chrome thing
#voxpelliUpside of components, one can hide away the details behind a simple abstraction and update it to use better implementations as standards evolve
#miklbI'm not 100% on any of it, but the approach sounded interesting
#tanteknor would I count it as "relatively easily install, use, maintain, depend on" for an independent. Typical Drupal installs I know are maintained by teams, or Drupal enthusiasts.
#[chrisaldrich]tantek: I've been noodling around with it recently. Earlier versions are easier to install. 7 is reasonable, though 8 is making the hurdle higher.
#[chrisaldrich]there's been some recent chatter on Twitter about better indieweb/drupal integration.
#tantek[chrisaldrich]: that's worth indicating in the project summary for context, that you personally have been experimenting with it
#aaronpktantek: unfortunately that's from mediawiki core so there's not a lot I can do
#[chrisaldrich]I'm hoping to update some of the Drupal related stuff shortly as I feel like there are people out there working on related things, but who need to be "welcomed into the fold" a bit more.
#tantekok I'm done with my silo feature documentation for the day
#[chrisaldrich]coming into the indieweb wiki and not finding much can be a turn-off to potential adopters...
#tantek[chrisaldrich]: we've had a few Drupal folks showing up occasionally to IndieWebCamps over the years. Lovely people. Always welcomed, just haven't seen much uptake / interest from that community
#tantek[chrisaldrich]: ^^^ bunch of people, tools, sessions already documented there
#tantekwould be interested to know what new things you are seeing
#[chrisaldrich]drupal has been tracking more to the "corporate" recently and especially with Drupal 8, but there are many on the hobby-ist side we could/should reach out to.
#tantekand that's a better page to work on improving for Drupal than the /projects page which has a lot more work to do
#[chrisaldrich]in tandem, I've been playing around with Backdrop CMS, a fork of drupal 7 that's more WordPress-ish, since last fall. It's much nicer for independents
#[chrisaldrich]they've been paring down a lot of bulk out of drupal core to make it easier to use/manage
#[chrisaldrich]and making it less dev-heavy and easier to install and use. (Their goal is to make it as easy (easier?) to set up than WordPress.)
#tantekchrisaldrich it *was* unwieldly and I started attempting a clean-up without completely wiping it
#tanteknote the header below which you added Drupal
#tantekin general I'm a bit leary of adding projects that people aren't *already* using live as their own personal web sites
#tantekbecause that's how it got unwieldly in the first place
#tantektoo much "here's a cool project either everyone should check out or has something indieweb related but I don't use personally for anything interesting"
#tantekbasically we need a place for experiment projects and see alsos that's not distracting from users who just want a simple list of options to use / install to "get on the indieweb"
#[chrisaldrich]I think that paring down some of the content under each of the CMSes and moving it to their direct pages may be the best way to go -- thus the reason I didn't add much underneath it.
#[chrisaldrich]I suspect it's better to have examples of people on the individual project pages is probably better than duplicating the details on /projects.
#LoqiHere are some projects you can use to get your site on the IndieWeb, improve your IndieWeb support, or take a look at to get inspiration for your own project! https://indieweb.org/projects
#[chrisaldrich]At the same time, putting some editorial work in ranking things from possibly a /generations perspective (from 4 to 1 in decreasing order) thus making the bigger, easier to use projects at the top would be wortwhile.
#KartikPrabhu[chrisaldrich]: I don't think a "rank" is necessary. that page already has more in the "experimental" section that others
[eddie] joined the channel
#[eddie]I second that thought, ordering from Gen 4 products down to 1 seems like a great way to list it
#KartikPrabhuthere are only 8 projects that are not "experimental"
#[chrisaldrich]While it's nice to have all those experimental projects for /plurality and PR, they could each be one liners that point to other individual pages. I suspect that a gen1 user coming to this page is more likely to be interested in open source things like Drupal for their widespread use and flexibility than they might be for some of those experimental pieces which aren't available at all to them.
#tantek[chrisaldrich] you're evaluating two different things
#tantekstart with reading that and seeing if it makes sense to you, or if you have suggested improvements to that approach
#[chrisaldrich]I've read it all. I think my issue is that may be one of the longest and biggest pages on the wiki with a lot of duplicated material which lives elsewhere on the wiki.
#KartikPrabhutantek: might be a good idea to move all the "Experimental" stuff to its own page /experimental-projects to clean up the initial clutter?
#tantekKartikPrabhu: maybe - but then you've created two problems
#tantekI'd rather see individual projects on /projects evaluated and moved to their proper destinations
#KartikPrabhuhopefully 2 smaller problems that are more mangeable than trying to edit that whole page at once
#tantekinstead of take half the mess of /projects and create a new mess that no one will clean up
#[chrisaldrich]Two separate problems. I'm thinking about someone who comes in and knows phython and is looking for something they could help extend rather than building from scratch if they don't want/need to.
#tantekand if instead you'd like to make a better resource for "someone who comes in and knows phython and is looking for something they could help extend", work on improving the /python page
#[eddie]tantek: That makes sense. Maybe move all “experimental/in development” to it’s programming language page (python, php, javascript, etc). Maybe adding a note higher at the top to the extent of “If you are looking for projects or building blocks in development for a particular programming language, you can find more here: https://indieweb.org/programming_language”
#[eddie]That would prompt people who know specific languages to be able to find projects by language, through the individual language pages
#[chrisaldrich]I think the best guiding advice is: "The projects page is focused on providing a clear flat list of projects you can set up on your own site to join the IndieWeb."
#[eddie]tantek: So here’s the question if chrisalderich (yep, I just volunteered you, haha. j/k) and I were to work on de-cluttering the page, you would recommend starting at the bottom (abandoned, then other, and so on). On a per project thought of where it belongs on the Wiki. Do we need to post somewhere for feedback before we move a project? or how should that process work? Or does the diff just cover that anyone interested can reply to the change
#tantekyou'll likely get a quick answer, e.g. maybe some historical context, or no answer, which means it's likely abandoned and be documented purely as historical (no need to put on the /projects page)
#tantekeddie is that call-out to Developers at the top of /projects more like what you were thinking?
#[chrisaldrich]I think in most cases a lot of them could have most of the details moved to individual pages and pared down to a one liner with a link to that page where they're better documented anyway.
#tantek[chrisaldrich] I agree most of them can have stuff moved to individual pages
#tantekI would not bother with even a one-liner if they are abandoned / obsolete
#tantekanyway, I think what eddie said makes sense as a general approach to make incremental progress at least
#[eddie]yep, tantek, that’s exactly what I was thinking
#[chrisaldrich]Project pages could also have something like [[Category:python]] too so that they're not as lost (or perhaps more searchable).
#tantek[chrisaldrich]: sure, we haven't had much luck with using/maintaining MediaWiki Categories but if you think they'll help we can try
#[eddie]Wow, look at chrisaldrich go! I'm headed out for the evening, but I'll look into seeing what I can do to help clean things up over the next couple of days as well
#gRegorLoveGWG or anyone else: Know if the wp-json endpoint for webmention has issues with additional params, like vouch=? I got a 500 error sending a wm to someone.