dckc, [tantek], nickodd, [tw2113], B, gRegorLove_, [chrisaldrich], [schmarty], gRegorLove__, gxt, [fluffy], cweiske, [jeremycherfas] and swentel joined the channel; nickodd left the channel
#[jeremycherfas]Huh? Garmin US no longer allows login via VPN. I can almost under that for tv streaming, but for an activity monitor?
kingpapa joined the channel
#@iamhirusiYay! I have added support for mention-of #webmentions on my site! :D
It was so simple to extend it from how I handle a webmention reply. Structured data from an API + structured development = win. :) (twitter.com/_/status/1278955984023482368)
moppy, KartikPrabhu and gRegorLove__ joined the channel
#petermolnarI run this regularly, though originally for scenarios if the source repository disappears, not because I thought gh will cascade-delete repos
#sknebelthats something I also haven't found a really satisfying solution for. one way could be moving branches to unique names before overwriting them (and maybe accept the risk of there being conflicts later if the name is random enough), or using the data-sharing options for local clones to create a new copy of the repo and only update on of them, keepign the old states around
#sknebel(https://git-scm.com/docs/git-clone has a bunch of options for referencing the data files from other local clones of the same repo, so having multiple ones doesn't necessarily mean duplicating all files. or use lower-level dedupe.)
#sknebelbut its all stuff that needs careful figuring out
#cweiskeI solved that the same way I do for normal server backups: full backups that use hardlinks if files are identical to save space. rsnapshot does this.
#cweiskethat allows me to go back to earlier versions
#sknebelright, that's an example of lower-level. and a good candidate for being reliable
#LoqiAnnotations are comments (including marginalia), highlights or any other interactions that add to (part of) a post, typically added by individuals other than the author https://indieweb.org/annotations
#jackyugh I'm trying to remember the name of the API that Google uses to fetch your account for signing in
#jackyit's like the sign in functionality that appears on Medium
#jackyI want to see if there's a way to spec that into something that's friendly to federation to make signing into a IndieWeb site a bit simpler (a click instead of having to explicitly type in your domain name)
#LoqiSubToMe is a button that content publishers can put on their sites which enables users to subscribe to their feed in whatever reader they choose — sort of a universal follow button https://indieweb.org/SubToMe
#aaronpkjacky: see that for some prior art on a way to do it
#LoqiIt looks like we don't have a page for "indieaction" yet. Would you like to create it? (Or just say "indieaction is ____", a sentence describing the term)
#LoqiA web action is the interface and user experience of taking a specific discrete action, across the web, from one site to another site or application https://indieweb.org/webaction
#jackyyou could pack it all in some JS and have a <universal-follow-button> component
#aaronpkbut where is the data stored, that's the problem
#aaronpkthe reason it works with subtome is because the data is stored associated with the central domain, so when it's embedded in other sites those other sites can get access to it
#aaronpkwithout the central domain, each domain would have its own isolated storage container for that custom element
#jackyRight and I don't know if that's a bad thing? Like it can be very small (like just a JSON dictionary) and it can call out to a index if it needs to "refresh" its index
#aaronpkno i mean that you come to a new site and it has no way to show you your domain until you enter it
#aaronpkalso doing anything like this in JS is actively fighting the anti-tracking stuff browsers are starting to lock down on, so I would be wary of investing too much time into it
#[chrisaldrich]I've noticed that a few things I (and others) have bookmarked to news.indieweb.org recently have come through without much context. Is there some secret microfomat and data we're missing that would remedy this?
#[chrisaldrich]I've had a few examples like Jacky's recently too... but I wouldn't really want it picking up my commentary on something I've bookmarked.
#[chrisaldrich]If people want to read that, they could click through my link, but the point is to highlight the other person's post.
#[chrisaldrich]I also need to look into how I'm marking up my read posts so that I can syndicate those as bookmarks to indienews. I suspect indienews doesn't parse for read-of, so if I added a secondary bookmark-of, it would catch that as a fallback for post type discovery and I'd be okay? Or maybe I could leave it off and it may default to bookmark-of finding nothing else?
#[chrisaldrich]Generally when it comes to these things, I'm almost 99% sure I'm not doing it right though...
#aaronpki think i could reevaluate how indienews gets its content
#aaronpkit was a very early project before many of the other tools were very mature, so the assumptions I made back then may no longer apply
#[tantek][chrisaldrich] more importantly we should make sure that the markup reflects the publishing intent. And only secondarily work on how should things like indienews consume them
#[chrisaldrich]does resubmitting bump the submission date and time and put things back to the top of the list or just update the old post?
#@aaronpk@dcuttridge We've been working hard on fighting the idea that being on the IndieWeb is only if you build a website from scratch, so we may have gone a bit far in the opposite direction with that page. Will try to clear it up! (twitter.com/_/status/1279083522481729537)