#GWGsnarfed: I've been soliciting thoughts again. I have this plugin I've started which replaces the comments template with a Facepile and mf2 compliant comments.
#GWGAfter some thought around here, I'm thinking of adding webmentions
#GWGWhat else fits into a plugin that takes over Wordpress comments of an Indiewebish nature
#snarfedGWG: another thing that'd be useful to work on is intelligent handling of text in likes and reshares
#snarfede.g. they should go into the facepile if there's no text, or if the text is just something like "X likes this," but if they have meaningful text they should probably display as a comment, maybe with an additional "like" indicator
#snarfedthat'd be good work that complements your and pfefferle's existing stuff
caseorganic joined the channel
#GWGsnarfed: The parsing is done by the semantic linkbacks plugin.
#snarfedgreat! and webmention sending and receiving would be done by wordpress-webmentions?
#snarfedwhee, the CMS developer tar pit is even bigger now :P
jedahan, caseorganic, voxpelli, hugoroyd_, ngoldman, benward and kylewm joined the channel
#Loqikylewm: snarfed left you a message 2 hours, 42 minutes ago: nice job handling webmentions for updated sources. rare to see that actually implemented
#JeenaI often wondered why not all likes/favs got to my page but kind of assumed they just were private on facebook or something so bridgy couldn't see them
#Loqibarnabywalters: snarfed left you a message on 5/14 at 10:38am: funny. i added the fb event oauth scopes after you signed up, so your bridgy access token doesn't include them. if you re-signup for bridgy, it should start backfeeding the rsvps
carlo_au, alistair, john3213, caseorganic, pfefferle, scor, Sebastien-L and jonnybarnes joined the channel
#barnabywaltershas anyone here had any poor experiences with ZIP as an archival format?
jedahan joined the channel
#barnabywaltersmy archive of HTML+headers of pages I’ve linked to is growing rather large — around 160MB
#barnabywaltersat the moment it’s not a big deal, but as it gets larger I’m considering storing it in a ZIP archive permanently, and reading/writing it using PHP’s ZipArchive class
#barnabywaltersand in the future I might implement something which goes over my old posts, checks for broken links and displays the (unstyled) archived data instead
#tantekif you're really looking to make the file standalone (not requiring special knowledge to interpret), you could translate the HTTP headers into <meta> tags
#tanteki.e. if you ended up serving that actual file from some webserver and then did the archive process on it again, you should end up with the same result
caseorganic joined the channel
#tantekI believe with the addition of the meta status code we did for webmention handling, that should allow you to capture all the HTTP headers as meta http-equivs
#rasculbarnabywalters xz might be better for compression
#Jeenasnarfed, I fixed the problem by adding the uid, and I see I got 54 likes from Twitter + Facebook on this image https://jeena.net/photos/9 but none from the indieweb! ^^
#Loqirascul: tantek left you a message 18 seconds ago: you can tell Loqi "!tell tantek something"
#rascul!tell barnabywalters you might look at xz for compression instead of zip it usually compresses things quite good, 7z is also pretty good with compression rates
#Loqibarnabywalters: rascul left you a message 18 seconds ago: you might look at xz for compression instead of zip it usually compresses things quite good, 7z is also pretty good with compression rates
#rasculbarnabywalters with xz you'll probably want to put everthing in a tarball and compress that, but iirc 7z has both a compression and archive format
#barnabywaltersrascul: the advantage of zip in this case is a) that it can archive a whole folder structure and b) that I can read/write it from PHP without extracting
#brettantek: I think I started the webmention wikipedia article XD
#luxagrafquestion for those of you using filesystem-based publishing systems, are you storing your comments/webmentions outside of your post/note/whathaveyou and if so how?
#rasculbarnabywalters i dunno about php's xz support though
#bretrascul: yes they are for newbies, but lower barriers is a good thing right?
#aaronpktantek: yeah cassisproject.com works for me now. yesterday i was getting the error on cassisproject.com and cassisjs.org but the wiki was fine, was just a problem with the redirect
#luxagrafbarnabywalters: are your regular comments stored the same way? or do you do webmentions only?
#Loqiaaronpk: tantek left you a message 55 minutes ago: cassisproject.com WFM - perhaps a transient PBWorks error? si(gh)los
#tantekaaronpk very odd - perhaps my DNS provider was having issues
#barnabywaltersluxagraf: the only non-webmention comments I have on my site are old ones imported from diaspora, all of which just have fake/approximated URLs
#luxagrafbarnabywalters: gotcha. i'm trying to figure out the best way to store and then lazy load both on-site comments and webmentions with as few requests as possible
caseorganic joined the channel
#tantekluxagraf: aside, if you're looking for backcompat in micro-markup for G/Y/B, uf v1 works for that
#luxagraftantek: thanks. that was my first plan, but now I think I'm leaning toward v2 + microdata because i see google eventually abandoning everything they don't control
#tantekluxagraf, example markup/data mechanisms that Google has abandoned, which were their own made-up stuff that they "controlled": Google Base schema, Google Data API/Elements,
#luxagraftantek: might be, i don't remember. i just remember they were a pain to hide in my userstyles file
#kylewmluxagraf: just aesthetics really... I wanted my post files to be more text content than structured data (with the intention that I might edit them by hand some day)
#tantekbret - google has abandoned opensocial but not before it took on a life of its own in an enterprise-run opensocial foundation which limps along to today. not sure who or what is actually using it live on the web.
#luxagrafkylewm: that's pretty much my thinking as well, but it means i need some way to associate the two, which adds a layer of complexity that might be even worse long term
#luxagrafkylewm: I'm always worried my entire publishing system is progressing something like the architecture of the winchester mansion
#barnabywaltersluxagraf: wanting to avoid difficult-to-maintain complexity was one of the reasons I went with URLs as the way of matching up responses to content
#barnabywaltersit also gives me extra incentive to keep my URLs working :)
#kylewmluxagraf: hahah, yes ... a valid concern that i share. going through a phase of feature proliferation right now, hopefully in the future i'll remove the stuff i don't use/need
#tantekTo be clear about one thing, I still intend to focus on this community first (including contributions here in IRC and on the wiki). I was asked to help W3C hopefully make some (re-)use of IndieWeb tech (and community practices).
#luxagraftantek: can you say who else is participating?
#tantekluxagraf - so far just the folks named in the charter page I linked to
#tantekbut that's how these things start at W3C, and then they send it out to the companies / members of W3C to officially approve and say whether they will send a representative
#tantekaaronpk, any early reports on Swarm? indieweb connotations?
KartikPrabhu joined the channel
#luxagraftantek: once that charter is up it might be interesting to add a community group that's someone related
#luxagrafthat's worked out well for responsive images
#tantekluxagraf - in a case where there isn't already a community, I would agree.
#tantekhowever, I'd say #indiewebcamp serves that purpose already
#luxagrafabsolutely, i was just thinking community group might have more pull with the W3C, which seems to discount things not under its auspices (from what I've seen)
#tantekright - I think that's one of the reasons they asked me to co-chair, was to bring more pull from IndieWeb into W3C
#kylewmluxagraf: re: previous conversation about frankenstein architecture. my biggest itch right now is that i want to generalize {notes,replies,likes,reposts,articles}, so that they can be more than one thing (e.g., an article can be in reply to something), and new types (e.g., "bookmarks") can be added without changing code
#luxagraftantek: come now, don't ruin the standards soap opera by getting consensus :-)
#tantekkylewm - feel free to add to that analysis on the /posts page
#tantekI think it's a challenging problem that none of us has quite figured out yet
nloadholtes joined the channel
#aaronpkit's probably time to follow up on that post now that i've actually created a bunch of content since then and have implemented photos
#luxagrafkylewm: back when friendfeed was big and everything was a "life stream" my cms became more abstracted like that
#kylewmtantek: cool, thanks for the pointers! right now all my stuff has the same data structure and presentation internally, so it's weird to me that i have these types hardcoded. can definitely see how photos/checkins/events would merit their own top-level type
#luxagrafbut i ended up dumping that idea and going back more specific content types because I realized 90% of what I did was articles anyway
jonnybarnes joined the channel
#jonnybarnesoh, wheres snarfed when you need him :P
#kylewmI think my goal is definitely smaller than "unify everything" ... I really just want to get rid of reply/repost/like as separate types in my code
#luxagrafkylewm: i can see that. those sort of seem like things inherited from silos that might not make sense outside of them?
#kylewmjonnybarnes: need snarfed for activitystreams knowledge?
#luxagrafreply always makes sense, but, in my cms at least that's just a url field. If that field is filled, the post will get in-reply-to class if not then it won't
#jonnybarneskylewm: for bridgy stuff, I've just moved VPS so my site was down for a little while, which just happened to coincide with someone favouriting one of my tweets
#jonnybarnesso the webmention that bridgy sends currently has a u-like-of for the twitter link and a u-like-of for the short link
#jonnybarnesnormally it also has a u-like-of for the full link as well
#jonnybarneswithout my site wont actually parse the webmention
#luxagrafnevermind the fact that I've never actually used said field, it's there, just waiting.
#kylewmoh interesting, so it cached the failed redirect
#aaronpkwow so many things I like about the new swarm app
#tantekKartikPrabhu: microdata is maintained now only in the WHATWG HTML "living specification" which contains all the WHATWG extensions to HTML in various stages.
#KevinMarksAnd shipping support in their browser extensions
bnvk and brianloveswords joined the channel
#aaronpkand on the complete opposite end of the spectrum, I'm super disappointed that instagram dropped foursquare's venue database and replaced it with facebook. the venues are *so bad* now
#JeenaSadly now they send the likes for both of their posts to bridgy which sends it to me with different uid:s so many of the likes appear double on my page :-/
#Jeenabut why can't they make it invisible for bridgy?!!!!
#kylewmaaronpk: the first one sent the webmention with target=..#fragment, but my actual post was rendering the u-like-of without the fragment. so the urls didn't match
#tantekluxagraf - that p-summary is on the comment being displayed on an article. The implication being that in general comments that are displayed may have summarized (or truncated) the actual content from the original reply post.
#luxagrafKartikPrabhu: not saying they didn't, or that testing is everything, but there's a whole lot of "we believe" in that post and not a whole lot of our "user testing has led us to believe"
brianloveswords and hadleybeeman joined the channel
#LoqiIRC is an abbreviation for Internet Relay Chat and is the primary discussion forum for the indiewebcamp community, in particular the #indiewebcamp channel on the irc.freenode.net server http://indiewebcamp.com/IRC
#LoqiA silo, or web content hosting silo, in the context of the IndieWeb, is a centralized web site typically owned by a for-profit corporation that stakes some claim to content contributed to it and restricts access in some way (has walls) http://indiewebcamp.com/silos
#aaronpkthat was the simplest definition of "sentence" I could think of
#tantek.comedited /Nginx (+154) "subhead re-arrangement, clustering, why, how to, move files without extensions to a how to, move TOC to more easily skim how to" (view diff)
#tantekaaronpk, Loqi could answer questions of the form "who uses xyz" simply linking to /xyz#IndieWeb_Examples to start with
#tantekalso would answer questions of the form "who supports xyz", e.g. who supports webmentions
#tantekquestions of the form "why xyz" or "why should I use xyz" or "why should I support xyz" could be answered by linking to /xyz#Why, e.g. "why should I support webmentions"
grantmacken joined the channel
#aaronpkgreps IRC logs for occurrences of those phrases
#pdurbinaaronpk: no fair. how can the rest of us grep the logs? :)
#tantekskip doctype, skip <html>, skip <!-- -->, if you see <head> then insert immediately after, if you see any other <tag> then insert immediately before.