#[snarfed]tantek: interesting. i assume you wouldn't similarly protect your twitter account due to mention spam because your followers don't normally see those mentions?
#[snarfed]kylewm: hah. maybe wrong repo. i think there is one implemented somewhere
#kylewm[kevinmarks]: is it (verbal expressions) more than a construction API for regexes? those always seem weird to me, like http://sqlkorma.com/
snarfed joined the channel
#kylewm(to be clear, that's a construction api for sql, not regex)
#tanteksnarfed, right, the problem is worst on IG so I would start with that.
#tantekI may take more silo profiles private in the future.
#snarfedpublic vs private culture varies widely across silos though
#snarfede.g. i expect it'd be a much more jarring change on twitter than on IG
#petermolnarre tantek I walked away from 500px for the same reasons you're about to private your IG
KartikPrabhu, Schnitz_ and [kevinmarks] joined the channel
#[kevinmarks]A more readable construction api for regex (l admit I like HyperTalk syntax, and python stuff like startswith() more than regex)
lukebrooker joined the channel
#Schnitzjesus just walking thru the tumblr developers forum, thats such a flaky bunch of people there. No wonder tumblr is the No. 1 fangirling vent
#Schnitzops this was meant to be for indiechat, sorry
#aaronpkdoes anyone tag posts as having happened at an event rather than at a place?
#aaronpkinteresting, "Pop-Up Magazine" was the event?
#Schnitzso snarfed getting back to my lil superfeedr question, you basically fetch any URLs anywhere in the DOM of the RSS right so it doesn't really matter where I put the webmention facebook public <a href> as long as its somehwere as a value and its shows up via superfeeds parsing?
#[snarfed]if superfeer passes that through to me in its content field, then sure!
#Schnitzcool no worries I'll just wait until superfeedr picks this up, nothing for you to do, just givin' it a try...
singpolyma, KartikPrabhu and ttepasse joined the channel
#aaronpkI've gone through all my post types now with the exception of one
tantek joined the channel
#aaronpk...my fancy "collection" posts, which I author by just adding URLs that point to h-entrys
#aaronpkI think I'm going to flatten those into blog posts for now, possibly for good. I might re-create that experience as a new way to author things in Quill.
#KartikPrabhudoesn't using a database just compound the filesystem ones
#KartikPrabhuif a filesystem is bad everything built on top of it inherits its badness and then some more. So if the filesystem is the base-level you have access to then use that
#acegiak!tell GWG I'm having trouble getting start and end stamps to work. are they working for you? The metadata is there but it's not showing up in the fields or the post display after saving
#Schnitzalright thanks (trying to get control over the RSS generator of Tumblr, any ideas?)
#tommorrisenjoying https://whatdoesmysitecost.com/ - loads your web page, tells you how much it'll cost to buy the data necessary to load the page in different countries.
#GWGacegiak: It should only show up with certain kinds, But it should save. I tested saving repeatedly.
#LoqiGWG: acegiak left you a message 40 minutes ago: I'm having trouble getting start and end stamps to work. are they working for you? The metadata is there but it's not showing up in the fields or the post display after saving http://indiewebcamp.com/irc/2016-01-04/line/1451911396808
#petermolnarKevinMarks I'm reading the filesystems post; there is no mentions of zfs' and btrfs' copy-on-write by default mechanism, which in theory, is pretty resilient.
friedcell, lewisnyman, endi, mxuribe, hs0ucy, tvn, ttepasse, marcthiele and snarfed joined the channel
#SchnitzI've put the webmention publish under category
#Schnitzthat would have been semantically kinda nice and ok, 'cos tumblr allows me to access that via tagging... adding a tag with publich webmention url to a post in tumblr seems kinda natural
#snarfedyou can see the publish processing details if you look at one of the logs
lewisnyman joined the channel
#Schnitzso basically you poll the RSS to see if anything is new, take the webmention call from there, THEN go to the HTML version and use h-entries, etc.?
#snarfedsuperfeedr polls the rss, it pings me when there's a new post
#snarfedpublish is a separate feature from blog webmentions (ie tumblr support)
#Schnitzno worries, totally with you, I'm your fan .-)
singpolyma joined the channel
#Schnitzvery interesting... but this also means you're actually making some sense out of my HTML representation and there I /should/ have some control it, since I /can/ control the HTML in Tumblr. This is awesome.
#Schnitzmaybe silly question, can you handle more complex posts then into FB like photosets, etc.? H-entry-wise this should be no problem to be marked-up, or complex threads, question is in what way can you get this into FB?
#Schnitznot too bad. Photo posts is something I should definitely be able to bugfix then with my current setting. I have something to check out / improve then. Thanks :-)
lewisnyman_, squeakytoy2, glennjones and shiflett joined the channel
#tanteklet's see if Anil picks up indieweb this year (we keep reaching out to him nearly every year!)
#tantekpetermolnar: I've kind of given up on clicking on hackernews / reddit links due to low quality of discussion. If there's something particularly interesting, please quote it inline! Thanks!
#petermolnarI will; that was too hasty to link anyway, and I'll pay more attention from now on
uranther joined the channel
#petermolnarby the way: I see tweets and notices going to the Old Ones of Blogging (anildash, h0d3r, etc) about blogging, none seem to reply or look interested, and I really wonder, why?
#tantekpetermolnar: they might be burnt out, or despondent.
#tantekfortunately for every old one of blogging, there are dozens if not hundreds of new ones of blogging, so progress is not bottlenecked on any particular gatekeepers.
#tantekmaybe that means it will get published tomorrow morning!
wolftune joined the channel
#Schnitzoh wow, webmention as a w3c thing? really?
#tantekSchnitz: perhaps you want to consider joining the W3C Social Web Working Group :) w3.org/wiki/socialwg
#tantekand yes, the Social Web Working Group *charter* included Webmention explicitly as one possible input (years ago), and last year decided to adopt it as an editor's draft, and then resolved to publish a first public working draft too! (all last year)
#Schnitzdon't do this to me, and official invite to get back to w3c... I had promissed myself to do something else in life, I feel very much flattered and tempted though :-)
#Schnitzno I'm sure, but getting to to REC? years?
#tantekso I expect to go through CR very quickly, if we can fight off all the random feature requests from architecture astronauts
#SchnitzI admire your faith, energies and optimims. Go for it!
#tantekwe're likely going to have a disposition of comments with lots of formal objections like ("doesn't include my pet hypothetical feature which is required for anything to work")
#aaronpkSchnitz: lol that sounds like my framework I called "Website Template Framework" (WTF) so that the lawyers had to write that in all the legal docs I was involved with
#LoqigRegorLove: tantek left you a message on 12/31 at 8:04am: we are a go for 2016-01-13, can you help out with creating pages for 2016-1? Thanks! (also feel free to expand the HWC template embed to redlink the whole year) http://indiewebcamp.com/irc/2015-12-31/line/1451577851070
#ben_thatmustbemekind of. so its not like its really duplicates at all
#ben_thatmustbemethey are all there, its the right number of them, but the way streams was written was just as permalinks to the likes, not any author info
#snarfedi definitely prefer that instead of omitting url so that people can click through to the silo post. (ideally end users should never end up on bridgy's own mf2 pages, which is what would happen otherwise.)
#tantekthat design detail (and reasoning) should be documented somewhere, perhaps even on /Bridgy
#tantekotherwise we'll have more implementations with such bugs
#LoqiBackfeed is the process of syndicating interactions on your POSSE copies back (AKA reverse syndicating) to your original posts https://indiewebcamp.com/backfeed
#snarfedhaving said that, i kinda wonder if webmention.io should handle this better. bridgy definitely provides unique uids for each like, but webmention.io doesn't serve the uid, so if consumers want to de-dupe, they have to fall back to de-duping on url
#snarfed...which breaks in this case, since i didn't think url needed to be unique
#snarfed(ideally it would be, of course, but silos gonna silo)
#ben_thatmustbeme!tell tantek i had a chance to look at that issue again, it would have shown the author if it had a type value for the author. if there is no type, its not in the correct format, and thus ignored
#Schnitzsnarfed, always a first, this might work, reblogging pgotos inside tumblr and while reblogging, adding the webmention post url... basically 'shooting' stuff out of one silo into anther by just adding http://brid.gy/publish/[silo] when reblogging... pretty cool, should show up my FB test page soon
#Schnitzyou could always copy'n'paste manually, but thats little more 'active', and you feel 'you're doing something'. It being automated adds a new level of fluidity.
#Schnitzinteresting Wiki articles, thanks for sharing, snarfed
#Schnitzwhile I was writing this I found a possible super simple solution
#Schnitzsince your publish routine *does* fetch the post in HTML already
#Schnitzyou could still stay with superfeedr on check the RSS to see whether there is a new post
#Schnitzand get the post with your publish routine *regardless* of webmentions, just to see whether in the HTML representation there are webmention <a href=""> and then publish
#Schnitz... I had thought implementation wise, since you already have code behind those webmention url grabbing the post as HTML, you could use that and modify it to a 'webmention detect' URL, so when bridgy gets something from Tumblr you then first throw it against that and find the actual webmention URL in the HTML, but thats an implementation detail I cannot foresee of course, sorry for that
#SchnitzI am very happy to extend the wording of issue 590 text-wise to cover this idea as a possible way of dealing with this, if its any help, snarfed
Pierre-O joined the channel
#snarfedthey're definitely related! this one sounded like triggering publish specifically. 590 sounds like fetching HTML to get all URLs, including outside of content.
#Loqitantek: ben_thatmustbeme left you a message 2 hours, 34 minutes ago: i had a chance to look at that issue again, it would have shown the author if it had a type value for the author. if there is no type, its not in the correct format, and thus ignored http://indiewebcamp.com/irc/2016-01-04/line/1451939694582
#tantekben_thatmustbeme: I think I'm still not understanding where the problem might be - is there an assumption of what a particular post type has / doesn't have?
davidascher, shiflett and Lancey joined the channel
#davidascherhi y'all. tantek mentioned on twitter that people have been thinking about how to federate the stream, in response to a tweet of mine. I haven't written up what I had in mind yet, but I'm happy to start by reading what others have done.
#tantekdavidascher: a good place to start is what do you mean from (your) user perspective? what user feature are you talking about implicitly when you say "federate the stream"
#davidaschertantek: sure, let me try and blather a few lines, see if it helps.
shiflett joined the channel
#davidascherI'm assuming that I have a network of people willing to work in cooperation, and in particular happy to defer any questions of monetization.
#tantek(davidascher aside: each my @-replies to you actually originated on my own site, tantek.com, which then automatically syndicated those to Twitter so you would see it, a type of forced-silo-federation as it were)
#davidascherI've found that there have been some interesting UX advances in the world of blogging, both w/ medium and wordpress, which lead me while reading one piece to reading other related pieces.
#davidascherI'd like to explore whether there's a way to foster that kind of reading content that may be originally hosted on different, unsiloed blogs, but with the same continuity of reading that these silos have built.
#davidascherIOW, could I make ghost or wordpress.org find related content via an explicit author-managed or (ideally) reader-influenced network of related content.
#KevinMarksthe story cards idea is a bit like that
#tantekdavidascher: one thing the indieweb community has proven is that no separate REST endpoints nor RSS feeds are needed, if you do a little bit of work on your home page
#KevinMarksthat was based on the silo presentation of related stories
#tantekthere's alot of "what is the minimum viable ... " here
#tantekdavidascher: but to answer your question, I don't know of anyone yet on their own site where they have some sort of (semi?)automated "Related Reading" function to articles on *other* sites.
#tantekI've seen plugins that do automatic "related reading" from amongst your own posts, but you're clearly talking about something beyond that
#davidascheryea, I'm specifically looking at related reading as a way of dealing w/ the UX failure that (I consider) feed readers, but with the same silo-tunneling goal.
#davidascherThese still (and I understand why) assume that the reader is ok w/ shifting from site to site to read new content.
Schnitz_ joined the channel
#KevinMarksare you thinking of a reader presentation?
#davidascherWhat I'm wondering is whether we can turn a blog into a thing that doesn't just display the article from that one blog but also a front-end that can display content from other sites. It does presume that 'article' is a content type.
#davidascherIn particular, I'm wondering not whether it's technically possible (it obviously is, provided enough effort), but whether people have thought about ways to do decentralized recommendation systems beyond "webrings".
#snarfedi bet we could do it stumbleupon's API, or something similar
#davidascherI think I should probably write it up as a post.
#snarfedwe definitely like decentralized interactions here, but we don't mind using centralized services to mediate those interactions
#snarfedrecommendation engine is a great example. hard to do fully decentralized.
#davidaschersnarfed: yup, that could work. Another thing I was wondering about is local spidering of a twitter graph for example -- look at who the author follows on twitter, check their home page for appropriate content, suggest that.
#tantekdavidascher: a good way to start broader thinking about this is to document the existing "related reading" UIs that you are seeing
#snarfedsure! doable. just expensive and nontrivial for everyone to do themselves. begs to be amortized.
#davidaschersnarfed: q: how critical is it for "everyone to do themselves" in this community? I assume we're all using some OSS software and not requiring everyone to host baremetal and hand-craft html pages yea?
#snarfedyup. in general we support all different levels of hosting. the one constant is that people need their own domain. that's what enables owning your data, migrating, etc.
#petermolnarsnarfed about the owning the domain thing: I'm still sad that there is no way to actually "own" a domain; at best you're renting it for a while. Except for .onion, where you get a random generated string; that, in theory, you do own.
#petermolnar(that doesn't change the need of a domain though)
#snarfeddavidascher: sorry, yeah, that plugin is kinda deprecated. we now recommend that you install the webmention and semantic-linkbacks plugins individually instead
#tantekboth of those pages say to install the IndieWeb plugin first
#davidaschernone of those seem to do what I was actually trying to do, which is to let me add rel="me" to links on my homepage.
#tanteksnarfed, where on the wiki does it say "we now recommend that you install the webmention and semantic-linkbacks plugins individually instead" ? or how new is that recommendations?
#snarfed.orgedited /WordPress (-216) "/* Adding Indieweb Support/Elements to a WordPress Site */ remove indieweb plugin" (view diff)
#davidascherwhich the WP UI doesn't seem to let me do.
#snarfedtantek: i haven't followed all this in detail. GWG pfefferle etc are more up to date than me
#petermolnardavidascher you're right, that is not part of those plugins; it's usually part of the theme