#GWGThe Indieweb Plugin, which is just an installer for Indieweb plugins, declares only two plugins as ones you MUST install.
#aaronpkI just mean if someone has a plugin like Webmention handling or something and they want it to be more community maintained then that's fine for the IndieWeb org
#snarfedi'm with kylewm here. we see a steady trickle of people who see ugly/broken wm comments in wp because they have one plugin but not the other, or because the versions skew
#kylewmI take your point about modularity and reuse... it is nice to have semantic-linkbacks be abstract enough to support other notification mechanisms or other parsers
#kylewmperhaps that could be handled at a library/packaging level? so two plugins might both use the semantic-linkbacks "library"
#GWGThe original Indieweb plugin was a single plugin that included both plugins, but it wasn't maintained and was behind the dev versions, so it was replaced with a plugin that installed the other plugins
#GWGWordPress, by the way, has no dependency management
#gRegorLoveOk. If they remain separate plugins, how about a dependency check when installing WP Webmention plugin, "it's strongly recommended to install Semantic Linkbacks next"
#aaronpkif i was using wordpress, i would much prefer to read a tutorial about which plugins to install than installing a plugin that tells me to install plugins
scoates joined the channel
#gRegorLoveI think that's a decent flow. Given how much the term "webmention" is associated with indieweb, seems much more likely someone will install that plugin first.
#gRegorLoveSo a nice prompt (dismissable) suggesting installing Semantic Linkbacks next might clear up things.
#gRegorLove"Semantic Linkbacks" is gonna be jargon to most people, I think.
#GWGaaronpk: I just wonder if the wiki is conducive to that tutorial, or there should be a companion site.
#GWGgRegorLove: I don't know who named it, actually. I assume acegiak
#aaronpkthe wiki is a great place for that because it's community-maintained
#ben_thatmustbemeespecially with mysql backend being so picky about always storing in UTC, but then not really paying attention to zone offsets when inputting
#kylewmaaronpk: I've thought that it might be better for mf2util to have a bunch of utility functions like get_first_hentry get_author rather than trying to completely abstract mf2 away from the user
#KartikPrabhu!tell tantek: to file a bug to Feedly I have to Join their G+ community. Not inclined to do that. Communites on G+ are a big source of noise
#KartikPrabhuidea: someone sends a webmention through form, do AJAX stuff and if webmention accpeted update with response and scroll to that to show "accepted"
#tantekKartikParbhu you're basically describing a lighterweight form of comment/reply webactions that barnabywalters started working on
#KartikPrabhucweiske: how about comparing microformats properties?
#gRegorLoveReleased my ProcessWire webmention code! The core sending/receiving/processing is pretty stable, but I'm still developing /Vouch support as well as an admin interface to view/approve incoming webmentions, so I'm considering it "beta" for now.
#gRegorLoveI'll work on the readme tomorrow and share it with the ProcessWire community. Maybe we'll attract some new indieweb fans.
#KartikPrabhugRegorLove++ for new CMS webmention support
KevinMarks__, petermolnar, KevinMarks___, loic_m, keroberos, interactivist, lewisnyman and pfefferle joined the channel
#voxpellicweiske: I think diffing of individual mf-properties would be a bit too hard – rather just do edits as replacement of individual properties
#voxpelli!tell KartikPrabhu If you have realtime updates of WebMentions then all you really need to do to deliver feedback is to have the submit happen over Ajax – the update will come back when it's ready, which might take a while
#Loqipfefferle: snarfed left you a message 2 days, 10 hours ago: hi! looks like your semantic-linkbacks repos have diverged, and each one has useful new features (filters vs salmentions). mind picking one to be the single canonical repo, and then merge them? thanks in advance! http://indiewebcamp.com/irc/2015-07-19/line/1437355029097
#GWGpfefferle: Don't really have the moment, or you can't explain what you have in mind?
#pfefferleGWG thought it might be a good idea to upload some stuff/examples/hacks...
#lancey.spaceuploaded /File:quoted_tweet_example.png "An example of a [[quotation]] on [[Twitter]]. The quoted tweet is embedded in the status update rather than being the main focus, and the user can provide an additional caption, making it different from a retweet."
#GWGsnarfed put me down the unit test path, I found testing for WordPress Coding Standards at the same time. I failed.
#pfefferleGWG I started using the WordPress coding standard a while ago
#pfefferleGWG porting plugins is very time heavy ;)
#pfefferleGWG but there are a lot of good editor plugins to check the code every new save
#GWGpfefferle: I know. But I figured, rather than you and snarfed telling me I put the wrong spacing or style, I figured I'd let an automated tester tell me
#GWGAlso sets up for those unit tests that snarfed was encouraging for Micropub.
glennjones joined the channel
#GWGI started to set up all my projects to run them
mlncn and Pierre-O joined the channel
#Loqislack/cdevroe: benwerd: Congrats on the Known push.
#cdevroeAll that effort to get into Slack and I'm now finding it pretty difficult to "catch" up when I pop in in the morning. How do all of you "catch up"?
mlncn joined the channel
#cdevroe(Slack doesn't do a great job of keeping your marker, for some reason. I've sent them bug reports about it but I haven't seen any changes.)
#cdevroeGWG speaking to some locals here, we have a hookup to do a IWC in a trolley while it is moving from place to place. ;-)
#cdevroeI'm thinking of switching from Safari to Firefox today. My only wish was that Firefox was "more of a Mac app" and played nicer on OS X. E.g. Dictionary look ups on words, etc.
#LanceyWorkcould you build your own plugin to do that?
#petermolnarWordPress users: I could not sleep yesterday, so I wrote an audioscrobbler receiver ( server ) WordPress plugin in order to get rid of Last.fm in the future
#LanceyWorkhas anyone thought about an infobox or other template on the wiki that we can apply to post type pages that displays general information such as the microformats syntax or silos that implement it?
#tantek!tell KartikPrabhu cont'd from last night re: comments box: the idea is that every "native" blog comment box turns into a simple indie reply authoring micropub client, so you can write a comment inline when viewing a post on someone's site, have it post to your site, then show up right there on the page you're viewing.
#voxpellitantek: would the micropub client live in the browser then?
XgF joined the channel
#tantek!tell cweiske thanks for the heads-up re: /edit - only publishing example(s) so far. updated my post reference to /edit with a parenthetical note accordingly (like person-tag in 2014)
#tantekvoxpelli: oh that's a possibility too - was thinking the server you're viewing could be a micropub client that you give permissions to *only* post reply posts to your own site.
chalettu and keroberos joined the channel
#voxpellitantek: rather cumbersome to do the indieauth dance on each and every blog one wants to comment on :/ I will probably just set up indie-config to point to a reply form in Quill
#bearit could be like subtome - if js is enabled it's a box that allows content input and uses your local store config to post micropub; if js not enabled it's a redirect to a service to enable comments
#Loqitantek meant to say: indie-config is a more user-friendly solution
#Loqiindie-config is a method of using protocol handlers and postmessage to setup your indie website to both notify the browser that it can handle webactions and then do so https://indiewebcamp.com/indie-config
#voxpellione could build an indie-config that has the config stored in localstorage if one wants it to work like subtome :)
#tantekthinking the site showing you the post, just as it shows you webaction buttons, it could show you a cross-site iframe onto *Your* site for posting a reply
#voxpellitantek: yeah, but if you're not logged in to your site you may have to log in before and since it's an iframe you're not sure if the form you're seeing is being proxied through a phishing site or something :/
#KevinMarksso where would it get the micropub token?
#tantekvoxpelli: no - it would only show up if you're already logged into your site
#tantekwhich is still A LOT less work than having to login to EVERY SITE you want to comment on
#tantekif you're not logged into your site, then all you see is the "Reply" button
#KevinMarksthe iframe would have cookies from your site already
#voxpellitantek: yeah, I guess one could make it safe, so it could be a possible evolution
#voxpelliI wonder though how often silos have response forms inline versus on a separate page – especially on mobile
#cdevroeCould do a time limited cookie that does a round trip from your personal domain to the site you're trying to comment on to create a cookie there. So... if you're not logged in: 1) click reply, 2) takes a trip to your personal domain, sets temp cookie 3) reroutes back to origin domain endpoint, sets temp cookie 4) write reply, click send 5) posts variables to personal site, saves
#tantekvoxpelli, cdevroe getting "reply" user-flow as *easy* as possible, and at least as good as what silos do is *key* to making indie web interactions work well enough for people to actively switch.
#voxpellimain addition for both separate and iframed incie-config would be to specify a return URL – so the including page knows when the flow is done
#voxpellithen it would mostly be a matter of responsive webdesign to adapt the page to fit within an iframe – and perhaps to set some expectations around the size of the iframe?
#KevinMarksthe including page should get a webmention to know it is done?
#cdevroevoxpelli: It is possible to do stuff with XSS. S3 does it.
#voxpelliKevinMarks: well, in the case of navigating to a full page the original page will have been closed
#voxpelliand webmentions can have quite some latency
benwerd joined the channel
#voxpellieg. if my webmention endpoint gets flooded with mentions of the same domain I start throttling lookups
ttepasse joined the channel
#cdevroeaway for lunch... I'll be following along with this.
#voxpelliI correct myself regarding Twitter UI:s – apparently it's just Tweetbot that does it on a separate page
#tantekvoxpelli: makes sense - can you document what you're seeing in the Twitter App UIs? screenshots of the user flow?
#voxpellitantek: I guess, is there any place where we already have such UX/UI research at?
#voxpelli(Would need to figure out how to upload files to the wiki as well :P)
#kylewmvoxpelli: is there a mechanism for updating that list of twitter accounts periodically? i'd be interested in trying to wire up a little API app that creates a twitter list from it automatically
#tantek.comedited /comment (+141) "start fixing this page to be primarily about reply than comment as the name of the standalone post, before the comment split, start clustering reply vs comment" (view diff)
#voxpellikylewm: that should be doable, if nothing else then semi-automatically by me, but the service is built to run as a continuous crawler so I should put it up somewhere eventually anyhow
#voxpellikylewm: if an app we're to continiously parse irc-people for new users and then call https://github.com/voxpelli/relspider#apilookup asking for more data about that user, then that would pretty much solve it – also supports webhooks for callbacks if the user hans't been crawled already
#kylewmalso the neo4j is just down the road from me :)
schmarty joined the channel
#voxpellikylewm: just down the road geographically speaking?
#kylewmyeah, i left out "office" accidentally... there office is right down the road
#voxpellikylewm: only thing I haven't really solved that well in the current version of relspider is the recrawling. Then I tried to look into strategies for that. Aaand I found a rabbit hole. A very deep such :)
#voxpelliturns out recrawling is more or less an entire academic field :P
#voxpellikylewm: I can start a copy of the spider on heroku if you want to try and do some interactions with it?
#kylewmvoxpelli: thanks, that's ok though. I got pretty close to having it running on heroku myself, think I can get the rest of the way there when I have a bit of time
#LoqiDocker provides a way to package an application into a "container" that includes all the dependencies it needs to run https://indiewebcamp.com/Docker
#bearthe host os is moot for seemless updates of the container environment
#bearwhat does that is tools like kubernetes or fleet/swarm
#Jeenavoxpelli that is an interesting idea, I don't use WordPress because my servers got owned 3 times because I forgot some old wordpress installation was still running somewhere, but if it is in a container then even if they find some bug in WordPress it still wouldn't compromise the whole server
#bearkubernetes allows you to do rolling updates based on version info
#LoqiKartikPrabhu: voxpelli left you a message 10 hours, 5 minutes ago: If you have realtime updates of WebMentions then all you really need to do to deliver feedback is to have the submit happen over Ajax – the update will come back when it's ready, which might take a while http://indiewebcamp.com/irc/2015-07-22/line/1437552093574
#LoqiKartikPrabhu: tantek left you a message 2 hours, 5 minutes ago: cont'd from last night re: comments box: the idea is that every "native" blog comment box turns into a simple indie reply authoring micropub client, so you can write a comment inline when viewing a post on someone's site, have it post to your site, then show up right there on the page you're viewing. http://indiewebcamp.com/irc/2015-07-22/line/1437580908416
#KartikPrabhu!tell Jeena, yes hfeed2atom does not follow the entries to permalink pages yet. But if there is no 'content' property it adds a "read full post" type link
#Jeenabut I want to reat the full text in the feed reader not on the website ;)
#LoqiJeena: KartikPrabhu left you a message 1 minute ago: hfeed2atom does not follow the entries to permalink pages yet. But if there is no 'content' property it adds a "read full post" type link http://indiewebcamp.com/irc/2015-07-22/line/1437588564521
#KartikPrabhuJeena: of course that's for your reader to decide :) I don't want a h-feed to atom convertr to make that decision
#Jeenaand at least while having it as microformat you have the ability to check if there is a full post or not, once it is in the atom format that information is gone :)
#gRegorLoveben_thatmustbeme: Just noticed all the rel in-reply-tos on your homepage. Is that intended for the stream?
#KartikPrabhuthat is why I am happy with hfeed2atom not taking that responsibility. Just convert h-feed to Atom
#JeenaKartikPrabhu but fetching the link would mean that you get the whole html website with header, sidebar and footer, etc. and then you still have to parse the microformats to get the content
#KartikPrabhui see feed readers as "my website somewhere else" so I don't see the distinction
#JeenaI use RSS more often than not offline so I want to have the content right there not first when I look at the content, because then I often do not have access to the internet
#KartikPrabhuyes. so it is the responsibility of your reader to do the fetching
#Jeenafor hfeed I'd agree, for RSS hm, difficult because you never get just the content but you have to implement something to extract the content from the whole HTML which is not a trivial thing to do
#KartikPrabhuwhen a feed reader encounter what it thinks is a partial feed it can fetch the full thing for content (just like a human reader would for an hfeed)
#KartikPrabhuon a website also you never get just the content
#LoqiI: cweiske left you a message 3 weeks, 1 day ago: didn't check php-resque. I was happy with gearman. But now I just found out that I need scheduled tasks, which is something that neither gearman nor php-resque give me http://indiewebcamp.com/irc/2015-04-07/line/1428398114615
#Jeenatantek do you have an example of a offline website I could look at? I mean I want a nice UI which means I want to be able to start it on my phone from the home menu, or on the desktop as an application
#JeenaI wrote https://marketplace.firefox.com/app/feedmonkey/ which is a HTML/CSS/JS application which works offline. But I made it installable on the phone because I miss some detail in how to make it so it won't want to redownload the app from the internet when I get on the websige in the browser when I'm offline
#KevinMarksvoxpelli: I think the new twist is getting notifications when the client isn't running
#voxpelliKevinMarks: yeah, ServiceWorkers and the Push API should be able to handle that
#KevinMarksright, though Chrome only for now until the new Firefox shows up
snarfed1 joined the channel
#snarfedjust to play devil's advocate, Jeena, using existing "legacy" feed readers that have existing mobile apps with good offline support (feedly, newsblur, etc) is *totally* reasonable
#snarfedeven for reading indie sites, since we have services and libs like hfeed2atom and granary that convert h-feed to rss/atom
#snarfedi'd love to have a good mobile offline indie reader that supports h-feed…but making a good reader is a *big* project, especially with good mobile/offline support
#snarfedit'd be great to build one! but that's a really big ask, so i vote that you can be a good indieweb citizen and still use legacy readers for now
cdevroe joined the channel
#Jeenasnarfed yeah unless you want to read aaronpks or tanteks stuff in those readers because hfeed2atom doesn't get the rest of the partial feeds :)
#tantekaaronpk, Jeena definitely add both of those Quill / feedmonkey to the /offline page!!!
#KartikPrabhuJeena: neither do traditional feed readers
#snarfedJeena: hmm. i get tantek's full posts in a legacy reader without using any lib or service
#Jeenahehe ok I just took tantek and aaronpk as an example but I guess you get my point in theory ^^
#tantekusing AS/Atom - but I don't know of any "reader" that consumes them :/
#snarfedJeena: eh forget theory, it's all about practice. :P in practice the vast majority of indieweb sites i want to follow still publish legacy feeds
frzn_ joined the channel
#snarfed…and the rest will be solved soon with KartikPrabhu's service!
#KartikPrabhusnarfed: Jeena is saying that hfeed2atom does not fetch the full content for you. which is fine for me, since the original h-feed doesn't have it anyway
#tantekwhat's the best way to tell how many times someone "reads" one of your posts in a legacy feed reader? per post 1px gif bugs that are only in the Atom versions?
#snarfedKartikPrabhu: sure. i concede any detail like that. hfeed2atom is still way young and early yet
#snarfedmy point is that "build a pure indieweb solution" is a good way to get functionality, but definitely not the *only acceptable* way, especially for stuff like mobile offline readers that are very substantial projects
#KevinMarksthough I shouls add some proper caching
#aaronpkmore subtle would be to use a different image proxy for images in your posts for the html vs atom feeds
#tantekvoxpelli: more interesting is, how many times was the post actually displayed in a *feed reader* (as a rough approximate for # of times actually read by someone using a feed reader)
#KevinMarksah, so do i need to change useragent or something?
#Jeenaa couple of years ago I fucked up my awstats setup and never fixed it, so now I don't know if there is anyone reading my stuff and I feel free!
glennjones joined the channel
#voxpellijust such a thing like https requires any web based feed reader on https to load all posts through some kind of proxy – and if you do, then some kind of cache makes sense as well
#tantekJeena, the purpose for me would be to see how many "readers" would I lose by dropping Atom
#KevinMarksthat reminds me, I need to add a utm_ rewriter to my link parsing
#Loqivoxpelli meant to say: just such a thing like https requires any web based feed reader on https to load all post images through some kind of proxy – and if you do, then some kind of cache makes sense as well
#tantekand to see if there is any up/downward trend of people viewing my posts via Atom
cdevroe joined the channel
#KartikPrabhuknows exactly how many readers he loses by dropping Atom, maybe 2/3 and he might have already lost them because Feedly defaults to RSS which is now dropped!
#JeenaI'm just being silly, I know that there are good reasons for stats
#snarfedKevinMarks: maybe! try it. hard to say for sure, it'd take some digging. all i can say is it's coming from the servers, not caused by app engine
#KevinMarksmaybe I'm seeing wordpress blocking python's default useragent as a naive anti-bot tactic
#KevinMarksI though it might be a robots.txt thing with fetch() using the google crawler
#aaronpkspeaking of which, we should probably make sure we have clear specs for both publishing and consuming h-feed, h-entry, because right now it's kind of confusing
#tanteklet's start with documentation of the issues so we can make sure they're being addressed - i.e. anything you see as a source of "confusion
#aaronpkanother one is something Jeena and I talked about at IWC, is not wanting to publish a feed on your home page, but linking to one or more feeds instead
#dhalgrenrandom q: I see quite a lot of references to IndieWeb tech in the w3c's social web process. how is that relationship expected to go? I mean, will there be eventual interoperability with current oStatus, pump.io etc networks or what exactly is the roadmap/vision here?
#LoqiTypical use of the term roadmap does not refer to actual roads, maps, or maps of roads - instead, keep a personal priority ordered list of what you're specifically Working On, and for unsorted or vague desires, add them to an unordered Itches list, on your User page or your project's page https://indiewebcamp.com/roadmap
#tantekdhalgren, thus the roadmap question for you is, what's your personal site, and what's the very next thing you want to improve on it?
#dhalgrentantek: thx! I saw a gnu mediagoblin lecture yesterday, that's how I found these Social Web WG pages. I was simply plesently surprised by the apparent awareness of this community there; thought the federated social webs were going a rather separate way from the DIY spirit here.
#tantekwe're hoping to get out of the current /monoculture trap that so many other implementations / federated social webs have fallen into
#tanteks/federated social webs/federated social web attempts
todrobbins joined the channel
#Loqitantek meant to say: we're hoping to get out of the current /monoculture trap that so many other implementations / federated social web attempts have fallen into
#tantekwhether or not any particular monoculture can be patched to federated with other implementation is a good question. we can submit pull requests, and sometimes document attempts to do so, but so far progress with connecting /monoculture communities/efforts has been slow
#Loqitantek meant to say: whether or not any particular monoculture can be patched to federate with other implementation is a good question. we can submit pull requests, and sometimes document attempts to do so, but so far progress with connecting /monoculture communities/efforts has been slow
#dhalgrenindeed. though I see there really are tens of thousands of people that are free of silos using one of those federated social webs today, according to some partial census I found; that's alot of early adopters; I gather this community is much smaller, even though I'd have a far easier time implementing and interoperating with the IndieWeb.
#dhalgrenwell my own project, delayed by a babylon5 binge and moving to another place, exists on scratches of paper at this point -- I must correct that this summer :)
#tantekregarding the "tens of thousands of people that are free of silos using one of those federated social webs" - they're free as much as leaky lifeboats leaving the Titanic are "free" of an iceberg
#GWGtantek: I will have you know there was plenty of room in some of the those lifeboats
#tantekdhalgren - the problem being that each of those supposed "federated social webs" is its own monoculture due to being single-implementation based, which puts a lot of constraints / limitations on their growth / interop
#tantekthey're essentially disconnected islands, not a federation. and as long as that continues to be true, they will not be a practical alternative to large corporate silos.
#tantekdhalgren: do you have a personal site currently? even a domain name with one page?
#tantekand you have hit the proverbial nail on the head with "I'd have a far easier time implementing and interoperating with the IndieWeb" <-- that is the key
#tanteklowering the barrier to independent hackability, empowering people to interoperate without having to depend on a singular implementation
#dhalgrenGWG: nothing too fancy, but improving my haskell skills basically, and publishing the results as a reusable library and a blog (I've got too many writings I liked that ended up only as a deeply nested obscure reddit comment) - a statically generated blog skeleton using hakyll (that part I have but the code is terrible, was trying some idioms I didn't understand) , with frontend js using functional reactive programming (reflex lib prob
#LoqiReddit is a link aggregator and bulletin board site where community members may submit links and text posts, vote on the submitted entries, and post comments on them https://indiewebcamp.com/Reddit
#dhalgrenthat is, a backend for comments at least. Haskell is a pure functional programming language. I find it really elegant. Hakyll is a static site generator in it.
#tantekHaskell is a pure functional programming [[language]].
#gRegorLoveNo firsthand experience, but appears like the Reddit API will allow getting most things out. ^
#dhalgrenyy, I already have a few pages of myown comments in json on disk, it's nice
#dhalgrenre domain name, I'll take the official one my government is offering. no anonymity but first.last.from.hr , being free & secure (linked to personal ID). could out me if someone searches for my reddit profile on the basis of those sentences, but I figure its no longer such a big deal (hope?)
#KevinMarkshow does that work if 2 of you have the same name?