2014-07-10 UTC
# 00:00 aaronpk 2) tantek's server sends a webmention to mine for that link
# 00:01 JonathanNeal Most of the magic is working, but the certificate isn’t perfect. While the certificate is added to Mac’s keychain, and you can trust the certificate yourself, it is not trusted by default. I bet a simple configuration change could fix this, if any of you are up to finding the typo/bug.
# 00:01 aaronpk 3) my server notifies my phone via a push notification
# 00:01 aaronpk 4) I see the message, click the talky URL, and we're in a video call
JasonO joined the channel
# 00:02 aaronpk and of course this could all happen with private posts so the talky URL is only visible to the two of us
# 00:02 tantek aaronpk - great summary of indieweb based txt-notification plus webrtc call user flow!
# 00:02 GWG Should there be a markup for that?
# 00:03 aaronpk I suppose there could be markup for it, but as demonstrated by that use case it doesn't actually require special markup to work
# 00:04 bear I just filed an issue with the talky.io team asking if the lock-room feature can be enabled using only a single URL call (to allow a web form button to start one)
# 00:04 bear then aaronpk's step 4 above would have a URL to click that includes the locked room key
# 00:05 aaronpk although at that point I might as well just use the existing system with a UUID as the room name
# 00:06 aaronpk bear: feature request: if I visit a talky.io URL on my iOS device, provide a link that allows me to launch the iOS app to that room, since it doesn't work in iOS browsers right now
# 00:06 bear aaronpk - yea, I don't know if that works or not - part of why I filed the issue
# 00:09 donpdonp A Micropub is a very small, one room public house. - en.wikipedia.org/wiki/Micropub
# 00:10 GWG aaronpk: How would software interpret a communications request if no additional markup is added? The human could, but...
# 00:11 bear we already have added code to takly.io's web site to add a banner when being viewed by iOS browsers
# 00:11 bear so making it add an extra button or link for "hey, launch the app" should not be a problem
# 00:11 GWG I'm not sure if I am right on that
# 00:13 aaronpk GWG: I guess what's interesting to me is that we can do this right now without any additional markup.
# 00:13 GWG aaronpk: I see using Webmentions for communications requests...
# 00:13 GWG aaronpk: Can you think of use cases where you would want it to enhance the offering?
# 00:15 aaronpk maybe? but I'd be curious to test this out first and see how it works
# 00:19 tantek (though I think a specific "call" page would be more precisely scoped to what we're talking about )
# 00:19 Loqi tantek meant to say: (though I think a specific page would be more precisely scoped to what we're talking about )
# 00:19 bear hmm, the talky devs say that the banner that is shown only to iOS browsers has a "open in" link
# 00:20 GWG tantek: In your opinion, is a request for communications an invitation as defined in the wiki though
# 00:20 GWG I'm looking at invitation and event...
# 00:21 GWG Seems not quite applicable to something you want to happen immediately.
# 00:22 tantek and /phone become more of a "collection of features" page that lists various degrees of expectations of what features are in a "phone"
# 00:23 tantek (where we're going, we don't need phone numbers)
# 00:23 bret side note, sip server hosting is enterprizy and a pain
# 00:23 GWG tantek: Then you also need to define if call is inherently audio, video, or text
# 00:24 GWG tantek: Yes, but online, could be a 'video' call.
# 00:24 tantek since there is no video call "standard" (webrtc is just plumbing for now) - people call video calls by brand names, like "FaceTime" or "Skype" to most people means "video"
# 00:24 bear the signaling should be different than the actual connection
# 00:24 tantek GWG - "could be" doesn't matter when you're doing design in UI like that - majority / dominant cultural assumption wins
# 00:25 bear "I want to make contact" … "I can be reached by phone, skype, talky, sms"
# 00:26 GWG I'm just saying using a unified methodology with an optional medium seems to be simple.
# 00:26 GWG But...if I want to talk to bear on Skype...I could mean video, audio, or text
# 00:27 GWG Then bear goes and scrambles to find a web cam when I want tohave an audio chat
# 00:27 bear I worry about unified anything - it's a trigger word for me that i'm either over or under thinking something
# 00:27 KartikPrabhu GWG: does anyone say "lets have a text chat on Skype" or do they just say "lets Skype" ?
# 00:27 bear well, in my case it would be to scramble to find a shirt :)
# 00:27 tantek GWG - there are separate urls for skype video vs. IM
# 00:27 tantek see my URLs for People Focused Communication post
# 00:28 tantek bear++ for noting worry about unified anything meaning over/under thinking
# 00:28 bear it's one of the few positive things the UML wars of the 90's taught me :)
emmak joined the channel
# 00:35 GWG Okay, so I will accept the premise the form of communication is implied by the URL
# 00:35 GWG I withdraw that aspect of my comment.
# 00:44 GWG But I still think there is a variant of the invitation needed
# 00:44 GWG The URL for the chat service being the 'event'.
# 00:44 tantek GWG - you're going to have to document your use-case in more detail
# 00:44 tantek like aaronpk did above with his step-by-step how to make an indie-call
# 00:45 KartikPrabhu GWG: afais a call invitation can be an h-event with a u-url u-uid as the URL for the call
# 00:46 bret would that mean all my calls have permalinks?
# 00:48 GWG But couldn't the h-event be the talky session?
# 00:50 KartikPrabhu my point was we might not need any "new stuff" to make this call thingy work
# 00:52 GWG Is this a protocol or a use case?
# 00:55 tantek KartikPrabhu: I'd avoid jumping to format/protocol solutions until there's a documentation (on the wiki) of the use-case and user-flow supposedly being discussed.
# 00:56 tantek wouldn't want the leap to format/protocol to drive the interaction model
# 00:56 tantek which would be vice-versa from what's desired
# 00:56 tantek which is to have the interaction drive the formats/protocols necessary to make it happen, and nothing more
# 00:57 tantek GWG - a use-case describes what steps humans take when they're using a feature to interact with each other or their content. a protocol describes what data/code does both on publishing/sending and consuming/receiving ends.
# 01:08 GWG tantek: I'm agreeing with you. KartikPrabhu was talking about inventing a protocol. I was saying this seems to be an implementation of the existing format.
# 01:13 mko I worked on a concept for an app like that. Personal data marketplace.
KevinMarks2, caseorganic, snarfed and npdoty joined the channel
# 01:57 KartikPrabhu snarfed: is it posisble for bridgy to return the "like" URL on fav a tweet through bridgy instead of original post URL?
npdoty_ and npdoty joined the channel
gRegor` and KevinMarks2 joined the channel
paulcp joined the channel
scor joined the channel
# 02:56 KartikPrabhu snarfed: strange that Twitter does the same mistake with favs that G+ does with comments
# 02:57 snarfed it's generally agreed that g+ comments should have permalinks, but not at all that likes should have them
# 02:59 snarfed permalinks for discrete pieces of content like comments make sense, but likes aren't quite that
KevinMarks2 joined the channel
scor, brianloveswords and chrissaad1 joined the channel
# 04:17 bear indieauth/micropub question - with handling of webmentions I normally don't try to validate certs but I'm thinking for micropub auth I should only allow an authorization endpoint if the given domain has a valid cert. thoughts?
# 04:17 bear makes it a variable defaulting to false for now
# 04:26 bear hmm, that may not be possible as even aaronpk's own site fails https cert validation
addal joined the channel
# 04:53 JonathanNeal Well, I can now get back to it. I was obsessed with writing repeatable instructions on how to create trusted, local domains.
# 05:07 JonathanNeal Yes, but I finished this one, and now we can setup SSL for local development with wildcards!
# 05:31 KartikPrabhu JonathanNeal: fragmentions.js minified does not seem to work with async... suggestions?
tantek joined the channel
# 05:46 KartikPrabhu JonathanNeal: possibly. I recompiled fragmention.js and marginalia.js using closure compiler and it works
# 05:48 KartikPrabhu JonathanNeal: err closure seems to have messed up some of my code! any good minifiers you recommend?
# 05:48 bear aaronpk - this is the error that the python requests library gives me: requests.exceptions.SSLError: hostname 'aaronparecki.com' doesn't match either of '*.pin13.net', 'pin13.net'
# 05:49 bear notes that this is why he loathes SNI
# 05:50 aaronpk Yeah the https client you're using doesn't know how to do SNI
# 05:51 bear default python libs don't do modern TLS
# 05:51 aaronpk I feel like python libs and old versions of IE are the only things that don't support SNI now
# 05:53 KartikPrabhu bear: would appreciate if you could document getting pyOpenSSL to work... I had quite some trouble with it
# 05:54 bear KartikPrabhu - to solve this issue the following libs have to be present when requests loads: pyOpenSSL, ndg-httpsclient and pyasn1
# 05:54 bear what version of python? i'm using 2.7.6
# 05:56 bear it just worked with me doing a pip install of those
# 05:56 bear let me post the new requirements.txt so you can see it
# 05:58 bear ok, pushed the requirements.txt to my repo
# 06:00 KartikPrabhu does not do the fragmention thing and some marginalia are in the wrong place
paulcp joined the channel
# 06:00 JonathanNeal Incidentally, your referencing photos from non-HTTPS. Not sure what the internet gods think of that.
# 06:01 SamB JonathanNeal: they don't like it when you do it on an HTTPS page
# 06:01 SamB iirc even IE doesn't like when you do that ...
# 06:02 SamB and that was some OLD version of IE
# 06:02 KartikPrabhu JonathanNeal: yes that is a problem with avatars that I don't really host. itching to fix that bit
# 06:12 bear aaronpk - is this a none issue with indieauth calls: if you pass garbage data for client_id it just spins trying to verify everything
# 06:13 KartikPrabhu was ready to declare marginalia active on all posts, but ran into this JS quirkiness
gRegor` and cweiske joined the channel
# 06:47 cweiske did someone already plan a webmention+mf2-enabled comment hosting silo for websites that want to outsource their comment form and still be indie-compatible?
# 06:50 cweiske JonathanNeal, did you try to clone the public http url phorkie shows?
# 06:51 JonathanNeal cweiske: I was able to fork last night. What would you like me to do?
# 06:51 cweiske I mean a git clone on cli with the URL phorkie shows
# 06:52 KartikPrabhu JonathanNeal: cool! heading to bed now (early morning physics meeting) ... might bother you about fragmention.js weirdness tomorrow though
# 06:52 JonathanNeal Great. I’m starting to feel better so it would be nice to dive into that.
KartikPrabhu, paulcp, jsilvestre and michielbdejong joined the channel
eschnou, paulcp, petermolnar, pfefferle, Sebastien-L, shepazu and kylewm joined the channel
xtof joined the channel
# 08:26 bear \o/ successfully do a roundtrip indieauth login, auth and token validate with python flask
# 08:26 bear this was the precursor to me getting micropub support for my static site
mko joined the channel
# 08:29 bear KartikPrabhu - well, two weeks real time but I only worked on it some during IWC and tonight :)
# 08:29 bear cweiske - I need to remove the hardcoded urls and make them config driven - but yes
# 08:30 bear cweiske - i'll post my badly hacked code as a gist
# 08:30 cweiske i'd only try if I can login with my indieauth server
# 08:31 bear KartikPrabhu - this is going to be part of Dainin
# 08:31 bear cweiske - that's the hacky part - right now it only works with Indieauth
# 08:32 cweiske so you don't do authorization_endpoint discovery yet
# 08:32 bear it does - that is part of the changes I just made to Ronkyuu
# 08:32 bear to add two helpers for that - one to discover and one to make an auth token validate call
# 08:33 bear I would love feedback on how i'm doing with Flask
# 08:42 bear the index.html file needs to be in a templates/ dir relative to where the python file is placed
sankha93 joined the channel
# 08:43 bear it may - never thought of trying that
# 08:44 cweiske I installed flask and tried to start listener.py but that fails because ronkyuu is missing.
# 08:45 bear yea, it's a github only thing right now - one sec
# 08:47 bear this is still a big work-in-progress - need to start cleaning up the non-bear user experience
# 08:48 bear i'll update the requirements file now
# 08:48 bear oh wait - that is in requirements.txt
# 08:49 bear python setup.py install should have run thru the requirements.txt file and installed all dependencies
# 08:50 bear no - your finding all the dev-working-local glitches
# 08:54 bear what is missing for this is adding a url route to allow the auth token to be stored and recovered for agents to use
# 08:54 cweiske and it fails badly if I input the domain only as opposed to my url
Sebastien-L joined the channel
fofr joined the channel
# 08:58 cweiske but I have no idea if it does anything when I click sign in with my url
# 08:59 bear it should call out to your auth endpoint and let you validate
# 09:01 bear and if I remove that param from the auth request I now get: OpenID error: Unable to discover OP Endpoint URL
# 09:01 cweiske ok. I don't support authorization, only authentication
# 09:01 bear I have more work to do then to split these out then
# 09:02 cweiske if you're longing for micropub, then authorization makes sense
# 09:02 bear yes, micropub is the first itch to scratch for this :)
paulcp joined the channel
# 09:02 bear but my goal is to make ronkyuu support all types of indieauth
SamB__ joined the channel
friedcell joined the channel
# 09:08 bear changing response_type to id now gives: OpenID error: Unable to discover OP Endpoint URL
# 09:08 bear but this is probably from me doing localhost urls
# 09:08 cweiske it provides an indieauth api, but uses openid in the background
# 09:08 cweiske so I can auth with my openid server at indieauth sites
# 09:10 cweiske yes, and you have no openid link in your hompage's html head
# 09:10 cweiske while I have <link rel="openid.server" href="http://id.cweiske.de/" />
# 09:11 bear i.e. not front-end web development ;)
# 09:16 bear I updated the gist - but it's line 102: urllib.urlencode({ 'me': form.domain.data,
# 09:21 cweiske wireshark also does not show anything except the GET to cweiske.de
# 09:21 bear for me it brought up your SimpleID loging
# 09:24 bear yea, fixing that by having it read from config file - one sec
# 09:25 bear hehe - no worries - i'm loving the feedback
# 09:35 bear ok, added a config file and updated the code to load it - it now gets client_id and baseurl from config
# 09:36 bear i'm heading to bed -thanks for testing it and apologies for it being so rough to use
# 09:41 bear Loqi tell KartikPrabhu i've pushed a new version of ronkyuu to pypi - please submit a PR if you have any local changes so I can push that version if needed
adactio and glennjones joined the channel
paulcp, terminal2, bear, pfefferle_, jsilvestre and xtof joined the channel
scor and paulcp joined the channel
scor joined the channel
pbeaulieu, eschnou, xtof, Sebastien-L, paulcp, jsilvestre_, h1ro and pfefferle joined the channel
paulcp and gRegor` joined the channel
snarfed joined the channel
# 14:28 gRegor` He does appear to be selfkoolaiding
# 14:30 petermolnar I'd be glad to hear you opinion on it before proceeding with the actual syndication work
# 14:30 Loqi Ok, I'll tell them that when I see them next
# 14:32 Loqi GWG: petermolnar left you a message 2 minutes ago: https://github.com/petermolnar/posse - it does nothing apart from building syndication meta boxes & saving the values, but it took me a while to come up with an approach that's flexible enough yet still relatively usable
# 14:39 petermolnar ah, for a moment a thought Cambridge, UK, that would have been lovely :D
brianloveswords, chrissaad, wolftune, eschnou and paulcp joined the channel
indie-visitor joined the channel
# 15:31 Loqi Welcome, indie-visitor! Set your nickname by typing /nick yourname
# 15:33 aaronpk any wordpress people getting close to wanting to make a micropub endpoint to create posts within wordpress?
pbeaulieu joined the channel
# 15:49 aaronpk hopefully with some language diversity so it's not just a javascript week ;)
Sebastien-L and tilgovi joined the channel
caseorganic joined the channel
# 16:08 donpdonp if the goal is a new vhost with ssl of name 'localhost', it seems that entire procedure could be automated down into a shellscript
# 16:15 aaronpk ok how about changing the july 30th homebrew website club into more of a hack night for dev week?
paulcp joined the channel
# 16:19 donpdonp JonathanNeal: yeah, 'sed' is the usual way but such automated config munging is risky.
# 16:19 donpdonp creation of the self signed cert, and the vhost config file could be done w/o subsitutions
# 16:27 donpdonp i just mean that the vhost for localhost, is a new seperate apache config file
# 16:27 donpdonp so no risk of mangling existing config files with just that step
emmak, tilgovi, snarfed, paulcp, jschweinsberg, pbeaulieu and mayuresh joined the channel
# 17:03 mayuresh hello ... :-)
# 17:03 mayuresh is kylewm around?
# 17:04 mayuresh i did most of the rel=me things to my web page. :-)
# 17:04 mayuresh and yeah, i could sign in to indiewebcamp wiki.
# 17:04 mayuresh thanks for the help kylewm.
# 17:05 mayuresh and the cool stuff on the indiewebcamp wiki
KevinMarks joined the channel
# 17:07 mayuresh hi kevinmarks
cweiske, paulcp and dariusdunlap joined the channel
KartikPrabhu, j12t, dariusdunlap and danfowler joined the channel
# 18:57 kylewm !tell mayuresh you're very welcome, and all credit for indiewebify.me goes to barnabywalters and bnvk!
# 18:57 Loqi Ok, I'll tell them that when I see them next
mayuresh joined the channel
# 19:09 Loqi mayuresh: kylewm left you a message 11 minutes ago: you're very welcome, and all credit for indiewebify.me goes to barnabywalters and bnvk!
# 19:12 mayuresh bear: thanks for the ++ :-)
# 19:12 mayuresh do take a look at kathe.in
# 19:16 bret aaronpk++ for the dev week hack night
# 19:16 aaronpk !tell tantek I want to run some sort of indieweb event during http://pdx.devweek.org and am considering co-opting Homebrew Website Club to be more of a hack night or something. any thoughts on how that might work in conjunction with SF? or just not do joint that day?
# 19:16 Loqi Ok, I'll tell him that when I see him next
paulcp, tilgovi and gRegor` joined the channel
# 20:06 aaronpk archivebot is now crawling its way through the IRC archives
chrissaad1 joined the channel
# 20:17 bear I would say "time for an updated robots.txt" but most of them ignore it :/
# 20:17 aaronpk it's been pulling 3-5 requests/second since yesterday
# 20:18 aaronpk yeah robots.txt already tells bots to ignore wiki/* so that they don't recurse into the meta wiki pages and page history and such
# 20:18 bear archivebot is pretty greedy - but for reasons I can fully grok
# 20:18 bear it wants *everything* so it can be solid copy
# 20:19 aaronpk yeah. i just feel bad about giving it redundant content
# 20:19 aaronpk like every page links to a login URL that is unique to that page because it contains a parameter like returnTo=/events
eschnou joined the channel
# 20:24 kylewm aaronpk: how do you monitor that sort of thing?
# 20:27 aaronpk i'd publish the access logs, but then that would really throw ArchiveBot for a loop
brianloveswords joined the channel
# 20:30 donpdonp there's probably a web page with n digits of pi and a link to the page with n+1 digits
# 20:32 kylewm interesting that ArchiveBot ignores robots.txt
# 20:32 kylewm I have an old domain that is owned by someone else now who has a robots.txt file
# 20:32 kylewm and archive.org won't show any history for it as
hmans joined the channel
KevinMarks joined the channel
# 20:37 kylewm hi hmans, saw your tweet this morning about maybe supporting webmention, thumbs up!
# 20:38 gRegor` Welcome, hmans
# 20:39 hmans I built something and realized it's pretty much 100% in line with webmention, so I'm going to switch the whole thing to it very soon.
# 20:40 gRegor` Will it work with outside sites not running pants?
# 20:40 gRegor` Nice software name, btw. Makes for some fun sentences, haha
# 20:40 hmans Yeah, my implementation of Webmention should be 100% happy to talk to non-Pants sites and vice versa.
# 20:41 aaronpk You should name components things like Sweat Pants, Under Pants, etc
# 20:41 hmans I'm building a command line client called Panties... which is horrible.
# 20:41 gRegor` The minimalist theme for it: No Pants
# 20:42 hmans I should stop naming my projects after the first thing that comes to mind.
bnvk joined the channel
# 20:42 gRegor` How many people are running pants?
# 20:43 hmans I think there are 2 or 3 instances out there not run by myself right now. Hey, I wrote the first commit 2 weeks ago :P~
# 20:43 SamB not too many people are pants
# 20:43 SamB nevermind *running* pants
paulcp joined the channel
# 20:43 gRegor` Oh wow, I didn't realize it was that young. But hey, that's pretty good for two weeks old!
# 20:44 hmans I haven't been pushing people to install their own instances (the README is just a big warning right now), and I've been manually adding friends to my own instance. It's just too early for anything else.
# 20:47 hmans I haven't added any rel=me support yet, though :/ It's coming.
# 20:51 kylewm 2 weeks and already have users begging you not to change the specs, pretty good :)
# 20:51 kylewm I'm curious why "The part of the Wm spec I'd have to deviate from is where the recipient checks if their link is inside the referred document." ?
# 20:52 hmans Yeah, turns out I don't actually have to -- it's the part where the recipient checks if the link to their article is actually present on the URL provided by the sender. (I understand why it needs to happen, but Pants does a tiny little thing on top of that -- not a big deal.)
# 20:53 hmans Friendships on Pants don't need to be mutual
# 20:53 hmans so when you add someone to your friends list, your Pants will ping your friends when you post something new. Of course, your friends may not have YOU in THEIR friends list, so they pretty much decide on the fly on how to react to your ping.
# 20:54 snarfed hmans: understood. skipping the link check is generally fine. it's common practice for us here, but definitely not required
# 20:54 hmans (In the current implementation, the referenced URL will always be fetched, but then the recipient will check the URL against its own friends list)
# 20:54 snarfed the spec does say it's a SHOULD, but that may be a bit too aggressive
# 20:54 aaronpk that ping you mention is more like the ping of the PuSH spec
# 20:54 hmans In addition to that, stuff gets pushed without including a reference to another posts, so yes @aaronpk, I should look at that, too
# 20:55 hmans I only have very shallow understanding about PuSH, but it was my impression that it focused on hub-style structures? Apologies if I'm wrong, I'll need to do more reading.
# 20:56 aaronpk the original PuSH spec (which was tied to XML) did emphasize the hub structure
# 20:56 aaronpk but with 0.4 allowing arbitrary content to be pushed around (html or json, etc) it's a bit more useful now
# 20:57 aaronpk I'm not quite sure how the hub architecture would map onto yours, but I suspect it's compatible
# 20:59 hmans Q regarding rel=me/indieauth: what if I have multiple web identities (domains, like hmans.de and hmans.io), but I can only link to one of them from one of the indieauth-supported networks (eg. Twitter)?
_axx joined the channel
# 21:00 hmans I don't mind using just one of my domain as "my identity" when logging in to eg. the indieweb wiki, but I'm planning to allow pants users to log in on any other pants site (using their own site), so indieauth/rel=me looks pretty much exactly what I should be looking at.
bitraten joined the channel
scor joined the channel
# 21:16 aaronpk I don't know if I would call it sudden. Unless you're thinking in very large timescales.
# 21:17 mko Realistically, a lot of it is spawning out of the advancements in Node.js
# 21:19 mko When a new feature of Node.js enables something that wasn't previously possible, the Node.js and JS communities as a whole are so large and wanting *more* out of their skillsets that *someone* will build something with the new feature that already exists in other languages just because they can.
# 21:20 SamB KartikPrabhu: Mozilla had large parts written in JS even back in the late 90s, didn't it?
# 21:21 mko Not really. Node.js describes itself as a platform for a reason. It's an entirely new model of programming logic within JavaScript (not that Node.js is that new of a concept from a programming language theory standpoint, just that it's new in JavaScript).
# 21:21 SamB mko: what sort of model is it?
# 21:22 SamB say, what non-JS frameworks does it resemble
# 21:22 gRegor` "why this sudden obsession for building everything in JS?" asked the marginalia.js guy. ;)
# 21:22 SamB gRegor`: that's different
# 21:22 gRegor` (But yes, JS is being . . . overused)
# 21:22 mko The model is that it's a non-blocking model of data input/output that relies on an endless event loop.
# 21:22 gRegor` I know. I'm just teasing him.
# 21:23 SamB gRegor`: when you want to run code in the browser, you really don't have many choices
# 21:23 gRegor` Definitely a difference between Angular and all these other JS-hotness frameworks and marginalia.
# 21:23 SamB mko: hmm, so the new to JS part is ... not blocking?
# 21:23 aaronpk someone should write a Javascript interpreter for PHP so we can write PHP code and run it in the browser
# 21:24 SamB or that you get to interact with the system
# 21:24 SamB rather than being heavily sandboxed
# 21:24 SamB someone actually DID THAT?
# 21:24 SamB why the hell would somebody smart enough to do that be dumb enough to think it was a good idea?
# 21:25 SamB PHP seriously needs to DIAF
# 21:25 aaronpk no, people who write bad PHP should learn to write better code
# 21:25 SamB it might have started out with a good idea, but sounds like it went off the rails pretty quickly (no, not Ruby on Rails rails. Train rails!)
# 21:25 gRegor` invites everyone to #indiechat
# 21:26 aaronpk i've seen some pretty terrible ______ where blank is node, ruby, etc etc
# 21:26 SamB mko: PHP has SOO MANY unresolved inconsistancies
# 21:26 kylewm KartikPrabhu: do you know Atwood's Law? "any application that can be written in JavaScript, will eventually be written in JavaScript."
# 21:27 SamB and did you hear that they used to choose names for things so that they would not all have the same length, because their hash function was length based?
# 21:27 SamB and many of those names are still here even though they switched to a less stupid hash function?
# 21:27 SamB and how about those exceptions, hmm?
# 21:27 SamB kylewm: C -> asm.js, at least
# 21:36 gRegor` Hehe. #indiechat is booming now!
# 21:39 SamB so, what sites can I link to for the actual authentication?
# 21:39 SamB will launchpad work?
# 21:40 aaronpk the list of supported OAuth providers for indieauth.com is listed there
# 21:40 aaronpk I haven't added launchpad, but I probably could pretty quickly if it would be useful to enough people
# 21:41 aaronpk they have to be added one by one because each silo requires registering an API key
Crossdiver, caseorganic, danfowler, j12t, petermolnar, brianloveswords, sdboyer, KevinMarks, the_merlin, bear, peat, dariusdunlap, glennjones, muhh, irdan, saurik_, donpdonp, arlen, iboxifoo, hadleybeeman, bigbluehat, CaptainCalliope, ngoldman, piney0, amblin, hugoroyd, tallpaul_, mattl, benward, voxpelli, jden, rknLA, galfert, finchd, mlinksva, jtzl_, rhiaro, catsup, lukebrooker and scor joined the channel
brianloveswords joined the channel
scor and mko joined the channel
# 23:53 bret how do I enable the experimental wiki theme that came out of 2014?
KartikPrabhu joined the channel