2015-10-16 UTC
uranther, kronda, arlen, j12t, mdik, rabble_, almereyda, zachdonovan_, Vendan, scor, wolftune, brucewang, fahrstuhl, acegiak, rhiaro, renoirb, GWG, hidgw, raucao, braidn, Jay-, JasonO-, zachdonovan, daf, strangehill, catsup, FergusL, M-kegan, nitot, finchd-also, joskar, Phyks, moredhel, bol, Garbee, tallpaul, tommorris, jden, bergie, benward, Phae, KitB, anm, Unifex, ventilateur, neil1, bear, Tristitia, tbrb, halorgium, yakker, iboxifoo, _6a68, JonathanNeal, MMN-o, shalkydri, mattronix, danlyke, modem, danfowler, rascul, dietrich, camerongray, hmans, saurik, zz_tridnguyen, blueyed, voxpelli, sivoais, sandro, cjk101010, martinBrown, aaronpk, micro_, smcgregor, Jeena, KartikPrabhu, tantek, kerozene, kylewm, ehlovader, j4y_funabashi, lordabdul, michel_v, brianloveswords, paulproteus, ben_thatmustbeme, snarfed, 14WABFSH1, oddvar, bret, jacus, Kongaloosh_, gRegorLove, jonnybarnes, CaptainCalliope, Leeky, kegan, wagle, Guerillero|BNC, ramsey, minsky, rektide, R2ZER0, strugee, marjolein, MrClaw, M-Kodo, Zegnat, reidab, MylesBraithwaite, sensiblemn, ben_thatmust and sivoais_ joined the channel
nitot_ joined the channel
[ben_thatmustbem joined the channel
# 00:30 [ben_thatmustbem hmm, netsplit
j12t joined the channel
XgF joined the channel
snarfed joined the channel
# 00:51 ben_thatmustbeme hmm, i might have them marked up wrong, i''m getting a blank coment on a fetched comment
# 01:24 ben_thatmustbeme ever since i started using a code sniffer i now have this urge to go through and make rulesets for every project i come across
# 01:26 GWG I have the urge since I discovered it to make all my plugins comply with WordPress standards. There's a ruleset for it
# 01:27 GWG I even made the Micropub plugin compliant, but I don't think snarfed will accept the Pull Request unless it comes with unit tests.
sivoais joined the channel
bigbluehat, moredhel, scoates, shiflett and [number5] joined the channel
gRegorLove, tantek and [snarfed] joined the channel
# 03:37 [snarfed] indiewebify.me isn't very well maintained. we should probably wind down recommending it
# 03:38 tantek we used to have multiple people who could update the domain - is that no longer the case?
# 03:39 tantek maybe only two? last I remember both barnabywalters and bnvk did
# 03:40 tantek is the site out of date compared to the repo?
sammachin1, MylesBraithwaite and oddvar joined the channel
# 03:52 tantek snarfed, remind me again how github is better than a wiki?
R2ZER0 joined the channel
# 03:53 tantek is there anyway on github to say, let anyone I follow commit changes to this repo?
# 03:53 tantek without having to "add" maintainers or editors or whatever one-by-one manually to every single repo?
# 03:53 bear no, but adding collaborators with write permission is very easy
# 03:54 tantek bear - site unmainteance death by a thousand "easy" cuts
# 03:54 KartikPrabhu i wouldn't want people I follow commiting changes to my code by default
# 03:54 tantek KartikPrabhu: but we are even more liberal about changes to the wiki!
# 03:54 tantek the point is, indiewebify.me is not "your code" or any of ours - we should be treating it as a shared resource like the wiki
# 03:55 tantek and that's why the code is unmaintained - because people treat them differently
# 03:55 tantek neither snarfed, nor me, nor you can edit nor update indiewebify.me right now
# 03:56 tantek code should be community by default, and only if you're particularly paranoid should you then have to make your own you-only-controlled fork of that repo
# 03:56 bear github is a code wiki but with a more structured permissions model
# 03:56 tantek you mean, a more manual buttons to push to do anything permissions model
# 03:57 bear no, it could be automated but the permission is about who can submit changes because code is, on purpose, different than writing/docs
# 03:57 tantek whereas everytime we create a new page on the wiki - we have to do ZERO permissions maintenance work to allow people to edit it
# 03:57 tantek I think the purpose is more often a mistake than not
# 03:57 tantek it's a bad assumption that results in neglect
# 03:57 bear i'm sure some folks have had to be banned for vandelism
# 03:57 bear that is probably a byproduct of the auth
# 03:57 bear which, is how I would solve this problem
# 03:58 KartikPrabhu not a lot of people have an incentive to vandalise the indiewebcamp wiki. Wikipedia gets a lot of that and has to have mods and such
# 03:58 tantek don't care - point stands. our wiki is ZERO Permissions maintenence
# 03:58 bear if you have auth on the wiki and have a github id then have the bot add them as a collaborator
# 03:59 tantek KartikPrabhu: not a lot of people have incentive to vandalize indiewebify.me
# 04:00 bear what is the repo and how do I get access so I can help solve this issue
# 04:00 tantek it can be extended to any community group on github, of which there are many
# 04:01 bear so it looks like some of the regulars have committed to it...
# 04:01 tantek manually maintaining permissions and all that is for enterprise bureaucracy nonsense
# 04:01 tantek not for agile volunteer-based open-source etc. communities
# 04:01 Loqi Ok, I'll tell them that when I see them next
# 04:02 tantek point being I'd rather have someone commit a mistake and then someone else be able to undo it, than that someone not be able to commit the mistake in the first place
# 04:02 bear yes, that I agree with - I recently started to do that with some of my older open source libs
# 04:02 bear and found that I haven't had to roll back many changes - and a few of them I just explained why and they did the change
# 04:03 tantek because even if you stumble upon one, you have to ask for permissions (social barrier), or submit a pull request (psychological barrier of what if my work is rejected), or wait (psychological impatience barrier) etc.
# 04:04 bear I wouldn't say that it's completely crap - but in some cases it is a higher barrier for community than is required
# 04:04 tantek it's fine for top down hierarchical orgs / enterprise
# 04:05 bear which is where it got all of it's defaults - yep
# 04:06 [snarfed] I'll avoid the wider debate... but in this case, barnabywalters and bnvk have at least been around occasionally, and friendly, so if anyone did actually step up and ask to take it over, or get added to the hosting provider etc, I'm sure they would have
mblaney joined the channel
# 04:09 kylewm bear: invited you to the github group, i'm surprised you weren't already in there!
# 04:11 bear hmm, I suspect this may be a mf2 parser upgrade issue - but I won't know until someone check
# 04:11 bear i'll file an issue about the error the above person got
# 04:12 kylewm I've heard of maintainers that give commit access to anybody who submits a good pull request
# 04:14 bear I now give access to anyone asking or filing issues for some of my libs
# 04:15 Loqi Ok, I'll tell them that when I see them next
# 04:17 [snarfed] also worth noting: the long pole for letting communities manage apps like this usually isn't github, it's the hosting provider
# 04:18 bear yea, getting apps deployed is so much harder than it should be (but I also no why it is and that's why i'm doubly frustrated)
# 04:20 [snarfed] deploy-on-push and gating on continuous builds are both steps in the right direction
# 04:21 bear k, heading to the pub - my fav bartender tells me she has just tapped a new chocolate porter
# 04:34 mblaney speaking of problems, did anyone notice indiewebcamp.com/irc/today isn't updating?
# 04:37 mblaney kylewm, I've been working on my h-cards and was just looking at aaronpk
# 04:43 mblaney the strange thing is it does show me when people join, just not any real messages
# 04:47 kylewm it looked like it didn't make the websocket connection at all
tantek joined the channel
# 04:49 gRegorLove I'm working on figuring out updating syndication links after async Bridgy Publish. Tricky.
j12t joined the channel
# 05:00 mblaney kylewm I've been trying to debug pushstream.js, and I have a feeling it's actually working fine.
# 05:00 mblaney at a guess I would say the messages aren't being pushed from the server.
[aaronpk] joined the channel
# 05:02 [aaronpk] Well that's odd, especially since they're still coming to Slack just fine
# 05:05 kylewm ok maybe I am back in the world of the living. my messages weren't showing up in the log at all for a bit
# 05:06 [aaronpk] I wonder if it is somehow related to the netsplit earlier
[kylewm] joined the channel
kylewm joined the channel
# 05:20 [aaronpk] the damn server Loqi is on is so old it can’t talk to indiewebcamp.com SSL anymore
# 05:28 mblaney aaronpk++ it's smart that you got it working though! :-)
tantek, j12t, eschnou, nitot, loic_m, Jihaisse, gRegorLove, lewisnyman, petermolnar, elima_, cweiske and tvn joined the channel
sanduhrs and j4y_funabashi_ joined the channel
j12t joined the channel
# 08:33 j4y_funabashi_ and wondering if there is a way to mark that up so json parser will see the link as navigation to the next page of entries?
# 08:35 j4y_funabashi_ yeah sorry by json parser I mean a microformats parser (which gives you a json version of the page)
elf-pavlik joined the channel
# 08:36 KartikPrabhu yeah rel-prev and rel-next are those things. I don't know any feed reader that uses them though
# 08:43 KartikPrabhu oh I remember why I removed them. I don't know if prev = older or prev = previous in reverse chonological order
# 08:45 j4y_funabashi_ yeah I was just wondering about it because alot of the apis I have used include nav links
friedcell joined the channel
# 08:48 j4y_funabashi_ Yeah I agree these pages should be expanded a bit, I guess the order is up to us as publishers?
# 08:50 KartikPrabhu again, I'm the minority report on this. Most people are fine with the existing rel-next spec
# 08:50 j4y_funabashi_ although "next document in a linear sequence of documents" seems clear enough to me
# 08:51 Loqi KartikPrabhu meant to say: in fact tomorrow, I'm going to make all posts on my site in chronological order
j12t, cweiske, alexhartley and hardfire joined the channel
alexhart_ and [kevinmarks] joined the channel
# 10:48 [kevinmarks] Is the problem with indiewebify.me not so much a code access or as a deployment one
j12t and kerozene joined the channel
j12t, modem_, bol, friedcell and kerozene joined the channel
# 12:28 petermolnar while FB content may not be that important to the world, Instagram is, so from hereon I will try to spread the PESOS word
hs0ucy and j12t joined the channel
# 13:07 petermolnar I've recently realized I can drop 2/3rd of my custom icons font file, since unicode covers a significant amount of them now
# 13:08 GWG I am thinking of accomplishing something this weekend.
# 13:09 GWG Maybe I will finish adding Microformats parsing to Post Kinds. I have been putting that off.
# 13:10 cweiske did the unicode standard change such a big deal in the last months?
hs0ucy joined the channel
joskar, j12t, elf-pavlik, adactio and mdik joined the channel
hs0ucy, j12t, snarfed and mlncn joined the channel
alexhartley joined the channel
ttepasse joined the channel
# 14:59 aaronpk I'd be willing to set up automatic deploys for indiewebify.me if we moved it to my hosting account
# 14:59 aaronpk it'd be nice to have that be more up to date, and have more people be able to contribute to it
LanceyWork joined the channel
# 15:10 bear kevinmarks - for this specific case it may be a code even if only in the fact that it didn't handle a parsing error properly
# 15:11 bear aaronpk - I can definitely help with that, I will be having freetime over the next week or so
# 15:12 aaronpk cool. i've done automatic deploy from github hooks a couple different ways
elf-pavlik and j12t joined the channel
# 15:12 bear since it's a public repo you can use circleci or travis to do the deploy
indie-visitor joined the channel
# 15:13 Loqi Welcome, indie-visitor! Set your nickname by typing /nick yourname
# 15:15 bear I can't imagine it's a very memory hungry site
moredhel joined the channel
# 15:17 aaronpk looks like on the free tier it must be not running for 6 hours a day
moredhel, friedcell and [snarfed] joined the channel
# 15:38 aaronpk hm why isn't app engine one ofthe supported providers on travis-ci?
# 15:39 [snarfed] i also remember a push-to-deploy article on the Google cloud blog or nearby a while back
tvn_, scoates, j12t, e-lima, shiflett and lewisnyman_ joined the channel
wolftune and snarfed joined the channel
# 16:39 ben_thatmustbeme now i just need someone who supports salmentions to comment on something of mine, and someone else to comment on that
# 16:42 snarfed ben_thatmustbeme: if you connect u-url and u-in-reply-to, backfed facebook threaded comments will also work
# 16:45 aaronpk hmm thinking I should promote silo.pub on ownyourgram as a way to connect to those
tantek joined the channel
# 16:51 aaronpk interesting limitation of JSON post requests, there isn't a good way to also include a file
e-lima, gRegorLove and [shaners] joined the channel
# 16:56 [shaners] aaronpk: Could Loqi be re-written as a Hubot? (to be less Rubey and Goldbergy)
nitot_ joined the channel
# 16:57 aaronpk I will probably rewrite Loqi with lita.io since it seems to be a pretty flexible framework
# 16:58 aaronpk I have a lot of feels about the architecture of IRC bots
j12t joined the channel
# 17:02 bear I think I have at least 3 different python chatbot frameworks in the dusty corners of my harddrives
# 17:02 aaronpk i started rewriting Loqi in a node.js framework a friend made. that's the Loqi running everywhere except freenode now
# 17:04 bear my most recent attempt made use of redis for it's pubsub and storage
# 17:04 bear I've always wondered how php would perform if used as a command line language
# 17:05 bear redis pubsub slinging json blobs == super simple and very flexible
# 17:06 aaronpk you can even go pretty deep with process control and forking in php
# 17:06 bear yea, I thought it would be - all of the things that I don't like about php are all related to mod_php and it's web usecases
tvn, shiflett and snarfed joined the channel
# 17:29 snarfed anyone know why the indieweb github org's membership is private?
# 17:29 aaronpk probably because that setting was added after the org was created
# 17:30 bear most orgs leave it up to the members to advertise their status
# 17:30 snarfed indieweb is pretty explicitly open in general though
# 17:31 voxpelli still good to be able to chose if you want to advertise it on your profile or not
lewisnyman joined the channel
# 17:31 bear IMO it should be up to the individual
# 17:31 voxpelli (me is not a member of it though so doesn't affect me)
# 17:32 bear do what - make your membership public?
# 17:32 rhiaro you go to the list of members and the setting is next to your name I believe
# 17:33 bear you will notice that owner's can change other normal members but cannot change other owners visibility
# 17:33 bear this is all part of the org security changes that they have been rolling out lately
# 17:33 [shaners] I’m in town for a couple few days (currently at MIT Media Lab).
# 17:34 gRegorLove I can't seem to find the conversation about mf2 for abstracts linking to the paper in PDF format. Currently I'm using u-url for the abstract's page. I see KartikPrabhu uses u-url to link to the PDF, though.
# 17:34 tantek shaners - yes, rhiaro is basically right there at MIT
# 17:36 tantek shaners - yeah, you and rhiaro should definitely meet. she knows the good local vegan places.
[kevinmarks] joined the channel
# 17:36 [kevinmarks] Would using u-content make sense? Hm. The name/summary/content split is trickier when the content is in a less parseable form.
tantek joined the channel
# 17:39 tantek kevinmarks: what's the use-case? or are you still talking about the citing someone else's paper thing?
alexhartley joined the channel
# 17:41 [kevinmarks] I thought gregorlove was talking about abstracts linking to pdf papers
# 17:46 tantek rhiaro: interesting that your posts still display BST when you're in EDT :)
# 17:47 tantek I don't even know what the right answer is. Hence more of a question ("interesting") than any kind of judgment.
nitot joined the channel
# 17:47 rhiaro I create posts with a timezone, but somewhere in the heart of my site it gets nuked to a timestamp before it comes back out again
# 17:48 rhiaro Though I'm not sure if I timestamp it before I store it and it gets reset to BST anyway
# 17:48 tantek most systems nuke that info because programmers tend to be all about the Z
# 17:48 voxpelli best time of the year to fix it now – when the US and EU exits summer time at different times
# 17:49 rhiaro and cos my server is in the UK, when I don't specify date for a post it defaults to BST
# 17:50 tantek voxpelli: yes, a good time for testing. it's like two weeks right?
# 17:51 voxpelli yeah, Sunday next week for EU and Sunday the week after that for US
# 17:52 voxpelli great time for US/EU coop, an entire week of an extra hour of overlap :)
elf-pavlik, j12t and mlncn joined the channel
# 17:58 aaronpk lol yeah proper timezone support is the first thing I did on my site
# 17:59 aaronpk i don't see what's controversial about the timezone the post is authored in is the one that is displayed
# 17:59 tantek e.g. from a UX perspective, still curious which is better, showing timezone in the perspective of the reader, or the creator of the content? and when/why?
# 18:00 tantek from the reader's perspective, it's *harder* for them to do the timezone math to figure out if the post was minutes ago, hours ago, etc.
# 18:00 tantek and *confusing* when they see posts from the future
# 18:00 aaronpk that's better than the majority of the defaults which is timezone information is dropped, or hardcoded to one timezone always
# 18:00 tantek just questioning "proper" as that makes it sound like a provably solved problem
# 18:01 aaronpk i might add a little bit of javascript that lets viewers show all timestamps in their browser timezone
# 18:01 aaronpk well frankly I do think it's solved, since this way I don't lose any timezone information
# 18:01 tantek I've considered JS for time since display (and parenthesized creator timezone absolute time)
# 18:02 tantek aaronpk: no the presentation problem is not solved
# 18:02 voxpelli if the time matters it should, if possible, probably be a relative time rather than an absolute one then
# 18:02 tantek voxpelli: relative time alone is worse. e.g. screenshots, old tabs etc.
# 18:02 tantek Pretty sure aaronpk has a canonical article on this :)
# 18:03 voxpelli yeah, absolute ones are for reference, relative ones for relating to
# 18:05 tantek was going to say, ever since the mid 2000s beginning of the end for the XML stack, there's been a lot of loose abandoned x's floating around the nets.
j12t joined the channel
# 18:09 aaronpk i'm not convinced i'd want to write on that with a sharpie tho
# 18:11 aaronpk anyway yeah I agree that displaying local times is a great job for a reader
cleverdevil joined the channel
KartikPrabhu joined the channel
# 18:14 aaronpk apparently they have an onclick event for the state dropdown so it didn't actually register that
# 18:15 aaronpk told me to enter a state when I submitted it, but then the form came back prefilled with the state chosen
nitot joined the channel
moredhel and benwerd joined the channel
[kevinmarks] joined the channel
# 18:28 benwerd Yes, her domain slipped - she's going to repoint a new one
# 18:29 GWG I haven't worn a shirt with anything on it in years
# 18:29 kylewm oh i remember her mentioning that a while ago, it was going to be like $250 to renew?
snarfed1 joined the channel
# 18:35 GWG benwerd: Is it available in a hat? Button? So on?
# 18:35 aaronpk appengine is not returning any error messages but is clearly not doing what it should be doing
# 18:36 snarfed1 aaronpk: happy to look if you want
# 18:36 aaronpk trying to run `dev_appserver.py .` it looks like it's running, but fails with "ERROR:root:php failure (255) with:" and then the HTML of the page
# 18:37 snarfed hmm, ok. sounds somewhat php specific, so i may not be that much help
# 18:37 snarfed consider skimming the php getting started tutorial to see if anything jumps out at you
# 18:38 tantek bear++ for trying to push to fix the indiewebify github / deployment situation
# 18:38 tantek snarfed++ for trying to push to fix the indiewebify github / deployment situation
# 18:39 tantek aaronp++ for volunteering to host to help fix the indiewebify github / deployment situation
# 18:39 tantek aaronpk++ for volunteering to host to help fix the indiewebify github / deployment situation
# 18:41 aaronpk snarfed: the "hello world" works fine, so it's defintiely something about the project
# 18:44 aaronpk snarfed: do you have to restart dev_appserver.py when you change source files?
Tino_ joined the channel
# 18:47 aaronpk narrowed it down to something with one of the files
# 18:49 aaronpk i will just do a binary search to find the line of code that's causing the error
[snarfed] joined the channel
# 18:55 aaronpk it looks like we can do github->appengine without travis-ci. is there a benefit to using travis in the middle?
# 18:55 [snarfed] at least the wm sender returns an error now instead of hanging
# 18:59 aaronpk ah well there are tests, so let's see if i can hook it up to travis
# 19:03 aaronpk hmm i'm gonna have to commit an API key to the repo so that it can deploy
# 19:06 snarfed i think github lets you gate push triggers on its "checks," like travis, so maybe use github's push-to-deploy instead of travis's?
# 19:07 aaronpk i would if i could read that blog post on how to use github's deploy
# 19:12 snarfed looks like that's probably good, and hopefully doesn't require creds in the repo
# 19:12 bear with travis you don't commit keys to the repo
# 19:12 aaronpk "For a minimal configuration, all you need to do is add the following to your .travis.yml:"
# 19:12 bear you encrypt them using their command line tool and then reference it in the yaml config
# 19:14 aaronpk oh it's encrypting it with the travis public key, that makes sense
nitot joined the channel
# 19:45 aaronpk snarfed: it's telling me "Please use Google Developers Console to activate the API for your project." but there are like 9929418 APIs in the developer console
# 19:46 aaronpk it's apparently not "Cloud Deployment Manager API"
# 19:46 snarfed i've never actually set up push-to-deploy, or travis
# 19:47 aaronpk i have no idea. this error message doesn't exist on google either
# 19:48 bear you have to enable GCE Auth via the API
# 19:49 bear yea, GCE is the paid side of app-engine
# 19:50 bear they offer a free trial with restrictions but when you use up the free... billing hits
# 19:51 aaronpk but...i don't have to pay for it if i just deploy to app engine
Lancey joined the channel
# 19:51 bear yea, the GCE part is bad info if your targetting app-engine
# 19:52 bear so you have hooked the github up to Google Source Repository?
# 19:52 snarfed to rephrase: some app engine and google cloud features are free but still require you to add a credit card
# 19:54 bear is catching up - opening up tabs for travis and app-engine now
tvn_ joined the channel
# 19:57 bear k, visit the google developer's console and either create or select the project you made for this
# 19:57 bear select the permissions item on the sidebar
# 19:58 bear you will see a list of service accounts
# 19:58 snarfed aaronpk: sounds like you should add bear to the google project as a dev or owner :P
# 19:59 aaronpk oh yes i'd be happy to do that too. i'd like to learn enough to do this tho
# 19:59 bear k, so you already added a credential for a service account of type json?
# 19:59 aaronpk yep, and encrypted that and stored it in the repo
# 20:00 bear is the yml file for travis in the repo?
# 20:01 bear let's go see what travis is outputing
# 20:03 bear i'm working thru how we did this at &yet but not having direct access now to that internal data is slowing me down
# 20:04 aaronpk snarfed: so if i enable billing, it's still possible to use google compute engine on a free tier?
# 20:04 snarfed i think there's a free trial but then you pay, like EC@
# 20:05 snarfed the travis/deploy api part, i'm not familiar with
# 20:06 snarfed but in general don't be afraid to plug in a credit card, if you're just using app engine and some api(s) you won't get charged for small usage
# 20:06 aaronpk well it didn't ask me for a card to launch this on app engine myself
# 20:06 snarfed right. it might for some part of push-to-deploy though
paulproteus joined the channel
# 20:09 aaronpk yeah thats just the interface that shows me a thousand different APIs I could enable
moredhel joined the channel
# 20:14 bear bah - I just realized I was still using my &yet credentials - they still haven't cleared me from things -- so i'm wiping my browser cache and reloading google cloud
# 20:28 bear there are 7 APIs enabled by default for all new projects
# 20:29 bear I'm wondering if it's the deployment manager or even the compute engine one
# 20:31 bear it doesn't hurt to enable all of them - try that maybe?
# 20:31 aaronpk i can try compute engine, but if compute engine and app engine are different things I don't think it's going to work
# 20:31 aaronpk I can't enable compute engine until I add billing info
# 20:32 kylewm a bunch of weird stuff just popped up on Pestagram tagged #indieweb
# 20:34 aaronpk dang, same error with the compute engine api enabled
# 20:35 kylewm !tell kevinmarks: pestagram is being weird, every url redirects to / and is blank
# 20:35 Loqi Ok, I'll tell them that when I see them next
KartikPrabhu joined the channel
# 20:35 bear I suspect it's something that the gae command line tool enables but isn't being setup in the travis flow
lewisnyman joined the channel
# 20:38 bear i'm going to setup a command line test deploy and then work thru that to see if I get any clues
paulproteus joined the channel
nitot joined the channel
# 20:43 snarfed aaronpk: gross alternative: set up a cron job somewhere that runs git pull; appcfg.py update
ttepasse joined the channel
# 20:44 aaronpk yeah that was my first plan without using appengine or travis
# 20:44 aaronpk well, actually i was gonna have it deploy from a github hook
cleverdevil joined the channel
petermolnar joined the channel
# 20:55 kylewm i only know how to get them on the command line with `heroku logs` (iirc)
# 20:58 snarfed aaronpk: agreed on trying for the ideal way. i often set myself a rough time limit to get that working, and if i pass it i fall back to lower tech
# 20:59 kylewm if nothing else, woodwind polls it every hour which would try to keep it awake more than 18 hours
nitot joined the channel
[kevinmarks] joined the channel
# 21:09 [kevinmarks] This is exactly the original rss use case, when it was for netscapes home page
# 21:15 Zegnat So, they invent AMP because they need a webpage format that is easily embeddable (iframable), and then when they need to embed content cards they need a different format?
# 21:17 Zegnat I saw that after I typed, sorry. The argument still stands though, I think
tantek joined the channel
# 21:25 Zegnat kevinmarks, I still think it is weird to create separate JSON files no matter who proposes it. Especially as AMP has a standing example of JSON-LD on the actual page for publishing metadata. Even setting the mf argument aside
# 21:29 [kevinmarks] Ignoring the json part briefly, the case for a separate suggested pages url makes some sense. It looks like a h-feed of h-entry's to me, but maybe the photo centricity means something else makes more sense?
# 21:34 tantek aaronpk++ for volunteering to host to help fix the indiewebify github / deployment situation
# 21:37 tantek kevinmarks: re: https://github.com/ampproject/amphtml/issues/657 so I think I dare you to post a comment to that with a *MINIMAL* parsed h-entry result that has the same feature set as they are attempting to reinvent. And the point out BTW this can be automatically generated from HTML as well.
# 21:38 tantek Especially since someone already asked "some standard microformat ... ?"
KartikPrabhu joined the channel
# 22:09 tantek have you missed all the discussion of u-featured over the past year?!?
# 22:14 tantek more documentation of examples welcome! (e.g. aaronpk ^^^)
# 22:15 tantek question is, do any /reader implementations do anything special with u-featured? kylewm ?
scoates and nitot joined the channel
# 22:39 kevinmarks this story cards thing feels liek it's worth doing the full microformats process ecercise on
scoates joined the channel
kerozene joined the channel
mlncn joined the channel