#indiewebcamp 2014-05-25

2014-05-25 UTC
#
KartikPrabhu kylewm: do you know how to get headers from a WSGIRequest object? request.headers does not seem to work
#
kylewm KartikPrabhu: no, sorry...looks like django specific stuff
#
KartikPrabhu oh hmm
#
KartikPrabhu aaronpk: any way to investigate what headers indiepost is sending me? There does not seem to be a 'Authorization' header at all
#
aaronpk Are you running behind a proxy? Some of them don't forward the auth header unless explicitly enabled
#
KartikPrabhu I don't think so. will check
lionzan joined the channel
#
KartikPrabhu aaronpk: apparently I had to enable that in the Apache configuration!
luxagraf joined the channel
#
KartikPrabhu is able to verify both taproot and indiepost micropub clients
tracend and jjuran joined the channel
#
@veganstraightedge Informal #indieweb dinner in SF tonight at 8p? @t @benwerd @jlsuttles @erinjo who else? (twitter.com/_/status/470362745271889921)
#
GWG Evening
#
GWG Still working on webactions and my comment plugin. May need some wisdom
#
lionzan after a long evening of work (it's 230am in the UK) I managed to validate h-card and h-entries!!
#
lionzan on lionzan.me (Jekill)
#
KartikPrabhu nice work lionzan
#
lionzan thx!
#
KartikPrabhu GWG: what are you doing with webactions?
#
GWG Trying to add them to my plugin
#
GWG http://tiny.n9n.us/2014/04/25/video-embed-test/
#
GWG This is the test page
#
GWG I took out the comment form thanks to you, KartikPrabhu
#
GWG The design I'm using is temporary
#
GWG But I took my syndication links and ripped them out of the theme and added that to the plugin.
#
GWG By having them in there, I can do the webactions + webmentions form instead of the comment form
#
KartikPrabhu hmm neat. the "reply on G+" doesn't seem to do anything yet
#
GWG KartikPrabhu: You've noticed.
#
GWG Thus the help.
#
GWG My plan was to try, when there isn't an intent like Twitter has, to pop up the syndicated post in a box.
#
KartikPrabhu oh hmm G+ does not support posting so you might not be able to do that
#
GWG KartikPrabhu: I don't need it to post.
jedahan joined the channel
#
GWG And it does support a share intent.
#
GWG Hit the Post about this on Google+ link
#
GWG I can't get a +1 or a Like so easily
#
KartikPrabhu yeah I see. you want the "reply" to similarly go to a comment box on G+?
#
GWG KartikPrabhu: Seems like a good compromise.
#
GWG Both Facebook and Google + support embedding posts.
#
GWG So, I just have to figure out a way to set up a popup in an embed such that I don't have to load the Facebook/G+ javascript except in the embed.
#
GWG What annoys me is that Facebook wants an app ID for loading half their things.
#
GWG My whole goal is to pop up an instance of Facebook so I don't have to support all that stuff.
#
GWG If only I'd bothered to learn Javascript
#
GWG I stopped at PHP
#
KartikPrabhu GWG: I couldn't find a way to do the comment on G+ thing... sorry. These silo APIs are annoying. manybe someone else knows how to do this
#
GWG KartikPrabhu: My intention was to pop up a window with the post embedded in.
#
GWG https://developers.google.com/+/web/embedded-post/
#
GWG Same with Facebook
#
KartikPrabhu oh isee
#
GWG But, that is more a javascript thing, and I admit I never got around to learning more than a few things with Javascript
#
GWG I may just stick with the popups I have
#
GWG This is still me just playing with it.
#
GWG So, I thought I might be able to get some other ideas.
#
KartikPrabhu GWG: here is a page that explains how to do a popup with javascript code: http://www.quirksmode.org/js/popup.html
#
GWG I have a simpler idea to try.
#
kylewm has anyone had experience using facebook's API to tag people in messages?
#
GWG Okay. Making progress here.
#
GWG Just wrote an embed_facebook function
#
KartikPrabhu GWG: good good.. :)
#
GWG And will add an embed_gplus function
#
KartikPrabhu kylewm: sorry. I avoid silo APIs like the plague :)
#
GWG Once I have the functions, I can figure out where to embed.
#
GWG kylewm: I'm sitting here trying to figure out how to use Facebook without an app id
#
GWG So I'm the wrong person to ask.
jedahan joined the channel
#
KartikPrabhu GWG: i suggest going the whole 9 yards with one function before writing another one. That way you know all the stumbling blocs and are ready next time :)
#
KartikPrabhu that is in my limited experience
#
GWG Well, Wordpress already supports Twitter embeds. So I don't need to add that.
#
GWG Basically, I'm using these sites own embed code
#
GWG So I'm not writing it per se.
luxagraf joined the channel
#
aaronpk !tell lionzan congrats!!
#
Loqi Ok, I'll tell them that when I see them next
#
GWG aaronpk: I've been using your instagram feed to test some things, as I don't use it. I hope you don't mind.
#
aaronpk GWG: uh.. sure?
#
aaronpk like what?
#
GWG I'm working on syndication links. I needed an Instagram link.
#
GWG You had one
#
GWG I don't have an Instagram account.
#
GWG I wanted to cover all the sites that backfeed using Bridgy.
#
aaronpk cool
#
GWG It seems the silos aren't cooperative about what I want to do, so I'll have to improvise.
#
GWG I just wrote an embed function in PHP before remembering you can't cause it to display an embed on-click without javascript, which I never got around to learning.
nloadholtes joined the channel
#
GWG So, either I change approach or I learn some Javascript.
#
GWG The easiest approach seems to be to write a self-contained piece of PHP that, when called with a URL argument, pops up the embedded version of a specific post.
#
GWG Then that forms my webaction fallback
paulcp joined the channel
#
aaronpk just discovered a public URL on jawbone.com for all my sleep data
#
aaronpk I suspected they probably had a URL for it and was able to find it by clicking the "share" button from the app
#
GWG aaronpk: That's good. How are you sleeping?
#
aaronpk GWG: I usually lie down in bed first, then shut my eyes
krendil joined the channel
#
GWG aaronpk: You know what I meant.
#
aaronpk :D
#
GWG Quantified self is full of interesting concepts.
paulcp joined the channel
#
GWG Okay. Repository updated.
#
GWG I may be getting the hang of this.
#
KartikPrabhu micropub success! see first two posts here: https://kartikprabhu.com/notes :D
#
GWG KartikPrabhu: I still think Notes should be higher on your home page.
#
KartikPrabhu GWG: I have been thinking about restructyring my homepage to include notes... don't know a good way to do it yet is all
#
GWG KartikPrabhu: I meant the link to the notes
#
KartikPrabhu oh! naah I like the navigation at the bottom
#
GWG Why not add it to this?
#
GWG You could start with some of the latest articles below
#
GWG or jump right into the archive, or explore some of my photography, painting. How about a little short fiction and haiku? Maybe some thoughts about the web…
#
GWG Reference notes in that paragraph?
#
KartikPrabhu oh! hmm yeah good point :D
#
KartikPrabhu thanks
#
GWG I try.
squeakytoy and paulcp_ joined the channel
#
GWG KartikPrabhu: Looking forward to meeting you in NYC
#
KartikPrabhu same here GWG... should be a fun weekend
#
GWG KartikPrabhu: I don't think I can recognize you from your photo though
#
GWG But they gave out nametags last time
#
GWG Jeremy Zilar has a button maker.
#
KartikPrabhu ha! I am an ephemeral shadow
#
GWG You can be one with a nametag
#
KartikPrabhu I am sure once everyone is introduced we shouldn't need nametags
#
GWG KartikPrabhu: I might confuse you with other ephemeral shadows.
#
GWG Or cats.
#
GWG I saw a cat is attending
#
KartikPrabhu oh yeah chloeweil... :P I am sure cats are diff for shadows :P
#
GWG I think she was at the last one, but on Sunday. I was only there on Saturday, and remoted in for a few hours Sunday morning
#
KartikPrabhu in all seriousness though I don't actually have any recent photo of myself. I am usually on the other side of a camera and that is the most decent "selfie" I could take
#
GWG KartikPrabhu: I have the same problem. My picture is probably two years old for that reason. Need to update. I should shave and such first. Look presentable.
#
KartikPrabhu !tell tantek: West is leading 8-7 again... something must be done #gamification
#
Loqi Ok, I'll tell him that when I see him next
#
GWG KartikPrabhu: Got friends who are indie-curious?
#
KartikPrabhu not around these parts... most of my friends are busy being physicists...
#
GWG KartikPrabhu: I haven't had any luck either. I know people who respect the concept, but not enough to come somewhere/do something
#
KartikPrabhu at the moment it would seem indieweb would be for devs/bloggers/journalists
#
GWG KartikPrabhu: I'm trying to work on that. I'm limited though. That is why I started trying to go universal with my work.
jonnybarnes joined the channel
#
aaronpk KartikPrabhu: congrats!!
#
KartikPrabhu thanks :) will tidy it up tomorrow but it is a working code
#
aaronpk i'm thiing maybe I should rename my app
#
GWG To?
#
aaronpk it should be a bird name, like Falcon ;)
#
GWG Too common.
#
aaronpk (Falcon is tantek's)
#
GWG How about Turkey? No one names things after the Turkey.
#
KartikPrabhu kite
#
GWG How about the Kakapo?
#
aaronpk how about Raptor!
#
KartikPrabhu ha!
#
KartikPrabhu go for it :P
#
aaronpk aw raptor.io is taken
#
GWG Be like everyone else....pick a flashy avian
#
KartikPrabhu ?
#
KartikPrabhu dolph.in is taken too :P
#
GWG I suggested a Kakapo.
#
GWG Or a Turkey.
#
GWG Why does everyone go mainstream?
#
aaronpk raptors are cool
#
KartikPrabhu I don't I am naming things after obscure mathematical structures :P
#
GWG Such as?
#
KartikPrabhu "Parallel Transport" , connection, bundle ... the last two are not public yet
#
aaronpk ooh there are lots of good names in combinatorics
#
aaronpk raptor.enterprises lol
#
GWG KartikPrabhu: How about Prabhubility?
#
KartikPrabhu my file storage database "pullback" still in progress
#
aaronpk wtf there's a .kiwi TLD
#
aaronpk RAPTOR.ZONE
#
aaronpk !!!
#
KartikPrabhu GWG lol! I can't name something after a thing I am bad at :P
#
GWG Kartikgraphy?
#
aaronpk GWG++
#
Loqi GWG has 2 karma
#
KartikPrabhu ha!
#
aaronpk raptor.farm oh man
#
KartikPrabhu rap.tr?
#
KartikPrabhu is .tr a tld?
#
KartikPrabhu yup turkey
#
GWG KartikPrabhu: This doesn't work with my name.
#
KartikPrabhu idno Davidology seems nice
#
GWG My last name is hard to work with
#
KartikPrabhu Shanskape?
#
KartikPrabhu aaronpk: rap.tr is takjen too...
#
GWG KartikPrabhu: The e is not silent, by the way.
#
KartikPrabhu yeah i know :)
#
GWG Wasn't sure.
netweb, dybskiy, chrissaad, KartikPrabhu and friedcell joined the channel
#
@voxpelli RT @benwerd: Owning your own site is the only way to guarantee you'll reach your audience. http://stream.withknown.com/2014/for-content-creators-theres-no-guarantee-youll-reach-your-audience #indieweb (twitter.com/_/status/470477328616022016)
eschnou, whaity, catsup, jonnybarnes, ottery, erikmaarten and lionzan joined the channel
#
Loqi lionzan: aaronpk left you a message 8 hours, 17 minutes ago: congrats!!
#
lionzan thx aaronpk!
#
lionzan Moving on Webmentions now
#
lionzan and having problems with https://github.com/indieweb/mention-client-ruby
#
lionzan but I'd guess I'll have to wait for aaronpk to wake up!
nloadholtes and ttepasse joined the channel
#
jonnybarnes did aaronpk write that? I though he develops the PHP librarys
#
lionzan hi jonnybarned, no idea, I just need to understand how to make that ruby client work with my Jekyll implementation
aboynejames joined the channel
#
@lionzan Testing Webmentions https://twitter.com/lionzan/status/470533443739475969/photo/1 (twitter.com/_/status/470533443739475969)
friedcell and erikmaarten joined the channel
#
lionzan I don't get the context in which I should use the commands shown in the Usage section of https://github.com/indieweb/mention-client-ruby
#
lionzan it's not command line I guess...
#
@lionzan Testing Webmentions. I mention lionzan.me and let's see what happens! Of course nothing happens! It must be a post… (http://lionzan.me/2014-05-25/testing-webmentions/) (twitter.com/_/status/470550095826407424)
jonnybarnes and aboynejames joined the channel
#
@Ladyfire_News genshi.box a file storage drive and web server completely secure through AES 256-bit encryption https://www.indiegogo.com/ #indieweb #freedom (twitter.com/_/status/470556961344200704)
#
@julien51 RT @benwerd: Owning your own site is the only way to guarantee you'll reach your audience. http://stream.withknown.com/2014/for-content-creators-theres-no-guarantee-youll-reach-your-audience #indieweb (twitter.com/_/status/470563023011065856)
#
david.shanske.com edited /User:David.shanske.com (+66) "/* Features in Development */" (view diff)
#
GWG I need some help with copy
#
GWG http://tiny.n9n.us/2014/04/25/video-embed-test/
#
GWG The Comment on This Post and Webmention section. I want to refine the text.
#
lionzan I sort of understand when I should run mention-client-ruby, I have to uncover how!
#
lionzan it should be after I upload the site to S3, or probably at the same time as mention-client-ruby... I'm confused
#
jonnybarnes lionzan, is your site essentially a static site?
#
jonnybarnes being generated by something like Jekyll
#
lionzan yes absolutely, it's in Jekyll
#
jonnybarnes then you'll need to run the mentions client after you've updated the static content
#
jonnybarnes because
#
jonnybarnes the client will send a webmention
#
lionzan yes, that's what I thought,
#
jonnybarnes lets say for example it sends it to me
#
jonnybarnes it sends a source value and a target value, target being the thing on my site you're mentioning
#
lionzan and source my PUBLISHED post
#
lionzan right?
#
jonnybarnes the source is thing on your site that does the mention
#
jonnybarnes yup
#
lionzan so that's how I understood, good
#
jonnybarnes I can't grab the source and display it on my site if you havent published it on yours yet
#
lionzan but now operationally, how do I run the mention-client-ruby stuff ?
#
jonnybarnes sorry lionzan, I use PHP myself
#
lionzan as you see, I'm stuck at a very primitive level
#
lionzan yeah, no prob :)
#
lionzan I'm gonna get it!
#
jonnybarnes I think bear has a static site
#
lionzan ben ward and bret comnes are with Jekyll
#
lionzan msg benward hi I want to get webmentions on jekyll and I'm stuck, can you help? Thx!
#
lionzan shut, missing /
#
voxpelli lionzan: I pretty much have webmentions on Jekyll – feel free to ask me anything
#
lionzan hi voxpelli, thx!
#
lionzan so, how do I make it happen? :)
#
lionzan I thought of using
#
lionzan mention-client-ruby
#
lionzan it seems relatively easy...
#
lionzan consider I'm no CS
#
voxpelli lionzan: what do you want to do? send or receive?
#
lionzan well, send, to start with. because to receive I have to have a running process on the server huh?
#
voxpelli lionzan: I only receive, using my external endpoint https://webmention.herokuapp.com/
#
lionzan I see. how do you set up so that the webmention to your post gets to that server?
#
voxpelli lionzan: if you want to send automatically with Jekyll, then it depends oin how you publish your site – if you eg. host it on GitHub Pages you need to do something upon the webhook that is sent after the page has been generated
#
voxpelli lionzan: You just follow the 3-4 steps on that page :)
#
lionzan wait, let's stick at what you do with your website i.e. how do you set up an external endpoint?
#
lionzan ops, let me see
#
voxpelli if you get stuck, then here's the source code of my blog as a reference: https://github.com/voxpelli/voxpelli.github.com
#
lionzan ok I see, super easy, I'll impelment immediately
#
lionzan now to the sending. yes I have to send out the webmention after publishing.
#
lionzan i would llike to do an experiment, can i try to mention a post of yours?
#
voxpelli lionzan: I guess you can testmention http://voxpelli.com/2013/12/webmentions-for-static-pages/
#
lionzan working on it, I'm editing the publisher ruby to include the commands as per https://github.com/indieweb/mention-client-ruby > Usage
erikmaarten joined the channel
#
kylewm the Facebook API is officially a bummer. looks like you can't mention people without creating and getting approved a custom 'story'
#
GWG kylewm: Is your brain available?
#
kylewm GWG: haven't had coffee yet, so like 50% available ;)
eschnou joined the channel
#
kylewm GWG: but yes
#
lionzan voxpelli any idea on how to include that snippet in a .rb file? it says S3Website::Uploader::Webmention but I have webmention gem installed
#
voxpelli lionzan: which snippet?
#
GWG kylewm: I'm looking for thoughts on copy
#
GWG kylewm: And since you are the unofficial tiny.n9n.us poster child, that you might help
#
GWG This is the paragraph I think I could do better on. "This site does not offer a way for readers to post comments directly on the site. Instead, this site accepts webmentions, so you can respond on your own site and link back to us. We also support the import of comments from the sites we syndicate to."
#
lionzan the one at https://github.com/indieweb/mention-client-ruby > usage
#
kylewm GWG: hmm, it's not bad. the shift from 'This site' to 'we' is a little off-putting , but i'm not sure how to fix it without going passive voice (comments will be imported...)
#
GWG I was trying to be general for the plugin, but I see your point.
#
GWG How about this?
#
GWG "This site does not offer a way for readers to post comments directly on the site. Instead, this site accepts webmentions, so you can respond on your own site. Comments from your own site and the sites this content is syndicated to will be imported into this page."
#
aaronpk jonnybarnes: lionzan: I dabble in Ruby
#
voxpelli lionzan: I'm not sure where you should put it if you want to auto ping from your ruby
snarfed joined the channel
#
kylewm GWG: iterating on that, how about? "This site does not offer a way to post comments directly. Instead, readers are encouraged to write on their own site, sending a webmention to notify me of the reply. Comments from other sites where this content is syndicated will also be imported into this page."
eschnou joined the channel
#
kylewm lionzan: are you using https://github.com/laurilehmijoki/s3_website to publish to s3?
#
lionzan kylewm yes!
#
GWG kylewm: Edited again
#
@NZN Sovereign Identity in the Great Silo Forest http://blogs.law.harvard.edu/doc/2013/10/14/iiw-challenge-1-sovereign-identity-in-the-great-silo-forest/ .@dsearls #vrm #iiw #indieweb #indieIDP #makered #makerfaire (twitter.com/_/status/470592796164759552)
#
GWG This site does not offer a way for readers to post comments directly on the site. Instead, readers are encouraged to respond on their own site, sending a webmention or using the below form to notify me of a reply. Comments from the sites this content is syndicated to will be imported into this page automatically.
#
GWG I think that might do it
#
kylewm lionzan: do you have
#
lionzan kylewm do you have suggestions on how to automate webmentions publish?
#
lionzan in my setup with s3_website push
#
aaronpk wow your site is hosted on s3?
#
aaronpk i don't know of anyone else here who's doing that yet!
#
lionzan absolutely yes!
#
aaronparecki.com created /S3 (+22) "r" (view diff)
#
kylewm lionzan: detecting new posts seems like the tricky bit. since you want to run the s3 uploader, then run webmention sender on all new posts ... i don't suppose you already have a mechanism for that? :)
#
kylewm GWG: that looks good to me
#
lionzan that's exacly what I am missing. i also have the pieces to do it, but I'm no good at Ruby and I don't know how to do it!
#
snarfed morning all
#
snarfed hmm. i tried u-syndication links on a post yesterday but bridgy didn't find them. time to debug…
#
snarfed https://snarfed.org/2014-05-24_international-beer-festival-tonight , https://www.brid.gy/facebook/212038
#
lionzan "the pieces" being https://github.com/indieweb/mention-client-ruby > usage. I just have to be able to call that from within the upload process. I know the exact line of code, but it seems i need to make the module aware that I want to use the webmention ruby gem
#
GWG kylewm: Can I pick your brain on a followup question?
#
GWG Hello, snarfed
#
aaronparecki.com created /Amazon_S3 (+257) "stub with dfn and indieweb examples" (view diff)
#
lionzan aaronpk I will try to write something on the wiki to document how I did it
#
aaronpk cool!
#
kylewm lionzan: the easiest thing i can think of is to have a script you run after uploading. it would fetch your site and look for permalinks, invoke the webmention sender for each permalink that has not been seen before, and write out the 'seen' permalinks to a file so you can avoid sending them again next time.
#
lionzan but I need help with BASIC Ruby, how do I call mention-client-ruby from within a Ruby Module?
#
aaronpk does this not work? "client = Webmention::Client.new url"
#
lionzan kylewm, that's exactly what the script here does. https://github.com/indieweb/mention-client-ruby > usage
#
aaronpk oh have you added the gem to your Gemfile?
#
lionzan I'm stuck with a much omore BASIC Ruby issue :)
#
lionzan aaronpk, I did "gem install webmention" and it worked
#
lionzan now how do I use it within a Module?
#
aaronpk ok yeah you probably need to add it to your Gemfile, then when you run the script with "bundle exec" it'll find it
#
lionzan OK, can you please help me out here? you mean the Gemfile of what? or there's just one Gemfile? (this tells you my level with Ruby :[] )
#
aaronpk there is one Gemfile for your project, in the root folder
#
aaronpk is this up on github somewhere?
#
lionzan no, I haven't been able to make it work from github (dumb-me)
#
aaronpk ok let me find some other jekyll site for reference
#
lionzan root folder of the site or of what?
#
aaronpk your source code probably looks something like this? https://github.com/jekyll/jekyll
brianloveswords joined the channel
#
lionzan that's how my jekyll installation should look like, but I just installed following a tutorial and i am not working with that.
#
lionzan I just call from command >jekyll build
#
lionzan and it does the job
#
lionzan then I call > s3_website push and the site is on S3
j12t joined the channel
#
lionzan of course I have a local version of the site, but it doesn't have any Gemfile
#
aaronpk k let me try something
#
lionzan yep
#
aaronpk lionzan: what module are you trying to run it in?
#
lionzan Module S3Website (which I installed the gem and works OK)
#
lionzan class Uploader
luxagraf joined the channel
#
aaronpk you know I should just make a command line tool for the mention-client
#
lionzan I want to put the code of https://github.com/indieweb/mention-client-ruby > usage after line 92 of module S3Website, Class Uploader
#
aaronpk here? https://github.com/laurilehmijoki/s3_website/blob/master/lib/s3_website/uploader.rb#L92
#
lionzan but I want to call it immediately after S3Website uploads each file, otherwise i don't know how to select which URLS webmention (th enew ones) and which not (the old ones)
#
lionzan exactly!
#
lionzan with the following lines:
#
lionzan url = "http://lionzan.me/#{upload.details}"
#
lionzan client = Webmention::Client.new url
#
lionzan sent = client.send_mentions
#
lionzan puts "Sent #{sent} mentions"
#
lionzan but I get an error msg
#
aaronpk i see...
#
aaronpk i think there is a better way
#
aaronpk the "upload_files" method returns a list of new files: https://github.com/laurilehmijoki/s3_website/blob/master/lib/s3_website/uploader.rb#L59
KartikPrabhu joined the channel
#
lionzan Exception Occurred: undefined method `[]' for <AWS::S3::ObjectCollection>:AWS::S3::ObjectCollection (NoMethodError) Retrying in 3 seconds...
#
lionzan let me see the link you sent me
#
lionzan I see
realz joined the channel
#
lionzan we need to iterate on those
#
aaronpk oh sorry it actually jsut returns the count :/
#
lionzan but still from within the S3Module, which doesn't like to see this:
#
lionzan client = Webmention::Client.new url
#
aaronpk that's because it doesn't know about the library
#
aaronpk i'm still confused about something, how are you making changes to the s3_website code and running it from the command line?
#
lionzan I guessed so, how do I tell it?
#
lionzan good question
#
aaronpk presumably you have the source code for s3_website on your computer?
#
aaronpk if so, there is a gemfile there
#
lionzan I have the code installed, i edit as root, and I assume it gets compiled each time?
Sebastien-L joined the channel
#
lionzan yes I have the source, and yes, now I see that there i have also the Gemfile
#
aaronpk if you add gem 'webmention' to the Gemfile, then run whatever you're running with "bundle exec" first it should find it
#
aaronpk so the gemfile ends up looking like this https://gist.github.com/aaronpk/4a71102b9164369bc504
#
lionzan let me try and do it and i get back
#
aaronpk k
#
lionzan question: the Gemfile in the Jekyll gem or in the S3Module?
#
aaronpk s3
#
lionzan now it looks like this:
#
lionzan source "http://rubygems.org"
#
lionzan gemspec
#
lionzan ok saw your example, let me do it
#
lionzan done
#
lionzan let me try if S3Module works now
#
@chloeweil @mibester any interest in coming down to NY for Indie Web Camp? Loved your POSSE to Twitter article. http://indiewebcamp.com/2014/Guest_List (twitter.com/_/status/470602560886161408)
#
lionzan aaronpk I tried, there's still sthg wrong, i get this msg
#
lionzan Exception Occurred: undefined method `[]' for <AWS::S3::ObjectCollection>:AWS::S3::ObjectCollection (NoMethodError)
#
aaronpk that's a really strange error to be getting for that change you made... are you sure there aren't any other changes?
chrissaad joined the channel
#
Loqi [@mattfeifarek] Just listened to this http://caseorganic.com/articles/2014/03/11/2/ on the #indieweb
#
Loqi nostalgic & hopeful Silos are for grain, not us! @mdoudoroff @Michael_Marmor (http://twtr.io/mgrFHfKT4Y)
#
lionzan reste the file to original and it works, now I change just the lines after 92 and come back in a minute
#
lionzan ok, so I get this error
#
lionzan Upload 2014-05-25/testing-webmentions/index.html: Success!
#
lionzan Exception Occurred: uninitialized constant S3Website::Uploader::Webmention (NameError) Retrying in 3 seconds...
snarfed1 joined the channel
#
aaronpk ok that's better, that's more like what I expected to see
#
aaronpk what command exactly are you running from the command line?
rektide_ joined the channel
#
lionzan I just type >s3_website push
snarfed1 joined the channel
#
lionzan that is how you upload the files and it works
#
aaronpk oh... are you editing the gem source code that it installed?
#
lionzan yes, is it bad? i see that it affects the results...
#
aaronpk yeah that's not really how your'e supposed to do it
#
lionzan I started programming assembler on an 8088 to give you an idea of my understanding...
#
lionzan long ago
#
aaronpk :)
#
aaronpk ruby is a little bit different
#
lionzan in fact I don't get entirely how Ruby works
#
aaronpk sometimes I don't either
#
lionzan I guessed so ;)
#
lionzan of me not of you!
#
aaronpk ok here's a new idea for you:
#
lionzan yes!
#
aaronpk make a new folder somewhere and download the source of s3_website using "git clone git@github.com:laurilehmijoki/s3_website.git"
#
lionzan can't I just copy the folder I have already on my HD?
#
aaronpk hm sure
#
lionzan well, sorry, just let me do it :)
#
aaronpk then in the new folder with the source, add the 'gem "webmention"' line to the Gemfile
#
lionzan ok done
#
aaronpk ok actually I think I figured out a way for it to work in either case, regardless of how hacky it is to edit the gem in the system path
#
lionzan tell me
#
aaronpk find the lib/s3_website.rb file
#
aaronpk and add this to the bottom of the list of requires:
#
aaronpk require 'webmention'
#
lionzan done
#
lionzan shall we try?
#
aaronpk sure
snarfed joined the channel
#
lionzan not good yet. but different
#
aaronpk oh my
#
lionzan give me a minute
#
lionzan here it is:
#
lionzan it finishes self.upload_files at least until line 74.
#
aaronpk gotta run, back in a half hour or so
#
lionzan then before getting to line 92 for the first time it goes:
#
lionzan Exception Occurred: undefined method `[]' for <AWS::S3::ObjectCollection>:AWS::S3::ObjectCollection (NoMethodError) Retrying in 3 seconds...
#
lionzan OK talk to you then, I'm hooked here :)
brianloveswords and bnvk joined the channel
#
GWG CDN versions of popular libraries vs delivering them yourself. Any thoughts?
#
lionzan.me edited /Amazon_S3 (+1098) (view diff)
#
GWG lionzan: Visually appealing site.
#
lionzan.me edited /Amazon_S3 (+53) "/* lionzan.me */" (view diff)
#
lionzan GWG thx!
#
GWG lionzan: Any time.
#
GWG I ask a lot of questions around here. Always willing to try to answer some if I think I have something to add
#
lionzan I see, it's a great community!
#
GWG lionzan: I've found that.
#
lionzan I guess yo ufollowed the exchange with aaronpk, it's about sending webmentions while uploading each individual static post
#
petermolnar.eu edited /2014/UK/Guest_List (+246) "/* Creators */" (view diff)
#
GWG lionzan: I followed it. Just I use a dynamic site with static caching, so I didn't have much to contribute.
#
lionzan checked yr website, you're very active!
#
GWG lionzan: Am I?
#
GWG How active is vry?
#
lionzan :[]
#
lionzan it's individual, anything more than myself, i see as very active ;)
#
GWG I'm trying to be.
#
GWG I see so many people focusing on process over productin
#
aaronpk ok i'm back
#
GWG Take aaronpk, who is very active both in process and production?
#
bnvk snarfed: you around?
#
snarfed bnvk: hey, yup
#
bnvk snarfed: cool. I wanted to get your thoughts on archiving of social media data into flat file formats- especially private messages
#
snarfed bnvk: hmm! ok
#
snarfed i assume you've looked at what the fb, tw, and g+ archives currently do?
#
bnvk yah, the Facebook archive download is useless
#
snarfed doesn't have private messages?
#
bnvk it's a ridiculously long HTML page with no microformats, IDs or anything
#
snarfed oh, sure, but at least that's a start, right? you can munge it from there
#
bnvk I'd basically be writing raw HTML parsing and creating data structs from scratch
#
snarfed sure. still, less work than writing an api-based downloader yourself
#
bnvk nah, I was able to use the python fb lib and do it quite easily
#
snarfed and it scales? sounds good then, go for it
#
snarfed you might also look at socialsafe and thinkup
#
bnvk yah, not sure if either do private messages
#
snarfed we can find out :P
#
bnvk in Thinkup's case it would be storing it in a MySQL db tho
#
bnvk and I'm thinking more along the lines of flat files that are segmented in a meaningful way (by day) or week or somethign
#
snarfed ok
#
snarfed so your question is mostly how to lay out the directory structure?
#
bnvk yah, like what is best in terms of scaling that
#
aaronpk scaling to what?
#
snarfed data volume is probably the only scaling issue, since it sounds like bnvk isn't serving this, just d/ling it
#
bnvk aaronpk: individual users entire archive
#
snarfed bnvk: for a single user's data, i doubt you'll hit fs limits like inodes or files per directory, so just do what makes sense to you
#
bnvk scaling for easy retrieval / parsing / etc...
#
bnvk right, files per dir will not really be a thing
#
snarfed bnvk: is this a one time dump? or ongoing, rsync style?
#
bnvk well, ideally it should support both
#
aaronpk I always like the YYYY/mm/dd/HHMMSS.ext format
#
bnvk I'm gonna delete all my Fbook data and walk away, I figure others won't do that
#
aaronpk or add a unique index after that if your'e worried about more than one thing per second (which is definitely a thing in IRC)
#
bnvk aaronpk: is that how you archive your site's data?
#
bnvk Oh, other point of ref here is typing this data to users fbook ID, picture and profile link, which are downloaded separately
#
snarfed bnvk: i'd stick with the facebook object ids for that
#
aaronpk my articles on disk: https://farm6.staticflickr.com/5239/14080936910_37d1daa8be_o.png
#
GWG snarfed: Do you plans to add any additional backfeed options to Bridgy?
#
snarfed (unless you need ids that are unique in a wider namespace than just fb)
#
snarfed GWG: you mean, like other silos?
#
aaronpk metrics are stored with second-level precision: https://farm4.staticflickr.com/3711/14265440342_7964f7ff8c_o.png
#
bnvk snarfed: yah, I was gonna keep the object IDs to join it all together
#
GWG snarfed: Yes. I made the list of ones I'm supporting in my plugin based on Bridgy
#
GWG 4 of them
#
snarfed GWG: got it. i'd love to! people have asked for flickr and app.net (https://github.com/snarfed/bridgy/issues/70)
#
snarfed i don't plan to implement more myself, but i happily accept PRs!
#
aaronparecki.com uploaded /Special:Log/upload "uploaded "[[File:2014-05-25-aaronpk-articles-filesystem.png]]""
#
aaronparecki.com uploaded /Special:Log/upload "uploaded "[[File:2014-05-25-aaronpk-metrics-filesystem.png]]""
#
GWG snarfed: Right now, support in my plugin is a post meta box
#
GWG I haven't gotten to pulling in the data from Bridgy yet. I'm doing that separately.
#
bnvk aaronpk: so 2014 / 03 / 23-1.md is the 1st note on the 23rd?
#
bnvk a second note would be 23-2.md ?
#
snarfed bnvk: an alternative would be to use the fb object id as the suffix
#
snarfed or even a snippet of the text plus the id
#
aaronpk bnvk: yep
#
snarfed GWG: sounds great!
#
GWG snarfed: I'm seeing a lot of things come together. I'm pleased.
#
snarfed gwg++
#
Loqi gwg has 3 karma
#
bnvk snarfed: right… well so the thing about the messages is that the objects Facebook returns are strange
#
aaronparecki.com uploaded /Special:Log/upload "uploaded "[[File:2014-05-25-aaronpk-notes-filesystem.png]]""
#
bnvk it's a blob of JSON with an object ID and X number of back and forth messages between one or more contacts
#
bnvk but these conversations can span weeks or months
#
bnvk so like 2 msgs from each contact, then fast forward a month, then 15 messages, then 1 day later 4 messages
#
bnvk so perhaps date splitting + conversation ID
#
aaronpk hm so it sounds like the problem you're running into is specifically wrt threading of long-running conversations?
#
bnvk aaronpk: right
#
aaronpk as in, does it really make sense to segment posts by date if you're talking about a single conversation thread that spans a long period of time?
#
bnvk it's gonna be hard to tell an indexer "oh, these messages on these 3 days actually relate to one another"
#
bnvk like email uses reply to headers, Fbook has nothing of the sort
#
aaronpk google voice tries to do this intelligently, but most of the time fails
#
bnvk yah
#
aaronpk it'll group a bunch of SMSs into a single conversation view, and then sometimes starts a new conversation
#
bnvk I think that is a tiny bit better though
pbeaulieu joined the channel
#
aaronpk but I can't tell what it's using to determine when to start a new one, and it seems to get it wrong both ways a lot
#
lionzan hi aaronpk, meanwhile I activated receiving mentions via herokuapp
#
aaronpk it'll often start a new conversation when it should have been part of a previous one, or will add to an existing conversation when it should have started a new one
#
bnvk I think they are actually SMS headers, no? They might be using those just not exporting them to users
#
aaronpk imessage on the other hand takes the opposite approach, everything is just one long conversation thread
#
bnvk aaronpk: is that how it is? Have you ever actually looked at iMessage data on disk?
#
aaronpk i'm talking about just the interface
#
aaronpk i think it's all sqlite and haven't opened that up yet
#
bnvk gotcha
#
aaronpk chat.db
#
aaronpk yep
#
aaronpk sqlite
#
aaronpk wow I should be exporting this
#
bnvk hehe, nice
#
aaronpk crazy schema
#
aaronpk hm yeah good question, how would I want to archive this!
#
aaronpk I think the way my IRC client logs messages would make sense
#
lionzan aaronpk, I lost you! :)
#
aaronpk lionzan: ok sorry let me get back to that
#
bnvk aaronpk: do you think MD is the best format for saving data like this?
#
lionzan no, no pressure!
luxagraf joined the channel
#
aaronpk bnvk: probably not. and in my site my .md files all have a YAML header on top with all the meta data
#
lionzan anyway, there is an issue with accessing the details of th eupload after line 92
#
aaronpk example: https://github.com/aaronpk/p3k/issues/1#issuecomment-29836869
#
aaronpk you can see that post isn't really even markdown at all. it's really only the articles that are markdown
#
bnvk right
#
bnvk so that's a YAML header between the —
#
lionzan aaronpk, a weird thing happens with url = "http://lionzan.me/#{upload.details}"
#
aaronpk lionzan: ohhh maybe upload.details is not actually the URL, but an object
#
aaronpk try before that doing"`puts upload.details.inspect" and see what comes out
#
aaronpk er, bad punctuation, ignore all the "`
#
aaronpk bnvk: yeah and for my event posts there isn't even anything in the body of the post, it's just metadata
#
bret webmentions in jekyll :) http://lionzan.me/2014-05-25/testing-webmentions/
#
lionzan aaronpk, wait, just on line 92 it goes
#
lionzan print "Upload #{upload.details}: Success!\n"
#
lionzan and it worekes flawlessly
#
aaronpk and you see the URL there?
#
lionzan replacing #{upload.details} with the URL
#
bnvk cool
#
bnvk I think YAML makes most sense probably
#
aaronpk i don't know, i'm stumped. I think it might be better to do this the Right Way instead of hacking up the s3 gem
#
lionzan aaronpk, that's the output I get
#
lionzan Upload 2014-05-25/testing-webmentions/index.html: Success!
#
lionzan now I want just to prepend http://lionzan.me/
#
aaronpk yeah I see where you're going with that
#
lionzan ok, what's the Right Way anyway? :)
#
aaronpk well it's unfortunate that the s3 upload gem doesn't actually return the list of new files, but there's likely a way around that
#
lionzan actually there is one difference. in the print, print expects a string argument
#
lionzan whereas when I say url=whatever maybe it's not enough to interpret as a string?
#
aaronpk oh yeah that sounds right
#
aaronpk but if you have "#{upload.details}" in quotes it should turn it into a string
#
lionzan sorry i went back to the old approach. ok so th eway around not having the list?
#
aaronpk so I know it has a method to find the list of new files that it will upload, so you could try using that method in a separate command to find the new files
#
lionzan maybe that's th epoint, url doesn't want to be a strring, but a URL
#
lionzan that's amusing, ok I stop
#
aaronpk bnvk: what other data in the facebook message thread besides the date/person/text is there?
#
lionzan but I have no idea where to look for it
#
aaronpk bnvk: how about a simple HTML format like the irc logs?
#
bnvk aaronpk: not much, there's "name, id, date, message" for each item of who said what, and then a "paging" aspect and an "updated_time" for each conversation object
#
bnvk hrm, HTML, how so? Like output a whole page of a conversation?
#
aaronpk one page per day
eschnou joined the channel
#
bnvk that would be nice for simple displaying back of the information to end user
#
aaronpk yeah you could just open the file in a browser. and with mf2, it's easily parsed back into a machine readable format
#
bnvk but indexing in another app / styling it for your personal site or whever wouldn't that be harder to have to parse the HTML ?
#
aaronpk mf2 parsers do all the dirty work and you just get back a JSON data structure
#
aaronpk you can forget that it's HTML under the hood if you want and just treat it as json
#
bnvk hrm, interesting
#
bnvk what's the state of the Python MF parser?
#
aaronpk it's being used by several people now I believe
#
bnvk oh cool
#
bnvk problem is, it's more complicated to engineer in the immediate
#
bnvk creating a decent HTML method viewing the data starts to come back to actually designing an interface
#
aaronpk interface? nah
#
aaronpk keep it simple
#
aaronpk like this http://indiewebcamp.com/irc/2014-05-25/line/1401042520
#
bnvk hrm
#
bnvk so just whatever the segment of conversation is- make that the single entity as you have done here
#
aaronpk i'm imagining an HTML page like this http://indiewebcamp.com/irc/2014-05-25 which is one day of a private conversation
#
bnvk interesting
#
aaronpk if I can figure out how to extract everything from this sqlite db I might bust this out right now
#
bnvk and the MF parsed data of this looks like what?
#
aaronpk one message would look like this http://pin13.net/mf2/?url=http%3A%2F%2Findiewebcamp.com%2Firc%2F2014-05-25%2Fline%2F1401042520
#
aaronpk and you can stuff in arbitrary properties if you have other things to store
#
bnvk gotcha
#
bnvk what's the diff between content.html and content.value ?
#
aaronpk if there are html tags in the h-entry then content.html contains the tags, but content.value is the plaintext version
#
aaronpk the parser does that
#
bnvk gotcha
#
aaronparecki.com edited /p3k (+929) "/* Storage */ add notes and screenshots about how p3k handles storage of posts" (view diff)
#
bnvk I like the HTML / MF storage in theory, in execution it seems like more effort + complexity than the value tradeoff
#
aaronpk how so?
#
bnvk hard to pin down, just kinda one of those intuition feelings about things
#
aaronpk the built-in fallback to a human-readable archive is a big plus to me
#
bnvk the use case of an end user loading a .html in a browser… sure the web is everywhere and everyhing… but also plain text files / yml feels so much simpler
#
bnvk just "here is text"
#
aaronpk if you don't want to go the html route, then I would suggest literally plain text, not yaml or json
#
aaronpk curl this page: http://indiewebcamp.com/irc/2014-05-25
#
bnvk vs. here is text that must be loaded in a specific application and handled a certain way
#
bnvk yah, I like the idea of plain text, with yaml style headers
#
aaronpk plaintext logs: https://gist.github.com/aaronpk/c97a037749c2b195137a
#
aaronpk easily greppable, opens with any text editor or browser
#
bnvk yah, certainly lighter weight
#
aaronpk of course the downside is you can't easily store other data about each message, whereas html gives you a way to do that if you need
#
bnvk well, what if I did that same line by line style storage
#
bnvk but with USERID being declared higher up in a header of sorts
#
bnvk then it loses it's simple human readable-ness
#
aaronpk sure. or you could use email-header-like syntax for the user in each row: "Aaron Parecki" <11500459@facebook.com>
#
bnvk right, that's not bad
#
bnvk that's what I used store when using fbook / twitter auth before
#
bnvk I liked that separation aspect of the MF version that is URL source of convo / URL of person
#
lionzan aaronpk, still fighting, but at least i got the receive part to work! look at http://lionzan.me/2014-05-25/testing-webmentions/ and see our friend bret!
#
aaronpk lionzan: that's awesome!
#
lionzan I love it when it all works!
#
lionzan and that's why I want to make also the send work!
#
lionzan and I did a few more tests.
#
lionzan there's an issue INSIDE webmention gem
#
aaronpk oh?
#
aaronpk that I can probably fix
#
lionzan I tried passing this
#
lionzan client = Webmention::Client.new "http://lionzan.me/2014-05-25/testing-webmentions/index.html"
#
lionzan very brutal
#
lionzan it MUST work, but I get errors anyway
#
lionzan aaronpk so, I'm debugging as I can, here's what i did
#
lionzan inside clinet.rb of webmention i added print @#url print " in webmention\n" after line 15
#
lionzan then I launch s3-
#
lionzan and i get
#
lionzan http://lionzan.me/2014-05-25/testing-webmentions/index.html in webmention
#
lionzan Exception Occurred: end of file reached (EOFError) Retrying in 3 seconds..
#
aaronpk that is odd... let me test out the gem with your URL
#
lionzan ok
#
lionzan should I mention a post of yours in my http://lionzan.me/2014-05-25/testing-webmentions/ post?
brianloveswords joined the channel
#
aaronpk hmm I'm wondering if I should change the client to only send to links inside an h-entry
luxagraf joined the channel
#
aaronpk cause right now it's trying to send to literally all the links on your page
#
lionzan If you need testing, I mentioned your post http://aaronparecki.com/articles/2014/03/23/1/posting-photos-on-my-own-site
#
lionzan or to links that have an h-entry
#
lionzan ?
#
aaronpk right now it's trying to send mentions to every link on the whole page, including things in your footer and such. i'm changing it to only try to send to links that are inside your h-entry
#
Loqi [mention] https://brid-gy.appspot.com/comment/twitter/chloeweil/470602560886161408/470643765213147136 linked to http://indiewebcamp.com/2014/Guest_List (webmention)
#
lionzan ah sure, you're right
#
@rubygems webmention (0.1.2): A gem for sending webmention (and pingback) notifications http://rubygems.org/gems/webmention (twitter.com/_/status/470645489889988608)
#
aaronpk ok there's that change ^
#
aaronpk also it shouldn't crash anymore, but it seems that it's having trouble sending the webmention to heroku
#
lionzan what should I do
#
aaronpk i'm going to try to rip out some of the guts of the gem some more
#
lionzan should I run gem install webmention ?
#
lionzan again?
#
lionzan or should I uninstall first? I guess with gem uninstall webmention ?
#
aaronpk gem install webmention should update it
#
lionzan let me do it
#
jonnybarnes lionzan: just sent a webmention to http://lionzan.me/2014-05-25/testing-webmentions/
#
jonnybarnes though I come up as unkown :(
tgbrun joined the channel
#
lionzan hey jonnybarnes thanks, it worked! yeah I see, dunno why.
#
lionzan aaronpk, just reinstalled, let me try
#
jonnybarnes lionzan: I use a rel=author link that points to my homepage, the homepage has my hcard.
#
lionzan jonnybarnes, I rely completely on herokuapp
#
lionzan aaronpk apparently S3 now can't load the custom require for webmentions
#
lionzan after I reinstalled
#
lionzan aaronpk I'm lost...
#
aaronpk yeah that's what I was worride about messing with the system gems
#
aaronpk you're not really supposed to do that...
#
bnvk snarfed: it seems Bridgy is not sending a bunch of "date" from my Tweet activity https://brennannovak.com/notes/445 or do you think that is on webmention.io side?
#
@rubygems webmention (0.1.3): A gem for sending webmention (and pingback) notifications http://rubygems.org/gems/webmention (twitter.com/_/status/470650024104587264)
#
lionzan uninstalled and reinstalled webmention
#
lionzan gets better
#
lionzan maybe uninstall and reinstall also s3-
#
lionzan ?
tgbrun joined the channel
#
aaronpk maybe? not sure... I don't usually mess with things this way
#
lionzan tht's my problem, I don't know how things work and i look for a solution my way...
#
lionzan how would a prop tackle the issue of adding webmention to their implementation?
#
aaronpk i'm writing a command line webmention tool for this gem which may give you some more ideas
#
lionzan PRO I meant
#
aaronpk the main challenge you have right now though is getting the s3 uploader to tell you which files are new
#
lionzan yep
#
lionzan and export somehow this info to webmention
#
aaronpk I would start by looking for a way to get that out of the s3 uploader gem as is, and if not available, hack it into that gem (and maybe send a pull request in the process!)
#
aaronpk once you have that, you can use that list and use the webmention gem in a separate script
#
lionzan aaronpk ! apparently it worked! check your post http://aaronparecki.com/articles/2014/03/23/1/posting-photos-on-my-own-site
#
aaronpk that's from my testing here :)
#
lionzan ops :[]
#
snarfed bnvk: yeah, sadly i don't get dates for twitter favorites
#
snarfed they're not exposed in the api, so i scrape them from html, which doesn't have dates for them
#
aaronpk gah i want this function in ruby https://github.com/indieweb/link-rel-parser-php/blob/master/src/IndieWeb/link_rel_parser.php
#
bnvk ah, weird they give retweets date but not fav
#
bnvk oh, I see, you're already scraping ;)
#
bnvk will handle accordingly!
#
GWG aaronpk: You've attended a bunch of indiewebcamps, correct?
#
aaronpk yes
#
GWG How would you define apprentice?
#
GWG I want to get one.
#
aaronpk http://indiewebcamp.com/2011/Guest_List#apprentices_to_builders
#
GWG Is there a guide to being the Master to an Apprentice?
#
aaronpk actually http://indiewebcamp.com/2012/Guest_List#apprentices_to_creators
#
aaronpk hm good question
#
aaronpk not that I can find, but I'm sure tantek could braindump that when he's online next
#
snarfed apprentices are great for getting you coffee, picking up dry cleaning, etc
#
snarfed wait, didn't mean to say that out loud :P
#
GWG aaronpk: MAybe it needs a wiki page
#
KartikPrabhu snarfed: 21st century man!
#
GWG snarfed: I'll start working on it....
#
GWG There should be some medieval language in there
#
GWG "If thy desire is to be a Creator, thou must walk the path of righteousness."
#
snarfed nah, https://twitter.com/eviltantek is a creator too
#
GWG snarfed: Should I ask how that came to be?
#
snarfed iwc sf earlier this year, but i don't know details
#
GWG snarfed: Have you ever discovered any way to like on Facebook without an App ID?
#
snarfed you could grab your cookie from a browser session and make the http request manually
#
GWG snarfed: I want to do something ala the Twitter intents
luxagraf joined the channel
#
GWG Where the user clicks and is directed to Facebook to do the comment or the like
#
snarfed ah, got it
#
GWG So I don't have to load my page up with their scripts.
#
snarfed sure, tons of pages have buttons that do that
#
snarfed oh, i see
#
GWG snarfed: Their developer site shows a need for an appID now even with that
#
snarfed what's the harm in registering an app and including their scripts?
#
snarfed it'll take way more work to do a different way, and i'm not sure what you'd gain
#
GWG snarfed: Because I want to make it a plugin. Needing an app ID would create issues for third-party downloaders
#
snarfed just embed your app id
#
GWG Their embed codes don't require an App ID.
#
snarfed sorry, i meant, hard code your app id in your plugin
#
snarfed so all users use the same app id
#
snarfed not ideal, but it works
#
GWG My current plan is to write a php script that takes a permalink as an argument and sends back a page, independent of the site, that can be popped up.
#
GWG With the embed of the post
#
snarfed sure, that sounds ok too!
#
GWG Then the scripts are only loaded on demand, not on every page
#
GWG Does that sound nuts though?
#
snarfed nah, it sounds fine. i'm still not really on board with the motivation, but it should work ok
#
snarfed gl!
#
snarfed of for a run…
#
GWG Am I the only one who hates the fact I'd have to load social sharing buttons on every page?
#
KartikPrabhu I hate it too so I don't have them :P
#
rascul why you have to do that?
#
GWG rascul: Do what?
#
lionzan aaronpk and all, time to go now, bye
#
lionzan thx!
#
aaronpk bye! i'll ahve something for you soon
tgbrun, krendil and cweiske joined the channel
#
@rubygems webmention (0.1.4): A gem for sending webmention (and pingback) notifications http://rubygems.org/gems/webmention (twitter.com/_/status/470666954194583552)
#
aaronpk sweet
#
aaronpk now there's a command line tool
#
aaronpk and it's pretty verbose which should help show what it's doing
#
KartikPrabhu aaronpk: should micropub clients help set the "slug" for a post?
#
aaronpk uncertain
#
aaronpk but it might make some sense
#
aaronpk especially because my posting interface has a field where I sometimes enter a slug
#
aaronpk and i definitely have customized it by hand past what would have been autogenerated
#
KartikPrabhu yes exactly why I asked :)
#
@rubygems webmention (0.1.5): A gem for sending webmention (and pingback) notifications http://rubygems.org/gems/webmention (twitter.com/_/status/470669448937234433)
luxagraf joined the channel
#
aaronpk already have a feature request for my new post interface
#
aaronpk ...which still doesn't have a satisfactory name
#
aaronpk how about "Penguin"?
#
aaronpk or "Flamingo"
#
aaronpk penguin.flights lol
#
aaronpk moves to #indiechat
#
kylewm.com uploaded /Special:Log/upload "uploaded "[[File:fb-post-news-publishes.png]]""
#
kylewm.com uploaded /Special:Log/upload "uploaded "[[File:fb-post-me-feed-failed-mention.png]]""
#
kylewm.com uploaded /Special:Log/upload "uploaded "[[File:fb-post-taggged-place-and-user.png]]""
#
kylewm.com edited /POSSE_to_Facebook (+1412) "/* Facebook API */ added research/screenshots about mention tagging" (view diff)
#
kylewm.com edited /POSSE_to_Facebook (+57) "/* Mention Tagging */ clarification" (view diff)
#
GWG kylewm: You going to cover comment/likes?
#
GWG Because Likes are the most popular thing I got
#
kylewm GWG: not sure what you mean?
#
GWG You are updating the POSSE to Facebook thing
#
GWG I've been looking at likes
#
kylewm is there uncertainty about how to POSSE likes? i can document what i have done if so
#
GWG kylewm: I'm looking at the Facebook system and trying to find ways to work with it without needing an APP id
#
kylewm I'm really disappointed about the mention tagging thing not working. I wrote this cool little address book so I could tag people on my site and have it automatically converted to the @screenname on twitter and @[fbid] on facebook.
#
kylewm GWG: oh, no clue there
#
GWG kylewm: Can I ask your opinion again?
#
kylewm sure
#
GWG http://tiny.n9n.us/2014/04/25/video-embed-test/
#
GWG Earlier I asked about the Webmentions...
#
GWG What do you think about the Comment part above it
#
kylewm so it looks good. one thing is that barnaby's plugin will replace the part that says [Twitter] with [Repost on kylewm.com], so it might say Repost on [Repost on kylewm.com]
#
kylewm i.e. maybe make the whole line 1 action
#
GWG It is one...but Icould put the text inside the tag
#
kylewm ya that's what i mean
#
GWG I may rearrange it. Not sure the list style works.
#
GWG Want to make the choices easy.
#
kylewm this is pretty cool though, having the intents without including their sanctioned button thing
#
GWG kylewm: That is what is so hard.
#
GWG kylewm: I can't do that for Facebook or Google Plus though
#
kylewm ugh, when I click share on facebook, my ugly mug pops up
#
kylewm yeah, I can understand why they would make it difficult. hard to track people if you don't include proprietary javascript or images on your page
lmjabreu_ joined the channel
#
GWG kylewm: I have a plan for that.
#
GWG But it isn't a good one.
#
kylewm what's the plan?
#
GWG I stole their embed code
#
GWG I want to write a php script that when sent a URL, returns a page with the embedded post.
#
GWG Then it will reproduce the popup
#
GWG Wish I could think of a better way
#
GWG I like the Twitter intents
#
kylewm so bring up the embedded post and let them click 'like' on it?
#
kylewm and why not a php script to make an api call?
#
GWG kylewm: Yes
#
GWG kylewm: Wouldn' that require an App ID?
brianloveswords and erikmaarten joined the channel
#
kylewm hmm, you could use bridgy publish to post the likes...
#
GWG My reasoning is that, as a plugin, I want to make it so a user has to do as little as possible to get the functionality working.
#
GWG kylewm: That wouldn't work if other people are liking things.
#
GWG It would all look like I was
#
kylewm oh, duh
#
GWG In order to make certain things simple I have to come up with something complicated
#
GWG Go figure
#
kylewm bleh, i'll let you know if i think of anything
#
GWG kylewm: See what I mean?
#
kylewm yep
#
GWG Trying to stick to links and not buttons is hard
#
GWG kylewm: Why is Twitter the only one that makes it easy?
luxagraf joined the channel
#
aaronpk shoot, php-mf2 isn't resolving protocol-relative URLs
#
aaronpk oh there's an issue already https://github.com/indieweb/php-mf2/issues/35
dariusdunlap joined the channel
#
Loqi [@benwerd] RT @mattfeifarek: Just listened to this http://caseorganic.com/articles/2014/03/11/2/ on the #indieweb
#
Loqi nostalgic & hopeful Silos are for grain, not us! @mdoudoro… (http://twtr.io/mhNtGmVkJx)
fmarier joined the channel
#
aaronpk what the heck, the test works but fails when it's running on pin13.net
#
aaronpk confirmed, bug in php's parse_url which was fixed in 5.4.7
snarfed joined the channel
#
aaronpk i've gotta get rid of that server
#
aaronpk !tell barnabywalters I don't even know what to do about this... is it worth fixing for php 5.3 or not? https://github.com/indieweb/php-mf2/issues/35
#
Loqi Ok, I'll tell him that when I see him next
#
snarfed GWG: you might be interested in how bruce schneier does social buttons: https://www.schneier.com/blog/archives/2013/03/changes_to_the.html
#
KartikPrabhu anyone using email as a "new post" notification?
#
GWG I am, I love Bruce Schneier
#
aaronpk KartikPrabhu: what do you mean by "new post" notification?
#
KartikPrabhu people sign-up with their email and you send then an email about your new post... email subscription would the word I guess
#
KartikPrabhu I was just asked about this for my site, nice this person stopped using feed readers after the Google Reader collapse
#
KartikPrabhu s/nice/since
#
Loqi KartikPrabhu meant to say: I was just asked about this for my site, since this person stopped using feed readers after the Google Reader collapse
#
aaronpk crystal beasley just specifically asked me if she could get email updates about indieweb events
#
snarfed KartikPrabhu: brett slatkin does, http://www.onebigfluke.com/ (newsletter link)
#
KartikPrabhu hmmm newsletters seem different, but in the same vein
#
snarfed yeah, just batched
#
aaronpk ah nuts... jawbone events don't have a public URL unless they're explicitly shared from the app... time to reverse engineer this
#
snarfed always fun
barnabywalters joined the channel
#
Loqi barnabywalters: aaronpk left you a message 17 minutes ago: I don't even know what to do about this... is it worth fixing for php 5.3 or not? https://github.com/indieweb/php-mf2/issues/35
#
barnabywalters good evening
#
GWG Hi, barnabywalters
#
KartikPrabhu evening barnabywalters
#
barnabywalters aaronpk: RE the URL parsing fix, I’m indifferent about patching software to work with outdated platforms
#
barnabywalters if it can be fixed with minimal effort and minimal future maintenance cost, it might be worth it
#
barnabywalters but I’d rather leave it there as extra incentive to use up-to-date builds of PHP
#
aaronpk yeah I'm not quite sure the best way to handle that even
#
aaronpk just more incentive for me to finally update that server
#
barnabywalters aaronpk: best not spend too many brain cycles on it then :)
#
barnabywalters social responsibility notice: for anyone who somehow has not seen it, read some of https://twitter.com/hashtag/YesAllWomen?f=realtime
#
barnabywalters </PAS>
#
barnabywalters or, PSA, whatever
#
GWG barnabywalters: Is your Web Action plugin ever going to go mainstream?
#
barnabywalters GWG: I have no supernatural powers of note
#
barnabywalters so I have no idea
#
barnabywalters but TBH I would guess not
#
GWG barnabywalters: I meant, not require dev mode to install
#
barnabywalters GWG: ah, well that’s a much more useful question :)
#
barnabywalters are you asking specifically about Chrome?
#
barnabywalters it should already be very easy to install in Firefox
#
barnabywalters ah, or I might be confusing it with my other extension, Weave
#
barnabywalters I’ll add it to the firefox addon store
#
barnabywalters Chrome is more awkward as they make you pay to authenticate yourself
#
GWG barnabywalters: Yes, but I'd go for Firefox too
#
GWG barnabywalters: They do?
#
GWG Did not know this
#
GWG $5?
#
barnabywalters GWG: yep. When I was submitting Weave to a bunch of addon stores I got annoyed with it, esp. as the upload experience was much worse and you can’t run extensions in non-dev-mode without it, which I take huge exception to
#
barnabywalters I did eventually give in and was going to pay it, donating 2x to mozilla, but their system didn’t work with my card
#
GWG barnabywalters: I'd donate a few bucks to the convenience factor, but I don't want to give you my card.
#
GWG No offense intended, but I've never actually met you. That would be a bit foolish of me.
#
barnabywalters GWG: totally understandable :) I might give it another go
#
barnabywalters credit cards seem to be quite hit-and-miss
#
GWG barnabywalters: If you have a way for me to kick in, such as Paypal or such, I'll be happy to do so
#
GWG I just use Chrome on a bunch of systems, and to start disabling security measures, even overly protective ones is a pain sometimes
bnvk and snarfed joined the channel
#
snarfed barnabywalters: i'm working on representing rel-syndication in activitystreams, found the down/upstreamDuplicates fields, searched the wiki, and your name came up
#
snarfed small world :P
#
barnabywalters okay, web actions extension is now on the mozilla addon store: https://addons.mozilla.org/en-US/firefox/addon/web-actions/
#
snarfed ooh congrats!
#
barnabywalters it’ll probably get rejected but at least it’s there for the moment
#
barnabywalters snarfed: thanks! hah yep, small community
#
barnabywalters Weave got rejected by the mozilla reviewers because it changes content on a page, which is what it’s for
#
barnabywalters which is fine, but I wish they had left it up with a warning rather than disabling it completely and leaving an ugly 404 page
#
snarfed aww
#
barnabywalters it’s a good prod to fix my page about it though
#
GWG barnabywalters: I'd add to the description
#
GWG If you want to avoid rejection
jjuran_ joined the channel
#
barnabywalters GWG: thanks, have done so :)
#
barnabywalters also made my page about Weave non-buggy: http://waterpigs.co.uk/extensions/weave/