2014-09-17 UTC
bitwit joined the channel
# 00:05 johnmorton kylewm: No luck yet. Thanks for the help. Try more after dinner :)
# 00:10 Loqi Ok, I'll tell them that when I see them next
alexhartley joined the channel
KartikPrabhu joined the channel
KartikPrabhu joined the channel
alexhartley, KartikPrabhu1, alexhart_ and parzzix joined the channel
# 00:52 parzzix Good evening everyone...
wolftune joined the channel
# 00:58 parzzix Who has a self hosted known? How is it as far as customization, themes, and plugins?
scor joined the channel
# 01:17 kylewm and an option to override/augment the css with the "Site Styles" plugin
# 01:20 parzzix kylewm, thanks....I'm going to install on a droplet...trying to decide if know is the best choice.
# 01:21 parzzix Was thinking red wind maybe
# 01:21 parzzix i'm a bit indicisive
# 01:22 kylewm red wind would probably only be fun for you if you wanted to hack on it
# 01:24 kylewm I would bet most everyone here can identify with your indecision though
# 01:24 parzzix it would give me a reason to learn python..lol.. the main thing holdong me back on known is PHP
# 01:24 kylewm I'm not trying to sell you on Known, but fwiw, it's pretty, well-structured PHP
# 01:25 parzzix the wicki had allot of other options...to many to try to judge
# 01:25 kylewm parzzix: have you gone down the rabbit hole of static site generators already?
# 01:27 kylewm I think we can probably cut the wiki/projects down to 3-4 more realistic options
# 01:28 kylewm Known, Wordpress + plugins, Publify, Jekyll or other static generator + webmention.herokuapp
# 01:29 parzzix hmmm...choices...Can wordpress cover everything?
# 01:32 parzzix kylewm, checking out publify on github..seems demo site is down or buggered somehow
# 01:33 kylewm hmm... well, neuro` posted a docker image recently that got it up and running on my machine in a few minutes
# 01:33 parzzix really.... what is your sites addy? just to check it out
# 01:39 parzzix I'm just creating a droplet and will install it..just to play
# 01:39 parzzix love digitalocean
RichardLitt, bitraten3, dysfun_ and alexhartley joined the channel
# 01:52 parzzix kylewm, the more I look, known is the most well rounded option
paulcp, KevinMarks, Mark87, yobj, Aeyoun, rknLA, voxpelli_, Leeky, dietrich, mlinksva, benward___, DanToml, bigbluehat, jden, CaptainCalliope, protman, acegiak, leth, npdoty and b0bg0d_____ joined the channel
djmadeira, fmarier and alexhartley joined the channel
fmarier and addal joined the channel
# 03:25 aaronpk just realized that because i'm using the pingback->webmention gateway this will be an interesting test to see if the expiring webmention endpoint cuts down on any pingback spam
RichardLitt joined the channel
wolftune and alexhartley joined the channel
# 04:31 KartikPrabhu GWG: managed to have a local dev install with local copy/mod of SemPress in it
# 04:32 KartikPrabhu now to see if wagtail has better interface or to modify SemPress to be a artist portfolio+blog theme
# 04:33 KartikPrabhu GWG: making portfolio type site for a friend and secretly indiewebbing it ;)
lukebrooker joined the channel
# 04:36 KartikPrabhu my personal focus is to improve Bundle, but don't want to push that on others for now
# 04:37 KartikPrabhu is that MSFT cashing in on latest trending hastags or is there some substance? ^
techlifeweb and fmarier joined the channel
alexhartley joined the channel
# 05:08 aaronpk anyone want to try sending me a mention to test this for real?
# 05:15 aaronpk I added some logging to see how many errors I get, and I log the current time vs expiration time
# 05:15 aaronpk so sohuld have some interesting stats in a few months
# 05:15 bear I have the beginnings of a indie-stats crawler
# 05:15 bear i'm generating a json blob each day of site info and mf2 parsed results
# 05:16 aaronpk oh yeah I need to add that checkbox for indieauth.com
# 05:16 bear working on a simple flask app to allow an indie-auth masked simple endpoint to let folks ping the site to get added to the list
yakker joined the channel
KevinMarks_ joined the channel
alexhartley joined the channel
wolftune, ngronald_ and ShaneHudson joined the channel
# 06:58 KartikPrabhu bear: is it possible to see the indie-stats data for domains you have already parsed?
alanpear_, petermolnar, pfefferle and eschnou joined the channel
alexhartley, glennjones and pfefferle joined the channel
ShaneHudson joined the channel
# 08:01 KartikPrabhu pfefferle: I successfully installed a local version of WP+SemPress... took some doing though
DWGinNYC joined the channel
# 08:02 DWGinNYC Hi all, I assume this channel is simply for the webcamp -- is there another channel dedicated to the indie web movement in general I can join?
# 08:05 hmans KartikPrabhu, no, it's a completely different app, doing something entirely different. It just has the same name.
# 08:05 hmans But considering that it's backed by Twitter, Foursquare and Square, I guess I can't just let it slide.
# 08:07 DWGinNYC I've been talking about this for YEARS. Great to finally encounter an organized, or at least, impressed movement about it.
# 08:08 KartikPrabhu DWGinNYC: nice... start simple. but folks here are also very UI/UX oriented
# 08:08 DWGinNYC Heh, well I have a fundamentalist example to make my point, but even I use tools nowadays. :)
# 08:10 KartikPrabhu DWGinNYC: I see you already have Twitter etc.. links on your homepage. If you add a rel=me to them, you can log in to the wiki at indiewebcamp.com (and other IndieAuth sites)
# 08:12 DWGinNYC I'll try it, what do I need to do exactly? Well, generally, I can figure it out. Just point me in the right direction. You mean, edit my Twitter code to include rel = me?
# 08:13 KartikPrabhu DWGinNYC: the link that you have for Twitter, you can change it so <a href="twitterURL" rel="me"> etc... and that should suffice
# 08:13 pfefferle hey KartikPrabhu … any problems with SemPress or general installation stuff of WordPress?
# 08:14 KartikPrabhu pfefferle: no Sempress problems, just getting all the simlinks stuff so that WP actually sees the local version of the theme I want to edit
# 08:15 KartikPrabhu pfefferle: my plan is to edit SemPres so it acts more like a artist portfolio site, instead of a generic indieweb theme. If I think of some useful additions to SemPress itself I'll send a PR your way :)
krendil joined the channel
# 08:20 pfefferle KartikPrabhu nice, but please use Github instead of a PR… So that I can’t lose the message ;)
# 08:21 KartikPrabhu pfefferle: PR = Pull Request on Github :P I've already made a fork there
# 08:21 KartikPrabhu DWGinNYC: I was actually talking about the link on your homepage. I actually have no idea where one would put a rel=me in a Twitter embed
# 08:26 bear KartikPrabhu - oops - i'll fix the link
# 08:29 bear I need to add some sort of index page
# 08:29 KartikPrabhu DWGinNYC: nice... now if your twitter also link back to your homepage you can log in using your hompage into the wiki
# 08:30 KartikPrabhu bear: neat... just bear.im/indie-stats/kartikprabhu.com should be good right?
# 08:30 bear but i'm also storing them using timestamps
# 08:31 bear bear.im/indie-stats/domains/kartikprabhu.com
tantek joined the channel
# 08:31 bear i'll fix the code so it adds the history list to the main domain json file
# 08:32 KartikPrabhu bear: neat! why is the mf2 having relative URLs? are you using an older version of mf2py?
# 08:35 bear i'll clear the history and generate a new set now
# 08:36 DWGinNYC KartikPrabhu, ah, if you mean the link at my Twitter profile page, it kind of does, but, it does via a self-written URL shortener. But I can change that if that's the place you mean.
Sebastien-L joined the channel
# 08:49 bear heading to bed - KartikPrabhu if you take the "polled" item from the domains.json for a domain - and strip out the :, - and Z from the time stamp you will get the last polled data
# 08:49 bear so yours is 2014-09-17T08:39:08Z == 20140917T083908_kartikprabhu.com
# 08:49 bear I haven't settled on a storage pattern yet
# 08:50 bear bear.im_history == list of filenames of history
# 08:51 bear tantek - just trying out different ways of storing the polled data i'm generating
# 08:51 bear KartikPrabhu - I may be calling mf2 parser() wrong?
# 08:51 KartikPrabhu bear: will look into it soon. since you are heading to bed we can postpone this issue for a while
# 08:52 bear the code is in the repo if you get a thought or yea, we can talk further about it later
# 08:52 bear it's very much in thought-code-test-repeat mode
kensanata joined the channel
# 08:53 DWGinNYC had to use www.davethewebguy.com versus davethewebguy.com
bnvk joined the channel
# 09:05 tantek !tell gRegor` your Getting_Started edits look great - ship it. We can fix any styling issues in-place.
# 09:05 Loqi Ok, I'll tell them that when I see them next
# 09:05 DWGinNYC Thanks, I'm leaving window open to resume at sunrise -- thanks for your help!
petermolnar joined the channel
eschnou and barnabywalters joined the channel
# 09:52 tantek I'm not sure how to troubleshoot Bridgy in particular for this
# 09:53 tantek and then we can leave a message for snarfed,kylewm to take a look
# 10:15 alexhartley tantek: So I just post my issue there? Sorry new to indiewebcamp!
# 10:19 tantek we use our User: pages as a place to collect what we're "Working On"
# 10:20 tantek go ahead and create your User: page with name, URL, photo if you like, and add a == Working On == section with "Bridgy setup" and the details you noted
# 10:37 Loqi Ok, I'll tell them that when I see them next
emmak and friedcell joined the channel
parzzix joined the channel
# 11:13 parzzix On known, when someone replies to a tweet or post and I want to reply to comment, I can just do it from known correct?
# 11:14 parzzix do I still just do @theirname or do i use @their site name?
elf-pavlik joined the channel
# 11:27 tantek elf-pavlik: excellent! how's your IndieMark support level?
Sebastien-L joined the channel
# 12:06 elf-pavlik tantek, i think quite low :( possibly quite few people control a domain name :D
pfefferle, brianloveswords, scor and Mark87 joined the channel
pfefferle_ joined the channel
jgarber, tantek and verdi joined the channel
# 13:05 tantek that's ok, there's a lot of progressive enhancement positive thinkers here :)
# 13:07 jonnybarnes on a sidenote this annoys me, why does berlin get its own TLD but my city of Manchester doesn't. Not fair!
# 13:08 lewiscowper a capital of a major European country with a million more people in it than a city in the north of England?
# 13:08 lewiscowper London has a TLD
# 13:09 jonnybarnes however I would point out that "Manchester" has about 3.2million people
# 13:10 elf-pavlik I think farmersmarket.manchester wouldn't make a nice domain name... maybe airport codes?
# 13:12 elf-pavlik farmersmarket.man :D
# 13:12 lewiscowper jonnybarnes, according to wikipedia it's 2.7 million in greater manchester
# 13:13 lewiscowper but yeah, I think the Berlin/London/Paris TLDs make more sense to figure out where demand lines
# 13:14 elf-pavlik just pulling all the h-cards based on list of people's profiles... i didn't look at bridgy in depth, bnvk just showed it to me
# 13:15 elf-pavlik i made small PoC where I can aggregate profiles which use http://schema.org/Person since i publish this way, but would also like to suppor fetching profiles people publish with h-card
# 13:16 tantek pulling all the h-cards based on people's profile is just the *from*
# 13:17 elf-pavlik now it makes people fillup google form to get listed, i can already list myself with data from my website
# 13:17 tantek why not have that site parse people's h-cards directly from their URLs?
# 13:18 elf-pavlik makes sense, i thought bridgy does something similar to directory listings as well
# 13:22 tantek oh interesting - that page itself could use h-card
PierreO, chrissaad and bitwit joined the channel
Sebastien-L and djmadeira joined the channel
danlyke joined the channel
# 13:53 tantek barnabywalters: all this JSON-* one-off formats have been done 10 years ago as XML-* formats.
# 13:53 tantek it's just that developers, especially new ones, forget
# 13:54 tantek in 2004, XHTML1 was perhaps considered the most modern
bitwit joined the channel
# 13:54 tantek barnabywalters: XML culture liked to reinvent things, just as JSON culture does
# 13:54 tantek note the reinvention of plenty of vCard/h-card vocabulary in that jsonresume
pfefferle joined the channel
# 14:05 tommorris tantek: I had some fun on twitter with the JSON Resume guys. They've managed to completely botch so many things it's unbelievable.
eschnou joined the channel
# 14:11 elf-pavlik tommorris, link?
# 14:12 tantek I'm just going to assume that the jsonresume people are "new" and missed the spate of XML-* vocabularies.
# 14:18 tantek the Ruby parser is a bit behind and could use some help
# 14:18 tantek elf-pavlik: what's your preferred backend language?
# 14:18 elf-pavlik i like JS and Ruby
# 14:18 elf-pavlik lately mostly Node.js
# 14:21 tantek elf-pavlik: what does it need beyond the JSON structure already produced by parsers?
# 14:22 tantek elf-pavlik: microformats parsers produce a canonical JSON output
# 14:22 elf-pavlik tantek, i haven't looked at mf specific json format
# 14:23 elf-pavlik on first thought if i just want to reuse vocab i woldn't need to use it at all
# 14:23 elf-pavlik and them mix it with any other vocabs available thanks to RDF / JSON-LD
# 14:24 tantek elf-pavlik: you can do all that mixing already in microformats and in the JSON output from microformats
# 14:24 tantek so I'm not sure what else JSON-LD adds (or what you need from it)
# 14:26 tommorris tantek, elf-pavlik: I've been meaning to write some JSON-LD stuff for microformats.org parser output. wouldn't be too hard.
ben_thatmustbeme and chrissaad joined the channel
# 14:28 tantek btw elf-pavlik,tommorris since this is a generic vocabulary/syntax discussion re: microformats, we should probably do it in #microformats
snarfed, gRegor`, hugoroyd, DanToml, tommorris, lmjabreu___, mattl, b0bg0d_____, Aeyoun, hadleybeeman, benward___, jden, jancborchardt, Garbee, CaptainCalliope, kronda_ and Phae joined the channel
# 14:58 djmadeira I’m new to the IndieWeb movement; how do most people go about doing @mentions on their site without creating a twitter dependency?
# 14:58 djmadeira for example, if I wanted to do POSSE on my site and support mentions, but not have those all break if/when twitter goes out of business
# 15:01 djmadeira seems like most of the indieweb stuff is tied to domains; could I use @bennannovak.com as the mention, use indieauth if supported to find the twitter handle, and make that the link dynamically using a CMS?
pfefferle joined the channel
# 15:02 tantek djmadeira: you can send webmentions directly to other indieweb sites
# 15:03 djmadeira right, but if I want to go POSSE
# 15:03 djmadeira and I want to basically tweet from my site (notes)
# 15:04 tantek POSSE is only for when you want to specifically stay involved with one or more specific silos
# 15:04 djmadeira How should I handle @mention ing? should I at all? should it always be a rel=webmention
# 15:04 tantek to stay in touch with your friends on those silos
# 15:05 tantek the indieweb @-mention equivalent is simply linking to someone's home page and then sending a webmention to them
# 15:05 tantek no need for rel=webmention those links, in fact that's not what that means
# 15:07 tantek rel=webmention is for discovery of webmention endpoints
stream7 and tilgovi joined the channel
Mark87 joined the channel
# 15:34 Mark87 I'm looking at the github, and I don't know how to install it. I just want to require() it, call a fetch() function and get back some php objects representing the data. Can it do that?
# 15:36 tantek barnabywalters: perhaps you could stub that on the wiki?
# 15:36 barnabywalters it’s already on the readme, I don’t want to duplicate it because then it’ll go out of date
# 15:37 gRegor` Composer makes it even easier to install.
# 15:37 Mark87 Well that is much easier. Why does the Installation readme go on about Composer and GPG keys?
# 15:37 Loqi gRegor`: tantek left you a message 6 hours, 31 minutes ago: your Getting_Started edits look great - ship it. We can fix any styling issues in-place.
# 15:37 gRegor` What is composer?
# 15:37 barnabywalters Mark87: because for anything with more than one or two dependencies, composer simplifies things hugely
# 15:40 Mark87 thanks, now im gonna see about getting this running
Pea1 joined the channel
RichardLitt and chrissaad joined the channel
wolftune, tilgovi_, danlyke and willnorris joined the channel
# 16:12 danlyke KevinMarks, I'm using Perl's XML::Feed. I could use some sort of proxy for the h-* stuff, but my life really doesn't need more distributed potential points of system failure...
tilgovi and paulcp joined the channel
# 16:18 ben_thatmustbeme starting a new webapp to give us at least partial indiemark bits as well as some stats on all sites on irc-people
willnorris joined the channel
# 16:23 ben_thatmustbeme well at least be able to list who has rel=me auth providers, how many, who has what endpoints (auth, webmention, token, micropub)
# 16:24 ben_thatmustbeme who has h-feeds linked off the main page. then just curiousity bits for like 'most posts' etc
tantek joined the channel
# 16:25 kylewm jonnybarnes: nice work on the localStorage stuff! looks like you went without a library?
# 16:27 barnabywalters ben_thatmustbeme: that’s exactly what indiewebify.me was started as (an indiemark tracker) but I haven’t been able to give it the attention it deserves
# 16:27 bret a+ to tantek for a great IWCUK14 writeup! :)
# 16:27 jonnybarnes well I found a project called store that seemed to simplify things across various browsers
# 16:28 bear ben_thatmustbeme - that is the same thing that i've started but I would love to split the work up amongst others
# 16:28 jonnybarnes native JS seems to be getting very good these days, with things like querySelector and eventListener
# 16:28 barnabywalters jonnybarnes: yup, with a few little syntatic shortcuts it can be very usable indeed
# 16:29 bret i still dont quite understand how to use jquery
# 16:31 bear ben_thatmustbeme - if you tell me what format you want the data from each domain in I can start deliverying that to you now
# 16:34 Mark87 tantek do you publish h-entrys on tantek.com?
# 16:38 bret barnabywalters: gahhhhh $ does not belong in JS!
# 16:39 bret i dunno actually. I dont know jquery clearly
# 16:40 ben_thatmustbeme bear, my original plan was to just curl and grep for interesting bits, but since it seems there is certainly interest in it, perhaps we should just start a fork of indiewebify.me's git repo
# 16:40 bret $().()$$$()$’$().$.ajax($$()))$$$))($$)
# 16:40 Loqi bret meant to say: $().()$$$()$’$().$.ajax($$()))$$$))($$)jquery
# 16:41 bret $().()$$$()$’$().$.ajax($$()))$$$))($$)
# 16:41 bear ben_thatmustbeme - i have already started it earlier this week
# 16:41 Loqi bret meant to say: jquery().()jqueryjqueryjquery()jquery’jquery().jquery.ajax(jqueryjquery()))jqueryjqueryjquery))(jqueryjquery)
# 16:41 kylewm this is not really a fair criticism of jquery... yes modern browsers have querySelector now, but that was not always the case
# 16:41 bear I have python code that scans each domain and stores data for it
# 16:41 jonnybarnes kylewm: the only real criticism I had is jquery was not more modular
# 16:42 jonnybarnes maybe theyve changed that now but it used to be one ginormous js file
# 16:42 bear ben_thatmustbeme - I was unaware that indiewebify.me had a stats component (or wanted to have)
# 16:42 jonnybarnes i remember reading a list of all sorts of browser inconosistancies that it fixed
# 16:42 bear ben_thatmustbeme yes :) - and it's python because it seemed a natural fit for the backend/cron nature of the task
# 16:43 ben_thatmustbeme well, they already have a lot of the parsing for testing various components of the site, the code would basically be just running through all those things for each site and storing results
# 16:45 bear I think it can be done in two pieces - we can write the stat parsing bits in anything really and just put them in a dir with a simple calling pattern
# 16:45 bear then I can feed the data to it, it generates numbers and populates a datastore
j12t joined the channel
# 16:46 gRegor` Mark87: You may have already seen this, but pin13.net is a good tool for testing parsing microformats from any site.
# 16:46 ben_thatmustbeme php can be very frustrating for all the inconsistencies in it. I am constantly having to check the order of built in functions
elf-pavlik joined the channel
# 16:46 bear yea, that is why I keep going back to python - very procedural
# 16:47 bear ben_thatmustbeme - what i'm doing now is gathering the html and also running that thru mf2py - then storing both with a timestamp
# 16:47 Mark87 gRegor thanks! I'm trying to figure out what I want to do with h-entry feeds, using phpmf2 to parse them, when it suddenly stopped showing me h-entry. I must have an error somewhere
# 16:49 bear if we work out a simple "here process this payload" call type - we can make it so that anyone can contribute a stat generating code bit in almost any language
# 16:49 ben_thatmustbeme this started out of curiosity over who uses the headers vs who uses html <head> for endpoints
willnorris joined the channel
# 16:50 bear I'll make a change to get a json metadata file that points to a file for headers, content and mf2 for each domain for each timestamp
# 16:50 ben_thatmustbeme that would be cool.. just need some central data store, which it sounds like you are already working on
# 16:50 bear I was using flatfiles just to avoid the yak shaving of picking a database
# 16:51 bear yea, I kicked the "how to store the stats" question-can down the path
# 16:52 rascul well, as long as the file will only be accessed via python
# 16:52 bear given the number of php types in the indieweb community I was sticking to json
# 16:53 bear in my editor window right now is a super simple flash app to deliver a json payload of info for a domain for a date
# 16:53 bear so that anyone who wanted to consume stats would just give a date range and a domain and they would get back the json of stats
# 16:56 bear right now it's not a lot - haven't fired it up for real yet
# 16:56 bear but I wanted to store the html in case a new mf2 parsing feature is added and then we could backfill older stats
snarfed joined the channel
# 16:58 ben_thatmustbeme okay, but we would script it to basically say on a daily(?) basis, go through and grab the latest data for each site and process a data point using script (X)
# 16:59 ben_thatmustbeme bear, i've had success in the past just with a descriptor file of some sort as the go between to allow any language. just have some metadata file in whatever format you want for the python that will tell it how to run my script via shell, and how to interpret results
# 16:59 bear ben_thatmustbeme - your describing the same method I normally use
# 17:00 bear my preferred choice for that is bash with a known command line pattern
Mark87, benward___, gavinc, saurik, tantek_ and JonathanNeal joined the channel
# 17:01 bear and it would loop thru a directory of crunchers calling each as needed
# 17:01 bear it is a neat project that is forcing me to use new tools :)
ben_thatmustbeme, aaronpk, sparverius and catsup joined the channel
# 17:02 snarfed yeah, learning can definitely "hurt so good" sometimes :P
catsup, danlyke, wolftune, addal, mdik, alanpearce and bear joined the channel
# 17:04 ben_thatmustbeme bear, if you can get me a sample. i'll work on starting some various stats modules that interest me
# 17:05 bear ben_thatmustbeme I sure can - it may not be until later in the day
# 17:06 bear but yes, I'll get that to you tonight for sure
# 17:06 bear that list was seeded from the irc-people list
ShaneHudson joined the channel
# 17:07 snarfed bear: i just realized bridgy has another, bigger list of domains you could use
# 17:08 bear naa - let me grab it using the code and parse it
# 17:08 bear so it can be part of the seed code I have for irc-people
danlyke joined the channel
# 17:08 snarfed you'd have to fetch each person's user page individually :/ you can if you want to, but the dump would probably be easier
# 17:09 bear unless you frown on that I don't mind
# 17:10 bear because then I can focus on other code and start gathering data
techlifeweb joined the channel
# 17:17 bear ben_thatmustbeme ^^ that is what i'm planning on doing tonight and tomorrow
# 17:17 gRegor` Cool to see momentum on this
# 17:22 snarfed ben_thatmustbeme: glad to hear you're working on this too!
paulcp_ joined the channel
# 17:26 snarfed !tell http://alexhartley.me/ sorry for the trouble with bridgy! it's actually set up with your FB account fine; the yellow warning banner is just a UI bug that i'm still debugging. apologies for the confusion!
# 17:27 snarfed !tell alexhartley sorry for the trouble with bridgy! it's actually set up with your FB account fine; the yellow warning banner is just a UI bug that i'm still debugging. apologies for the confusion!
# 17:27 Loqi Ok, I'll tell them that when I see them next
# 17:27 ben_thatmustbeme oh btw, snarfed, brid.gy did not like it when i changed from http to https for everything
# 17:34 snarfed kylewm: if we were to make bridgy de-dupe http vs https links, i wonder if that would go in PPD or somewhere broader...
npdoty joined the channel
# 17:45 kylewm snarfed: definitely sounds like a posse part discovery bug
# 17:46 snarfed kylewm: hmm, maybe…but, straw man, if someone includes an http link to their post, and then we discover the same url via PPD except it's https, we should probably merge them, right?
# 17:47 snarfed or if someone includes two copies of a url in their silo post, one http and https. (very very unlikely :P)
# 17:47 kylewm it might send two copies right away in that case
# 17:52 aaronpk where can I find more about the progress of indie-stats?
# 17:54 ben_thatmustbeme this is the second time today i've hit this issue, its like it has some issue with curling https urls
# 17:54 aaronpk hm I want to see if bridgy has trouble sending me webmentions.. guess I need to POSSE something
# 17:56 bear aaronpk - yes, i'll be keeping that page up to date with my progress
# 17:57 aaronpk i like the idea of being able to add crunchers in any language
# 17:57 bear because of the interest in this i'm moving it from my laptop to one of my servers tonight
# 17:57 bear yea, it's a pattern i've used in my Ops world before and it's really flexible and scalable
# 17:57 snarfed aaronpk: any reason you think bridgy's having trouble? i don't see any recent failures on your user pages
# 17:58 aaronpk snarfed: I changed to expiring webmention endpoint last night
# 17:58 bear and we can add them to a repo and be able to auto-deploy any new hotness
# 18:01 snarfed bridgy does cache endpoints, and i saw the discussion about that
# 18:01 aaronpk I see "Also setting webmention_endpoint to..." in the logs
# 18:02 aaronpk oh nice are you blanking out any parameters that include "token" automatically or something?
# 18:02 snarfed no explicit expiration, but in practice it's only 1-2 hrs or so
# 18:02 aaronpk so this means if bridgy tries to send another webmention after 5 minutes it will fail?
# 18:03 aaronpk every page request generates a new token that is valid for 5 minutes
# 18:04 aaronpk interesting that the webmention spec never explicitly stated whether it's ok to cache the endpoint for a domain
# 18:04 aaronpk it implies that the endpoint should be discovered for each post
# 18:05 snarfed we could consider an extra optional param somewhere in discovery that indicates ttl
ShaneHudson joined the channel
# 18:05 snarfed yeah, understood. but running a service like bridgy, the thought of not caching also makes me :/
# 18:07 aaronpk interesting that that's a sensible thing to do in the indieweb world, but not reasonable for multi-user domains
# 18:07 aaronpk so without caching, you'd need to make an HTTP GET to each target you're sending to. is that actually a lot more work?
# 18:07 snarfed aaronpk: hmm really? domain kind of implies that a single party is running it, so they'd just as easily have a single endpoint, and distinguish users based on parsing the source url
# 18:08 aaronpk I dunno I could definitely see how twitter might want unique webmention endpoints per user
# 18:08 bear because it's implied that it could change because it's discoverable
# 18:09 aaronpk there's also the possiblity of caching the webmention endpoint per target URL
# 18:09 snarfed re extra GETs being more work, the vast majority of links bridgy sees don't have wm endpoints, so i want to at least cache that. i could tighten the cache ttl for endpoints i do discover though
# 18:09 aaronpk i think caching a "no webmention endpoint for this domain" flag for several hours is totally reasonable
# 18:10 aaronpk tho you may want to invalidate that cache if someone manually clicks the "poll now" button since it's likely to screw up onboarding new users
# 18:11 snarfed i have to admit, it seems like in practice, we don't yet have any of the problem(s) that expiring endpoints would solve, right?
# 18:11 snarfed …so i'd probably want to talk a bit more before tuning bridgy for them
# 18:12 gRegor` caching is hard, let's go shopping
friedcell joined the channel
# 18:13 bear this is where I would implement an inverse ttl based on how often you've seen the domain
ben_thatmustbeme joined the channel
# 18:15 snarfed bear: huh. so, the more wms you send to a given domain, the shorter you cache its endpoint?
# 18:16 bear the more wm's a domain receives the longer the cacche
# 18:17 bear bah - never mind - caching is hard, i'm going shopping
# 18:17 Mark87 not that my opinion matters, but thinking about it overnight, I don't like expiring endpoints. They just don't solve anything
# 18:17 snarfed right now at least, i vote for the optional explicit TTL or no-cache param. not convinced about the other options yet
# 18:19 aaronpk to be fair, right now the spec says (implies) that the webmention endpoint should be discovered for each URL you send a webmention to.
# 18:19 aaronpk so expiring endpoints may not fall entirely within the spec, but unique endpoints per URL certainly does
# 18:19 Mark87 If you give an explicit TTL, you basically lose all the value of the expiration, which is to prevent the buildup of large lists. With an explicit TTL, I can automate refreshing my large list.
# 18:21 snarfed Mark87: sure. at a high level, i just feel like we're overthinking a problem we don't actually have yet
npdoty_ joined the channel
# 18:22 snarfed i know this was discussed a lot over the last 2d; i don't mean to revive that discussion. i'll back away slowly now :P
# 18:22 Mark87 I wonder is there a link to discussion of the spec while it was still being developed?
# 18:22 Mark87 not the last two days, but webmentions itself
# 18:22 aaronpk i'm logging stuff about my webmention endpoint now so i'll be able to share some stats soon
# 18:24 Mark87 I feel like all the problems would be solved if the spec called for the actual html of the page being POSTed to the webmention endpoint. Then it doesn't have to perform a GET back at all. I'm curious if anyone discussed something like that
# 18:24 gRegor` My StartSSL is apparently SHA1withRSA
# 18:24 bear that's a amplification attack in the making
# 18:24 Mark87 the same moderation would be used on the posted page as would be used on the retrieved page
# 18:25 jonnybarnes its fine for now gRegor` but Google Chrome will start complianing
# 18:25 aaronpk you can't trust the contents of the POST, so you'd have to go fetch the real URL to verify it anyway
# 18:25 gRegor` reads up on how to upgrade
# 18:25 Mark87 well I have though about that, in what way couldn't you trust the POST contents.
# 18:25 bear it kinda boils down to this in my brain - the webmention discovery is via GET - if the server returns ETAG or cache headers then they should be honoured
# 18:26 aaronpk Mark87: you can't trust the POST contents because you can trivially imagine getting a POST request that says "google.com likes your post"
# 18:27 gRegor` I guess I have to wait for renewal of my cert, or pay a fee to revoke my SHA1 now.
# 18:27 Mark87 @aaronpk ahhh, I was thinking about trusting the content of the post, trusting the origin of the content is an issue I missed
# 18:27 bear hmm, did not know about that webmention github repo - going to have to review it to see if ronkyuu is compliant
# 18:28 bear gRegor` or generate a new one with a www. or somesuch
# 18:28 gRegor` Not sure I follow, bear. Wouldn't that leave the non-www current as SHA1?
# 18:28 bear I do believe I walked thru webmention.org when writing it
# 18:29 aaronpk gRegor`: you can just stop using your current one
# 18:29 bear when you generate a non-wildcard cert you can have alt name - some vendors allow you to get a new one that way without forcing you to expire it
# 18:29 bear or yes, you can always just get a new one and move on
# 18:29 gRegor` I'm a noob to all this. I'll have to look into it more later.
# 18:30 gRegor` I think StartSSL charges for the new one, because it means revoking the old one.
# 18:30 aaronpk i've definitely got a few abondoned startssl certs lying around because I don't want to pay to revoke them
# 18:30 bear some vendors - like startssl - won't let you replace a cert that has the same csr
# 18:30 aaronpk no you can just generate a new one at startssl, it just has to have a unique subdomain
# 18:30 bear aaronpk - right, that's what I was trying to say with the www. comment above
# 18:30 aaronpk and since I don't actually use "www." I end up with things like "web." or "mail." when I need a new one
# 18:30 gRegor` But will that work with my non-www domain? I don't have www
# 18:31 aaronpk gRegor`: the startssl certs are valid for the second-level domain plus one subdomain
# 18:31 aaronpk they usually prompt you to put "www." as the subdomain, but you can choose a different one
# 18:31 gRegor` Ah, so I could pick a subdomain even if I'm not using it.
# 18:32 bear I have www1 thru www23 on one of my certs
# 18:32 bear because I keep testing different configs
# 18:32 gRegor` Nice. Good to know.
# 18:33 bear I also finally just got my org validated because free wildcard certs from startssl was worth that effort
# 18:33 aaronpk bear: yeah I'm considering doing the same. free wildcard certs <3
# 18:33 gRegor` I might just wait till next year
# 18:34 bear for a personal domain I would just wait
# 18:34 gRegor` I still haven't even forced https on all pages, heh
# 18:34 bear the only reason i'm doing it is because we have a security related business so I have to make sure our SSL stuff is *tight*
# 18:35 gRegor` I do want to understand ssl better, so might set up sha256 on another domain in the meantime
# 18:36 bear boo - my site is no longer A+ because of the sha-1 cert
# 18:37 bear ok, enough yak shaving and rabbit holes - time for some $dayjob work :)
eschnou joined the channel
parzzix joined the channel
RichardLitt, brianloveswords, npdoty and willnorris joined the channel
# 19:15 npdoty Loqi, tell benwerd and erinjo congratulations on the awesome Wired article
# 19:16 gRegor` Has to start with !tell
# 19:16 gRegor` !tell npdoty Like this
# 19:16 Loqi Ok, I'll tell them that when I see them next
# 19:16 npdoty !tell benwerd and erinjo congratulations on the awesome Wired article
# 19:16 Loqi npdoty: gRegor` left you a message 22 seconds ago: Like this
# 19:16 Loqi Ok, I'll tell them that when I see them next
# 19:17 npdoty ha, hober lurks for months at a time until mocking the dude reference
# 19:19 parzzix Does anyone know of a detailed tutorial for installing known?
KevinMarks__ joined the channel
# 19:28 KevinMarks__ Let's see if I can post an indie event and POSSE copy and edit the wiki on my phone while in line at the dmv
npdoty joined the channel
# 19:35 parzzix kylewm, I just checked out your site...red wind looks nice. Don't know if i'm technical enough to get it going though
# 19:36 kylewm parzzix: thanks, and yeah it's not really ready for general consumption yet... probably only interesting if you want to write a lot of python
# 19:37 parzzix lol...I would like to learn python, but probably not the best way.
# 19:43 kylewm ^this dude does a *great* job making it clear that indieweb isn't just known
# 19:45 kylewm they must've done a good job explaining that on TWiG
# 19:45 snarfed hey lazy indieweb: i'm looking to convert rendered HTML whitespace (specifically line breaks) to the plain text equivalent
# 19:45 snarfed e.g. if you have an article rendered in a browser, possibly including <p>s, <br>s, lists, etc., i'd like to convert that article's html to the equivalent plain text with line breaks that looks as similar as possible
# 19:45 snarfed (…aaand i don't want to embed a browser in my code to do it)
kbs joined the channel
# 19:46 bear yep ^^ was just typing kylewm's link also
# 19:47 ben_thatmustbeme well br,s and extra spaces are possible but at some point things could be styled by css
# 19:47 snarfed bear: your markdown parser renders to text as well as html?
# 19:47 bear yea, this type of conversion is always good for isolated chunks
# 19:48 bear oh - hah - I use html2text for that also!
# 19:49 snarfed in a single, self-contained python lib no less. better than i dared hope
# 19:49 bear yea, it's a great module - aaron really knew how to code
# 19:50 kbs apropos of nothing, and for the few people here who still use PGP :-) another little tool/idea to do the keybase "lookup-by-social-profile" features (without needing a separate service) over at https://github.com/kbsriram/keypan
# 19:51 kbs idea happily stolen from rel=me hacking efforts, so thanks (I think! :)
# 19:52 kbs hey snarfed and bear :) haven't given up my obsession with pgp yet, I guess
brianloveswords joined the channel
petermolnar joined the channel
# 20:03 kylewm yeah, it's weird that it doesn't say it's from "GitHub"
# 20:04 jonnybarnes so Im going to have to call Apple to get iMessage and FaceTime working on my laptop again :(
# 20:04 bear i'm constantly going "why are they emailing me about brid.gy...?"
# 20:06 bear wow - that aaronpk dude is always 3 steps ahead of us
# 20:07 bear if I hadn't got distracted by work...
# 20:08 kbs haha, aaronpk does think up cool stuff :)
# 20:09 kylewm amusingly i think aaronpk only gets github notifications via IRC
gRegor` joined the channel
# 20:12 bear I so need to get that going again, but with xmpp instead ;)
# 20:13 snarfed oh god no, for the record, i am *not* advocating that :P
wolftune joined the channel
# 20:30 elf-pavlik snarfed, hi :)
# 20:31 elf-pavlik nice work with: bridgy !
# 20:31 elf-pavlik i guess you need to deal with many APIs ?
krendil joined the channel
# 20:32 elf-pavlik searching for video...
# 20:35 elf-pavlik i recommend checking it out, i see it as nice way for programmable interaction
# 20:36 elf-pavlik if you like interactive playgrounds
# 20:38 elf-pavlik BTW how can i merge those thow accounts on bridgy?
# 20:40 snarfed there's no way right now. is there something specific you wanwhat would you want hmm. what do you mean by merge?
# 20:40 elf-pavlik i guess i don't understand this bridging concept... Branan just showed it to me
# 20:42 elf-pavlik we need to ask him to implement webmention :D
# 20:43 snarfed elf-pavlik: looks like you're set up on bridgy ok. your next step is to add webmention to one of your web sites.
# 20:44 elf-pavlik i think i better do it even just to learn it and provide in polyglot manner!
# 20:46 bret damn snarfed, thanks for saving me $500
# 20:48 elf-pavlik happy to see you implemented AS lib!
# 20:48 bret so brid.gy basically taskes AS data from activitystreams-unofficial and layers on a uf2 and wm?
# 20:49 snarfed a-u actually also does the to/from uf2 part itself
# 20:50 snarfed go for it! also available as a REST service, per silo
# 20:50 snarfed barnabywalters uses that to get twitter streams inside shrewdness
# 20:51 Mark87 elf-pavlik you're blowing my mind. This is something I've been dreaming up for a few months now, but using html5 microdata
# 20:51 bret your like the john macfarlane of social networks (pandoc author ;p)
# 20:52 elf-pavlik i think we could get by with simply defining JSON-LD context and frame
# 20:53 gRegor` "There have been no new Microformats released in the last 5 years and the mailing list traffic has been almost non-existent for around 5 years. From what I can tell, most everyone has moved on to RDFa, Microdata, or JSON-LD." Lulz
# 20:54 elf-pavlik gRegor`, i prefer RDFa over Microdata, also my homepage renders JSON-LD to RDFa which makes sweet templating engine! i need to pull out some stuff from http://viejs.org/
# 20:56 snarfed elf-pavlik: what's your homepage? wwelves.org? that just hangs for me :/
# 20:57 elf-pavlik hosting gifted by friends but occasionaly it goes down :(
# 20:57 elf-pavlik once my website gets back up you can also Content Negotiate for application/ld+json
# 20:58 aaronpk elf-pavlik: I'd really like to see some examples of this stuff on your own website. I hope you can get the server back up soon!
# 20:59 elf-pavlik can i with microformats make relation saying "web page with list of all people i know" ?
tecgirl joined the channel
# 21:01 kylewm I was wondering the same thing yesterday, whether there is a rel= value for an XFN page
KartikPrabhu joined the channel
# 21:03 kylewm (i.e., so you could find the contacts list from someon's homepage)
# 21:04 elf-pavlik kylewm, have you looked on linke to hydra:Collection i just shared?
# 21:05 elf-pavlik aaronpk, snarfed, Mark87 could you point me to someone participating in #indiewebcamp who digs and even better likes RDF? :)
RichardLitt joined the channel
# 21:07 Mark87 elf-pavlik I'm new here, but im sure somebody around here is interested in RDF
# 21:07 elf-pavlik Mark87, cool!
# 21:08 elf-pavlik i wonder how far i could get with complex datasets if i try to model them with microformats ...
# 21:09 elf-pavlik gRegor`, ha, good point!
# 21:10 elf-pavlik tommorris, cool!
# 21:12 KartikPrabhu elf-pavlik: I used to write RDF but then found that microformats is much easier and does the same thing
# 21:13 elf-pavlik KartikPrabhu, do you know examples of some complex datasets which use microformats? can construct graph from such data?
# 21:14 KartikPrabhu elf-pavlik: never needed to do that. Microformats only has stuff that people want to use on websites
# 21:14 KartikPrabhu also what is a complex dataset example? I am sure I can write it in a microformats syntax ;)
# 21:15 aaronpk how about start with something that is published on your site, elf-pavlik
# 21:15 elf-pavlik soon i'll add places i visited, events participated in, people i know, wishlists of things i can give or receive etc
# 21:15 aaronpk theoretical examples are not super useful, so I'm interested in things that you have actually published
# 21:16 elf-pavlik no content negotiation though ...
# 21:16 KartikPrabhu I am sure I have theoretical dataset examples that even RDF would find complex
# 21:16 elf-pavlik KartikPrabhu, of course but does it help us? ;)
# 21:18 elf-pavlik KartikPrabhu, I also need to publish data from my usage of public transport
# 21:18 KartikPrabhu elf-pavlik: actually most fields in that example you gave me have mf2 counter parts. Person = h-card Location = h-adr, id = uuid, name = p-name image = u-photo
# 21:19 aaronpk that html page could easily be turned into structutred data with a few choice mf2 classes! and if there isn't an existing rel value or property name then those would make great suggestions
# 21:19 KartikPrabhu elf-pavlik: what is RDF helping you with, that some HTML+mf2 couldn't?
# 21:19 elf-pavlik well, how do i express that i took trip on particular bus?
# 21:20 aaronpk so the real question is what do consumers of this data need to do with the data? because you can always just write a sentence and people can read it
# 21:20 KartikPrabhu elf-pavlik: express for whom? people? then just write it as aaronpk said
# 21:20 elf-pavlik analyze my usage of public transport to run it agains my records of contributions to such subsystem
# 21:20 aaronpk that's why there haven't been many new vocabularies developed for indieweb conversations to happen, because it all happens at the people layer.
# 21:21 KartikPrabhu elf-pavlik: if you want to use it only for youself then just make your own mf2 classes with h-x-* , with x denoting experimental
# 21:21 elf-pavlik KartikPrabhu, but why wouldn't I use RDF ?
# 21:22 elf-pavlik as you see people already do transit vocabs in RDF
# 21:22 KartikPrabhu I wouldn't because it is complicated to author compared to mf2 wirting
# 21:23 aaronpk if anyone were to reply to that and show the reply context, they'd render it as a note from me that says "Biked 2.2 miles in 11 minutes"
# 21:23 aaronpk but if they wanted to, could actually parse the individual data on the page, like avg pace, total distance, heart rate, etc
# 21:24 elf-pavlik aaronpk, i quess you know QS xtof :)
# 21:24 tommorris I publish RDFa on tommorris.org. Take a look elf-pavlik. ;)
# 21:25 KartikPrabhu aaronpk: that is a neat example! maybe you can give a image fallback of the map itself as a e-content or something?
# 21:25 elf-pavlik tommorris, plus Mozilla Persona - awesome!
# 21:26 aaronpk and he didn't even need to know about my "bike" vocab
# 21:28 elf-pavlik "distance": [ "2.2 mi"] so i still need to parse text to get data :(
# 21:28 elf-pavlik what data type you use for duration?
# 21:29 aaronpk other measurements are hard because you have to include the units
KevinMarks__ joined the channel
# 21:31 aaronpk don't give me a spec for web standards as a pdf lol
# 21:33 tommorris Can we get W3C to ratify a standard described by a Flash SWF file? ;)
# 21:34 KartikPrabhu why are units hard? <span class="h-measurement"><span class="p-type">length</span>: <data value="3" class="p-value">3</data><data value="m" class="p-unit">metres
# 21:35 KartikPrabhu seems easier than parsing the ISOstuff which was a nightmare while adding to mf2py
alexhartley joined the channel
# 21:36 elf-pavlik KartikPrabhu, i would use URI for meters
# 21:37 KartikPrabhu metres is a SI standard unit of length.... no need to use a URI for that
# 21:38 elf-pavlik what if someone abbreviates them to just 'm'
# 21:39 aaronpk KartikPrabhu: i think you got them backwards, <data value="meter" class="p-unit">m</data>
Pea1 joined the channel
# 21:40 aaronpk SI doesn't actually require one or the other, says to use the local version
# 21:41 elf-pavlik needs english++ ;)
# 21:43 KartikPrabhu elf-pavlik: we have been doing quite some serious collaboration using mf2 :)
# 21:43 elf-pavlik KartikPrabhu, great to hear it! :)
# 21:44 elf-pavlik KartikPrabhu, awesome! but i still don't see it as *very complex* data
# 21:44 Mark87 watch the video elf-pavlik posted the tweet about earlier. its good so far.
# 21:45 elf-pavlik Mark87, REST ?
# 21:45 aaronpk I don't see "*very complex* data" as a good thing :)
# 21:46 aaronpk if your data is that complex maybe you should work on simplifying the data model first ;)
# 21:46 tommorris I should start a website: Tom drunkenly reviews W3C specs.
# 21:46 elf-pavlik tommorris, LOL ;-)
# 21:46 elf-pavlik + screencast
j12t joined the channel
# 21:48 elf-pavlik KartikPrabhu, let's say we continue conversation once i have more bit more complex data published using RDF ?
# 21:48 elf-pavlik and really impressed by aaronpk metrics! :)
# 21:49 elf-pavlik i plan to publish very detalied personal consupmtion/contribution logs
kirilind joined the channel
# 21:52 elf-pavlik hi kirilind :)
# 21:53 gRegor` tommorris: My Drunk W3C, a la My Drunk Kitchen
# 21:54 tommorris The only problem with drunkenly reviewing w3c specs is how I get my laptop past security in da club.
# 21:57 elf-pavlik afk, rolling...
j12t_ joined the channel
# 22:02 elf-pavlik binbasti, jump jump!
paulcp joined the channel
KartikPrabhu joined the channel
# 22:07 KartikPrabhu elf-pavlik: I suggest that publishing data is not useful. It is one-sided. If anyone else is parsing and doing something with the data you publish that'll be agreat discussion
# 22:07 elf-pavlik KartikPrabhu, good point!
# 22:07 aaronpk so this is interesting, in the past 17 hours I have received 2 legit webmentions, 1 junk test webmention from myself, and 290 spam pingbacks to endpoints from older than 5 minutes
# 22:07 KartikPrabhu that is the thing that converted me to mf2 frankly, that both publishing and comsumption is easy
# 22:09 KartikPrabhu and as a personal anecdote about how easy it is to comsume mf2. I could contribute to mf2py in less than half a year of learning python
# 22:09 aaronpk oh interesting, a lot of the invalid webmention/pingback requests are GET requests
# 22:10 tommorris KartikPrabhu: I am sure Python rdflib has issues that need work. ;)
# 22:10 rascul KartikPrabhu i was processing mf2 with beautifulsoup4 before i discovered (or maybe someone told me about? mf2py
# 22:10 ShaneHudson aaronpk: I need to sort that out too, last week I had an issue where indiewebify.me spammed me. Need to do a duplicate check or something
# 22:10 KartikPrabhu tommorris: I have no motivation since I don't intend to consume RDF... selfdogfooding and all that
fmarier joined the channel
# 22:14 Mark87 wouldn't the most useful thing be to actually have a webmention form at the endpoint?
# 22:15 bret elf-pavlik where did you get the sweet avatar painting/drawing?
# 22:16 Mark87 @aaronpk KartikPrabhu haha I saw kartik's page and immediately thought of that
# 22:16 KartikPrabhu Mark87: I have a webmention form at the end of articles but will have to modify it to have a source field
# 22:17 bret Heh nice! Is there an english transaltion? or just google translate
# 22:17 Mark87 KartikPrabhu I need to find a way to implement webmentions somehow I think
# 22:18 bret Mark87 lots of ways to do that, but still a bit of work from scratch
# 22:19 Mark87 KartikPrabhu bret I don't have a blog because I'm not much of a writer, but I have a feed reader I wrote from scratch and It would be cool to use webmentions to reply to articles I like from twitter or a new blog if I get one
# 22:19 elf-pavlik KartikPrabhu, first to learn how the plumbing works
# 22:19 Mark87 bret nope. I really don't write anything haha. I just read
# 22:20 Loqi KartikPrabhu meant to say: in my experience every question has a wiki page :)
alexhartley joined the channel
# 22:20 Mark87 bret i'm an infrequent commenter I guess, but sometimes I like to interact
# 22:20 bret Mark87 well commenting with your own site is still writing, and definately a rad development. maybe see if you can play with Shrewdness?
# 22:21 gRegor` What is Shrewdness?
# 22:21 bret barnaby is writing a feed reader that lets you comment, but it needs a micropub endpoint
friedcell joined the channel
# 22:21 gRegor` From what barnaby said the other day, it's invite-only right now.
# 22:21 bret im not sure if there are any getup and go micropub endpoints
# 22:21 gRegor` His superfeedr bill is racking up, so not opening it up more at the moment
# 22:21 elf-pavlik KartikPrabhu, good point again! once i read it all i'll make sequence diagram for future generations ;)
# 22:21 bret ShaneHudson i forget syntax but google mediawiki redirect
# 22:22 gRegor` aaronpk beating us to things, as usual
# 22:22 bret elf-pavlik what do you like to use to make sequence diagrams?
# 22:23 Mark87 I think the first thing I need to do is implement login with indieauth
# 22:23 elf-pavlik gRegor`, i'll mark it as TODO
# 22:23 j12t_ The "semantic web" has been superseded. Don't worry about it :-)
# 22:24 gRegor` semantic web 2.0?
verdi joined the channel
# 22:25 KartikPrabhu Mark87: in the sense that no one actually does the semantic web thing anymore do they?
# 22:25 elf-pavlik calls tommorris for help :'(
# 22:25 Mark87 KartikPrabhu ahh, well I would say no one ever really started using the semantic web
# 22:26 Mark87 what is indiewebcamp if not a semantic web movement?
# 22:27 Mark87 KartikPrabhu well that's just nice manners really
# 22:27 Mark87 otherwise every coder and their pizza eating friend would be in here clamoring for everyone to adopt their half-baked idea for a specification
# 22:27 KartikPrabhu so let those good manners serve as motivation for you to post notes Mark87
# 22:28 Mark87 I found my old wordpress blog and was actually amazed at how much I did write in it
# 22:29 Mark87 my problem is I have a chicken and egg issue. I don't feel like writing to an audience of zero, but I won't have an audience if I don't write.
# 22:30 Mark87 KartikPrabhu haha ok I get the point. i'll start something. Maybe i'll start using my Known page I started after watching TWiG
# 22:31 KartikPrabhu ShaneHudson: have you started accepting webmentions on articles again?
friedcell joined the channel
# 22:31 ShaneHudson KartikPrabhu: Ah yes I have, I forgot to tell you! That *should* be working properly now :)
# 22:31 bret Mark87, i started with 0 audience, and still have a small audience, but I try to capture things that I am interested for myself, and that others might find interestig or useful too, like if I solve a problem that required putting together of many other resources
# 22:31 rascul Mark87 i write to an audience of myself, that way i know somebody reads it
# 22:32 ShaneHudson KartikPrabhu: Although still some improvements to make. And sending webmentions is quite buggy at the moment
# 22:32 bret the main thing is creating something of value for yourelf
# 22:32 ShaneHudson Hmm I need to open source my webmention plugin actually, will do that now
# 22:32 bret or maybe those around you that show interest
# 22:34 Mark87 bret rascul i'll set a goal maybe, write at least a tweet's-worth a day
# 22:34 bret has anyone tried to use twitter as a datastore?
# 22:34 bret Mark87 I try to only write when have something to say
# 22:34 rascul Mark87 if you have nothing to write about then i wouldn't bother writing
# 22:34 bret but usually i dont write it down then forget
# 22:35 ShaneHudson KartikPrabhu: Same here, odd. Will have a look to see what's going on :)
# 22:35 bret elf-pavlik hehe seems like a funny thing to try and do
# 22:35 KartikPrabhu cool! also maybe consider treating webmentions to www. and no-www as the same
# 22:36 bret im feeling the need for automatic posse in gitpub really soon
# 22:38 tommorris I have argued the toss over RDF and SemWeb in the past, elf-pavlik. Now there are things that occupy a more central part of my life. :)
# 22:38 Loqi microsyntax refers to short text conventions for conveying specific semantic meaning, such as an "@" prefix indicating a (user)name, or "#" prefix indicating a hashtag (both conventions popularized by Twitter) http://indiewebcamp.com/microsyntax
# 22:40 elf-pavlik #indiewebcamp rocks the house :D
# 22:42 KartikPrabhu aaronpk: could this be solved? using "what is" with a reference to a nick?
# 22:42 techlifeweb I write stuff down on my blog mostly for me. I figure if I'm having the problem then someone else might as well. Kind of funny how often I've searched my own stuff.
verdi joined the channel
alexhart_ joined the channel
# 22:48 kylewm the last salesforce hackathon i heard about didn't get such good press
alexhartley joined the channel
petermolnar joined the channel
jgarber and verdi joined the channel
lukebrooker joined the channel
# 22:53 elf-pavlik needs to catch some sleep - 1AM CEST
# 22:55 elf-pavlik night bret, @all great vibes here, i look forward to hang aournd more :)
indie-visitor joined the channel
# 22:56 Loqi Welcome, indie-visitor! Set your nickname by typing /nick yourname
# 22:57 RaspberryPiGuy thanks!
# 22:57 RaspberryPiGuy Would anybody be willing to help me setup Known on my LInux Mint box?
# 22:57 RaspberryPiGuy I'm having trouble finding sufficient documentation
# 22:57 kylewm bret: tl;dr is that they gave the $1 million dollar prize to someone who had been a salesforce employee for 9 years and had demoed the app long before the hackathon started, and then to deal with allegations of cheating gave a second $1 mil prize to developers from a company salesforce invested in
# 22:57 KartikPrabhu RaspberryPiGuy: maybe others who have setup Known would know.... kylewm?
# 22:59 kylewm you don't have to install from git anymore, as there is a proper zip release now on withknown.com
# 23:00 RaspberryPiGuy actually I have not. I've read what was basically a repost of that though.
# 23:00 bret kylewm salesforece rubs me the wrong way. I hardly know anyting about them, but the whoe "no software" thing really bugs me
# 23:00 RaspberryPiGuy I'll try to follow those instructions and report back. thanks!
# 23:00 bret Ha KevinMarks! the odds are in your favor so far
# 23:01 bret KevinMarks interesting, I read it as "just use our webapp!"
j12t joined the channel
wolftune joined the channel
# 23:02 RaspberryPiGuy well I was using nginx at first. I managed to get to the point where I specified the database info, but was getting errors connecting to it
# 23:03 kylewm ok, did you create the db and a user and grant the user privileges on the db?
# 23:03 RaspberryPiGuy I'm using apache now though since nginx was confusing the crap out of me
# 23:04 RaspberryPiGuy no, I could not create a database. I'm a total newb when it comes to databases.
reedstrm joined the channel
alexhart_ joined the channel
# 23:10 KevinMarks "no software" was a slogan defined against Oracle/SAP etc who sell you software for huge $ then maintenance contracts for it to keep it up to date
j12t_ and alexhar__ joined the channel
# 23:11 kylewm RaspberryPiGuy: so on the command line do "mysql -u root -p" and enter your password.
# 23:12 kylewm GRANT ALL PRIVILEGES ON known.* to 'yourusername'@'localhost' IDENTIFIED by 'yourpassword';
# 23:13 RaspberryPiGuy kylewm: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)
# 23:13 kylewm and then you have to load the schema... mysql -u yourusername -p < schemas/sql/mysql.sql
# 23:14 kylewm RaspberryPiGuy: I think that means mysql isn't running
tantek and alexhartley joined the channel
# 23:18 RaspberryPiGuy kylewm: it says unrecognized service. Do I need mysql-server installed? synaptic won't let me install it because it can't resolve dependencies. I'll try mongo if that's easier.
# 23:18 RaspberryPiGuy I do have mysql-client installed though
# 23:19 kylewm I think you're better off trying to get MySql working than using Mongo
# 23:20 RaspberryPiGuy kylewm: ugh. ok. I'm determined to make this work.
# 23:22 kylewm RaspberryPiGuy: let's go to DM for debugging linux stuff
# 23:23 RaspberryPiGuy ok how do I do that?
tantek, willnorris_, danlyke, jgarber, pauloppenheim, wolftune, brianloveswords and catsup joined the channel
# 23:54 tantek what kind of UI do people have for deleting comments (e.g. those received via webmention?) ?
scor joined the channel
# 23:57 gRegor` tantek: I don't have delete yet, but I do have a checkbox for "Display on site" so I can remove them from the frontend.
# 23:58 tantek so you store one bit in your data per comment saying whether to hide them or not?
# 23:58 tantek (I think that has to be done manually with Known for events)