#indiewebcamp 2015-10-19

2015-10-19 UTC
ventilateur, JasonO and tantek joined the channel
#
tantek !tell tbrb just making sure, no HWC Edi 2015-10-20? https://indiewebcamp.com/events/2015-10-20-homebrew-website-club
#
Loqi Ok, I'll tell them that when I see them next
#
tantek working on some indieweb slogans
#
tantek what do you all think of "Be a part of the open web, with your web site."
#
tantek ?
[shaners] joined the channel
#
[shaners] I’m luke warm on “be” as the verb. Feels too passive.
#
[shaners] Not sure a better one though.
#
[shaners] “Join” only works at the start. Isn’t the verb of ongoing.
#
tantek right, be is inclusive of joining, and staying active
#
tantek context: https://wiki.mozilla.org/WeeklyUpdates/2015-10-19#Wednesday.2C_21_October
#
tantek and previously was (improves upon) " Be independent with your web browser and your web site. ", e.g. https://wiki.mozilla.org/WeeklyUpdates/2015-10-05#Wednesday.2C_07_October
#
tantek which was a play upon the # ChooseIndependent marketing campaign that Mozilla did for a while, implying choose an independent browser as it were
#
[shaners] I get it / agree. It’s the right semantics. It just *feels* meh to me. :neutral_face:
#
KartikPrabhu "Open web needs you" with indiewebcat pointing intensely
#
tantek shaners - feel free to brainstorm alternatives, even those that don't seem "as good" - perhaps on a page like /slogan
#
[shaners] :thumbsup::skin-tone-2:
#
aaronpk lol those are always hilarious when they come through IRC. not really sure how to handle that. I guess at least it's descriptive.
#
tantek also in the logs
#
tantek is struggling with /Litespeed sending HTML as text/plain by default bug again
#
aaronpk if they're using the same names as twitter's emoji I could render them as images https://github.com/twitter/twemoji
#
tantek wow this is really confusing. h2vx.com and dev.h2vx.com use the exact same .htaccess now, and one serves its static index.html home page as text/html, and the other as text/plain. WTF
#
aaronpk they are both returning text/html for me
#
tantek huh - ok apparently I forgot to shift-refresh :/
#
aaronpk https://gist.github.com/aaronpk/d6c2e1e01df66f973689
#
tantek ok so that's one bug fixed
#
aaronpk curl++
#
Loqi curl has 2 karma
#
tantek speaking of indiewebify.me being bottlenecked on too few people being able to update it
#
tantek same problem with h2vx. :(
#
aaronpk we could try running it on google appengine too
#
[shaners] tantek: maybe move it the indiewebcamp org on github and add some more people as owners/admins/etc
#
tantek except it needs PHP with XSLT enabled/installed
#
tantek shaners it's already on the microformats org on github with multiple people able to commit
#
[shaners] run it on heroku and add github -> heroku deploys hooks?
#
bear where is it hosted?
#
aaronpk hmm yeah no xslt on appengine https://cloud.google.com/appengine/docs/php/#PHP_Enabled_extensions
#
[shaners] oh good. didn’t realized. :thumbsup:
#
tantek bear - currently on the same host I'm hosting tantek.com on
#
tantek I had to setup it quickly when Technorati's conversion service went down
#
aaronpk i'm tempted to make a thing where anyone can sign in with indieauth and be added as a member to get commit access to repos on github.com/indieweb
#
tantek ++ on that
#
aaronpk commit+deploy for whatever services are rigged that way
#
bear so it's the deploy that is the issue for h2vx - that depends on you want to enable access
#
bear s/on you/on who you/
#
Loqi bear meant to say: so it's the deploy that is the issue for h2vx - that depends on who you want to enable access
#
tantek not just who, but its runtime requirements make the "where to host" a non-trivial question
#
bear or the build could be to generate a tarball, store it in github releases for that repo and then just run a cronjob to look for a fresh copy
#
tantek bear - I got to the point of defining the deployment problem I wanted to solve (with easy UI/UX), and then was never able to (heard of no good solutions, ironically) for https://indiewebcamp.com/deployment#Goals
#
tantek specifically this was with/for H2VX as a use-case
#
tantek the easy update/revert is the key
#
tantek AFAIK no one actually handles that well
#
bear I can solve up to item #4 list right now
#
bear but it's the portable across web hosts that is a sticking point
#
tantek you mean you can only solve it for one particular webhost?
#
tantek that's no good
#
bear well, for a class of hosts - i.e. those that allow the container method
#
bear because all of what leads up to #4 can be done using something like CoreOS's rolling updates via versioned docker containers
#
kevinmarks h2vx being xslt also limits the number of people who can hack on it
#
tantek yeah
#
aaronpk lol yeah. that's the main reason I didn't even attempt to add stuff to it a while ago
#
tantek at this point might be worth rewriting it as a mf2 parser + extract the generating code from the XSLT
#
tantek bear - containers sounds like a heavy solution to what "should" be a "simple" thing - i.e. point to a specific github repo version/checksum and deploy, and have it keep a history so you can simply say "revert" and it goes back to previous one
#
bear it's actually a lightweight solution to the problem of service isolation and networking
#
bear :)
#
bear it is all about which side of the IT/Ops line you are standing on
#
tantek so it solves additional problems - that I get
#
bear I wouldn't expect a web-site-deploy use case to have to grok containers, I was just suggesting that is something I would use to implement that use case
#
tantek hence "heavyweight"
#
tantek it does more than is strictly needed for the use-case
#
kevinmarks there's a nice chunk of hcard to vcard code in feedparser
#
tantek nice chunk doesn't really cut it
#
kevinmarks https://github.com/kevinmarks/feedparser/blob/develop/feedparser/feedparser.py#L2468
#
tantek h2vx's generation code has been fairly well battle tested with multiple vcf and ics consumers
#
bear that is almost a straw man argument because you will always have a certain amount of ops-related tasks to do in order to have someone deploy a website on someone elses server
#
tantek regressing on that is the reason not to switch from it
#
kevinmarks plenty of tests https://github.com/kevinmarks/feedparser/tree/develop/feedparser/tests/microformats/hcard
#
tantek no amount of tests can compare to actual live use (over years) in the wild
#
tantek nor even assess how close the tests might be to real world wild usage
#
tantek bear - regardless - I defer to at a minimum capturing your solutions - if you could update /deployment with those suggestions - greatly appreciated!
#
tantek FWIW this is the source of the error:
#
tantek preg_match_all( '/FN.*?:(.*)/i', $Str, &$matches);
#
aaronpk i think you can just get rid of the &
#
tantek I don't think so - because the code after assumes that "matches" has been "filled in" by that call.
#
tantek which only works by reference
#
aaronpk no it worsk without
#
aaronpk i've never used the & and it works fine
#
tantek huh - php.net still has the & http://php.net/manual/en/function.preg-match-all.php
#
tantek oh the function signature ASSUMES the & already
#
tantek interesting
#
tantek hmm - going to do the "safe" thing and hotpatch just this one error instead of deploying dev.h2vx.com to h2vx.com to "fix" it
[aaronpk] joined the channel
#
[aaronpk] The function signature has the &, so you don't specify it at call time anymore
#
[aaronpk] That's why the error mentions call-time pass by reference
#
tantek aaronpk++ yes removing the & did it
#
Loqi aaronpk has 981 karma
#
tantek h2vx.com updated
#
kevinmarks is h2v mf1 only?
#
tantek yes
#
tantek even dev.h2vx.com (also updated / fixed)
#
tantek dev.h2vx.com is what is running the current state of the github H2VX repo
#
tantek.com edited /LiteSpeed (+111) "emphasize shift-reload to test htaccess fix" (view diff)
benwerd, snarfed, yakker and [aaronpk] joined the channel
#
[aaronpk] Cannot wait till my site rebuild is done. So many things I want to add to it!
#
[aaronpk] I'm trying to not let myself add new stuff yet until I get it back to the state of things that work now
[shaners] joined the channel
#
[shaners] aaronpk: same for me
#
ben_thatmustbeme I was tempted to do the same
#
ben_thatmustbeme But it's not really the overall architecture that is the problem. I just need to rewrite a pile of controllers
#
kevinmarks i am resisting the temptation to actually write some code fro mine, but I should switch it to an actual static host
#
ben_thatmustbeme Anyway need sleep
#
ben_thatmustbeme Goodnight all
#
Loqi don't let the bed bugs bite
[shaners] joined the channel
#
[shaners] night
[aaronpk] joined the channel
#
[aaronpk] Yeah I am normally not a fan of doing a full rewrite, but in this case the overall architecture and assumptions were really preventing me from making more progress with it
M-kegan joined the channel
#
kevinmarks using oauth dropins, and it feels like 'get me an h-card' would be nice
#
kevinmarks currently coding that in my app
[snarfed] joined the channel
#
[snarfed] hmm! interesting idea. I'd happily accept a PR for that
#
kevinmarks hm, that means I have to get your app building too ;)
#
kevinmarks name, uid (ie profile url), photo, url[], seem doable for most of them
#
kevinmarks I'm going to try login with 2 choices with current code first.
#
kevinmarks the other part of this is having to replace the appengine users api with my own cookie based one
#
kevinmarks which also might make sense to move into the dropins lib
#
[snarfed] hmm! that, I'd probably need to hear more about
#
[snarfed] tomorrow!
#
kevinmarks yes, that's a bit trickier and less developed so far
e-lima, Jihaisse, cweiske, sanduhrs, sammachin, Tino, petermolnar and pfefferle joined the channel
#
pfefferle good morning
#
Loqi pfefferle: GWG left you a message on 10/14 at 10:33am: Happy birthday. http://indiewebcamp.com/irc/2015-10-14/line/1444844029239
#
pfefferle GWG thanks :)
petermolnar and loic_m joined the channel
#
kevinmarks hm, I can login to G+ from my locl install, but appenginge gives me a 500 error
#
kevinmarks ah, found it. had to turn off openid support as they broke that.
#
@litenmendryg RT @bisonblog: Har Facebook lÃ¥st in oss för gott? Skriver hos @didigital_se om #IndieWeb och kampen om ett fritt internet. http://t.co/Nfid… (twitter.com/_/status/656027783986974720)
yakker, cweiske2, mlncn and elima_ joined the channel
#
kevinmarks wow, facebook gives you nothing in the API these days
#
petermolnar define nothing, please
#
kevinmarks their user json is {"name":"Kevin Marks","id":"10156180149405425"}
#
kevinmarks Google and twitter give you urls, images, links, description way more
adactio and friedcell joined the channel
#
petermolnar viva la lockdown mode :(
behind_you, Erkan_Yilmaz and tantek joined the channel
#
@rbonini RT @petermolnar: @rbonini @kevinmarks ownyourgram is just one solution, there are others listed at http://indiewebcamp.com/instagram (twitter.com/_/status/656054044004327424)
joskar, glennjones, fkooman, elima_ and Hyle joined the channel
#
adactio.com edited /events/2015-10-21-homebrew-website-club (+3) "/* Where */" (view diff)
#
adactio.com edited /events/2015-10-21-homebrew-website-club (-46) "/* Brighton */" (view diff)
friedcell and pfefferle joined the channel
#
GWG pfefferle: Saw the (linkback)_post issue.
#
pfefferle hey GWG
#
GWG pfefferle: Hello. Haven't seen you around in a bit
#
pfefferle GWG yes :( a lot of work these days...
#
pfefferle GWG any idea how to fix that problem?
#
pfefferle GWG nice work with https://github.com/dshanske/semantic-notification btw!
#
GWG pfefferle: I'm still hoping to get the new filters into core on that
#
pfefferle I also had a look at the semantic-comments these days, to try to re-use some of the code for a ficepile...
#
pfefferle s/ficepile/facepile
#
Loqi pfefferle meant to say: I also had a look at the semantic-comments these days, to try to re-use some of the code for a facepile...
#
GWG pfefferle: I need to have a look at the code in detail to try to figure out the problem. I'd been thinking a lot about this because of an issue I tried to patch in core re pingbacks. Right now, Semantic Linkbacks retrieves the HTML from a site a second time, after the pingback or webmention code already has. I wanted to try to open up a way for the data to be processed earlier, and still be able to go into comment meta, which it can't now.
#
GWG pfefferle: Re the Facepile, that code is the first thing I wrote when I got involved here, it needs improvement. Look at the mf2_s comment code, which is an improved version of some of the same code.
#
pfefferle GWG this would be awesome! I also had a look at the code when I started working on the SL plugin. Sadly the do all the parsing stuff directly in the code and have noo hooks for that at all...
#
GWG pfefferle: I'm actually trying to get that in.
#
pfefferle GWG thanks for the mf2_s tip... will have a look...
#
@adactio Brightonians—Homebrew Website Club is on Wednesday, 6pm @68MiddleSt https://indiewebcamp.com/events/2015-10-21-homebrew-website-club (twitter.com/_/status/656084172780498944)
#
GWG pfefferle: https://core.trac.wordpress.org/ticket/34141
#
GWG pfefferle: I looked at the Semantic Linkbacks and the Webmentions code to get some ideas on this. I'll keep trying, I guess.
#
@benjaminparry RT @adactio: Brightonians—Homebrew Website Club is on Wednesday, 6pm @68MiddleSt https://indiewebcamp.com/events/2015-10-21-homebrew-website-club (twitter.com/_/status/656084867483049984)
#
GWG pfefferle: The only solution I can think of so far is to split up linkback_fix. Need to look at it more.
tvn joined the channel
#
pfefferle GWG yes, please let me know if you have an idea!
#
GWG pfefferle: How is having it on comment_post causing the problem? Haven't researched it yet.
#
pfefferle It does not fire on an update
Zegnat, eschnou and friedcell joined the channel
#
@simmelj RT @ThatEmil: Nice to see some media coverage of the #indieweb idea (swedish article)! https://twitter.com/bisonblog/status/655687208532217856 (twitter.com/_/status/656092747363975168)
#
GWG pfefferle: Wasn't that the issue about firing on edit_comment?
sanduhrs joined the channel
#
pfefferle GWG yes, but if we do that on every update, you will not be able to edit the comment by hand, because it will always be overwritten by the linkback plugin... and it we have to add preventions to start no infinite loop, because the SL plugin also fires an "edit_comment"...
Tino joined the channel
#
GWG pfefferle: Will play with it and get back to you.
snarfed, yakker, pfefferl_, hs0ucy and hmans joined the channel
#
ben_thatmustbeme hello all
indie-visitor joined the channel
#
Loqi Welcome, indie-visitor! Set your nickname by typing /nick yourname
j12t and ttepasse joined the channel
#
snarfed morning ben_thatmustbeme
#
snarfed on the email deliverability tar pit, and why trying to own your email is a mug's game: http://liminality.xyz/the-hostile-email-landscape/
#
ben_thatmustbeme snarfed, i think what i'm going to do is just try to make a list of facebook's alternate versions of profile URLs
#
ben_thatmustbeme and hopefully converge my list to only a few possible
#
ben_thatmustbeme since most of them use the same ID
#
snarfed ben_thatmustbeme: ah, your goal is de-duping profile urls? not posts?
#
snarfed that's a bit easier
#
ben_thatmustbeme snarfed: yeah
#
snarfed one catch is app-scoped user ids though
#
ben_thatmustbeme i'm just dealing with unifying my knowledge of people
#
ben_thatmustbeme yeah, thats why its not going to be able to do all the work
#
ben_thatmustbeme so how does it work with app scoped ids, i'm just using the URLs from bridgy
#
ben_thatmustbeme trying to get it so i have one entry for each person across silos (and multiple urls per silo)
#
ben_thatmustbeme so only one image
#
snarfed profile urls from bridgy?
#
snarfed oh right
#
snarfed bridgy's profile urls are sometimes global user ids, sometimes app-scoped, depending on when the uers signed up for bridgy
#
snarfed i think FB redirects all profile URLs to use username now
#
snarfed (like i wish they did with posts)
#
snarfed so maybe just HEAD and follow redirects?
#
snarfed are you using rel-me to connect people's silo accounts?
sanduhrs1 joined the channel
#
snarfed oh right, nm, FB only redirects facebook.com/[app_scoped_id] to /[username] if you're logged in
#
snarfed yeah, this is kind of the point of FB app-scoped user ids, you're not supposed to be able to deanonymize and identify people across apps
#
snarfed hmm, i don't currently put people's web site in bridgy's user page h-card, but i should. you could probably use that
#
snarfed https://kylewm.com/services/mf2?url=https://www.brid.gy/facebook/212038
tvn and [snarfed] joined the channel
#
[snarfed] ... but that would only help with responses from people on bridgy, which is usually a minority
#
[snarfed] ben_thatmustbeme: so, you need something like rel-me discovery, ie *something* in common across the silo profiles
#
[snarfed] ... and actually i don't even think bridgy can look up an FB user's profile and web site URL if they haven't authed bridgy
#
[snarfed] so you may have to use heuristics like name :(
#
ben_thatmustbeme snarfed, yes, i'm going to have to do many by hand, i'm creating a system for that
#
[snarfed] (this has been quite a monologue :P)
#
ben_thatmustbeme to make it just suggest merging contacts type of thing
tantek joined the channel
#
[snarfed] agreed. especially for FB users who aren't actually signed up on bridgy, i think name matching is your only hope :/
#
ben_thatmustbeme well bridgy backfeeds me a link to their profile that works
#
ben_thatmustbeme thats all i really need
#
ben_thatmustbeme or rather their name and image and a link tot heir profile
#
ben_thatmustbeme i might be able to do some cleaver things with the image
#
ben_thatmustbeme hmmm
Geng joined the channel
#
[snarfed] the catch is that profile link requires you to be logged in
#
[snarfed] ah, image, sure, for people who reuse it
#
[snarfed] each silo does their own image processing though, so you'd need to do fuzzy image matching...ugh
j12t joined the channel
#
ben_thatmustbeme OHHHH, no no snarfed
#
ben_thatmustbeme i mean within FB
#
ben_thatmustbeme i have two profiles, one with their app specific id, one with their real user id
#
ben_thatmustbeme for their images
#
@Lottejackson RT @adactio: Brightonians—Homebrew Website Club is on Wednesday, 6pm @68MiddleSt https://indiewebcamp.com/events/2015-10-21-homebrew-website-club (twitter.com/_/status/656124587936227328)
#
ben_thatmustbeme just downloaded them from wget (to ensure i'm not logged in)
#
ben_thatmustbeme and md5sum is the same
#
[snarfed] oh within fb! sure
#
ben_thatmustbeme sweet!
#
ben_thatmustbeme \o/
#
ben_thatmustbeme thats the majority of the work
#
[snarfed] that's actually an interesting hack to deanonymize app scoped FB user ids in general, if it works
#
ben_thatmustbeme between silos i really don't have that many people, so its not a huge thing
#
[snarfed] worth publishing
#
ben_thatmustbeme well, you'd need to already have the user's real user
#
ben_thatmustbeme which doesn't help to generically go from anon -> real
#
[snarfed] well it lets you join two arbitrarily sized FB user sets across different FB apps
#
[snarfed] which is exactly what they meaned to prevent
#
[snarfed] s/meaned/meant/
#
Loqi [snarfed] meant to say: which is exactly what they meant to prevent
#
[snarfed] they could prevent that by watermarking images by app
LanceyWork joined the channel
#
aaronpk oh wow
#
aaronpk it wouldn't take much actually, i'm sure there's some sort of invisible data they could change in the image to change the md5 sum
#
[snarfed] it'd need to be lightly visible, otherwise you must render and hash the pixels
#
[snarfed] s/must/just/
#
Loqi [snarfed] meant to say: it'd need to be lightly visible, otherwise you just render and hash the pixels
#
aaronpk yeah, but even something as simple as adding 1 to the red value of a random pixel would be totally undetectable by eye but would change the hash
yakker joined the channel
#
[snarfed] right!
#
[snarfed] i guess you could then downsample, convert to b&w, or even just edges... i wonder how low res you could go before you hit collisions
#
[snarfed] eh. that's a project for a Friday, not a Monday :P
#
aaronpk eh, if it takes that much work to match photos it's probably acceptable
#
[snarfed] i dunno. it's still linear in the number of users
#
[snarfed] meh
#
aaronpk the fact that you can match profiles with the md5 sum of the image right now is worth writing up
#
ben_thatmustbeme [snarfed]: aaronpk yeah, any single bit change would throught the md5sum
#
ben_thatmustbeme but there are plenty of image matching algorithms out there
#
[snarfed] writing up, definitely. on my to-do list now, unless ben_thatmustbeme beats me to it
#
ben_thatmustbeme I'm not a good writer
shiflett joined the channel
#
ben_thatmustbeme just cite me
#
ben_thatmustbeme or mention me
#
ben_thatmustbeme or something
#
[snarfed] sure!
#
ben_thatmustbeme but yes, i have done image manipulation stuff before
#
ben_thatmustbeme if they are based on the same image, and the same size, water mark all they want, its still fairly easy to get a match
#
ben_thatmustbeme with some bit of certainty
#
aaronpk yeah that's why I think a simple fix like changing a pixel would be worth it, since it would throw off the trivial md5 match
ttepasse joined the channel
#
[snarfed] yeah, the hash is the key case to foil. comparing every pair is quadratic runtime, ie intractable for large sets
#
ben_thatmustbeme its still time N
#
ben_thatmustbeme based on N bits in the image
#
ben_thatmustbeme assuming no shift
#
ben_thatmustbeme shifts certainly add more, complexity, but even then there are low order magnitude solutions
yakker, tvn_ and j12t joined the channel
#
[snarfed] eh images will be bounded in size so that's roughly constant. i was thinking N users :P
#
@nobantu Join us 4 ALLthings #InternetIdentity @ #IIW #unConference: #VRM #CRM #indieweb #blockchain #OAuth #bitcoin #privacy https://twitter.com/idworkshop/status/655414283371573248 (twitter.com/_/status/656137594372100096)
#
[snarfed] ie what's the complexity of joining a set of n users with a set of m users
#
[snarfed] that's what fb cares about
#
ben_thatmustbeme ah, yes,
#
ben_thatmustbeme with md5sums you can just store the sums of them all
#
ben_thatmustbeme and just lookup
#
ben_thatmustbeme indeed
#
[snarfed] right
#
[snarfed] as opposed to comparing every pair
#
ben_thatmustbeme wonder if you could create a fuzzy image checksum
#
ben_thatmustbeme so not based on bits but approximate color values at positions
#
ben_thatmustbeme not a checksum at that point
#
aaronpk yeah you could resample the image down to like 20x20 with a small color space and that might do it
#
ben_thatmustbeme still comparing each pair but bringing that math down significantly
shiflett joined the channel
#
[snarfed] nah, you can still hash the downsampled, b&w, maybe edge-detected version
#
[snarfed] like i mentioned earlier :P
#
[snarfed] key question is how downsampled you can go before you start seeing collisions
#
[snarfed] eh, we can see if we get a response from fb
#
[snarfed] thanks for the idea ben_thatmustbeme and discussion aaronpk, this will be fun to write up!
#
aaronpk \o/
cleverdevil and awolf joined the channel
#
[snarfed] way more fun than fighting fb's id/url proliferation
tallpaul, KitB, anm, catsup, tantek and Zegnat joined the channel
#
tantek good morning #indiewebcamp!
scoates joined the channel
#
tantek great to see that HWC is on in Brighton this week!
#
tantek adactio++
#
Loqi adactio has 20 karma
catsup and snarfed joined the channel
#
tantek.com edited /MediaWiki:Sidebar (+2) "Brighton, Göteborg, Portland, SF" (view diff)
#
@jimgaynor .@moniguzman We talked about this earlier. POSSE: Publish (on your) Own Site, Syndicate Elsewhere. https://indiewebcamp.com/POSSE (twitter.com/_/status/656148193957154817)
catsup, j12t and [kevinmarks] joined the channel
#
[kevinmarks] You can change an image file without touching pixels, but the image processing will help. Down sampling might get you false positives from rainbow avatars etc.
catsup and snarfed joined the channel
#
snarfed fleshed out the h-cards on bridgy user pages, they now have full u-url and rel-me. example: https://kylewm.com/services/mf2?url=https://www.brid.gy/facebook/212038
benwerd, Geng and catsup joined the channel
#
tantek more on that candidate slogan - here's a real-world use in context to give you an idea:
#
tantek Be a part of the open web, with your web site.
#
tantek here: https://wiki.mozilla.org/WeeklyUpdates/2015-10-19#Wednesday.2C_21_October
#
tantek fellow HWC organizers (aaronpk, kylewm, benwerd, adactio, tbrb, voxpelli) how does that read to you?
#
tantek the "open web" aspect in particular is tailored for that audience (Mozilla) - which is expected to want to help keep the web open etc.
#
tantek beyond just being "independent" - which is the usual appeal we make
catsup joined the channel
#
voxpelli tantek: would replace "with" with something more active/actionable – perhaps "get"?
#
voxpelli Even "Be part of the open web, have a web site"
#
tantek "have" sounds more passive, like collecting something you buy
catsup joined the channel
#
aaronpk i think i like it, but the comma seems odd
#
tantek "with", by being absent of action, puts the focus of action back on the "Be" at the start of the sentence, thus overall making it more active
#
tantek aaronpk: yes, the comma seems odd to me too
#
voxpelli Might be the comma that makes me react
snarfed and catsup joined the channel
#
kevinmarks for a slogan, you want the stress to be on the end, and that sentence is oddly front weighted
#
kevinmarks "The open web is waiting for your web site" moves the flow to you more
#
tantek it's an iteration on "Be independent with your web browser and your web site."
gRegorLove joined the channel
#
tantek previous iteration: https://wiki.mozilla.org/WeeklyUpdates/2015-10-05#Wednesday.2C_07_October (and earlier)
#
tantek kevinmarks: "waiting for your" makes it sound like the open is not there yet, it also has the same issue of "have and collect" that I pointed out above
#
tantek the point of HWC is not to just check some checkmark
#
kevinmarks it being an exhortation definitely makes it odd to have a comma,
#
tantek e.g. "ok I added my website to the open web, I'm done"
#
kevinmarks "You're invited to the open web, bring your own website"
benwerd, catsup and j12t joined the channel
#
snarfed ooh yes. i actually like just "bring your own website"
#
snarfed clearly implies that you're using it to participate in a larger community
#
snarfed and short and sweet
#
ben_thatmustbeme Bring Your Own Website++
#
ben_thatmustbeme BringYourOwnWebsite++
#
Loqi BringYourOwnWebsite has 1 karma
catsup joined the channel
#
tantek snarfed, do you mean "Be a part of the open web, bring your own website" ?
#
tantek or just drop the open web part? (which was the whole point to be a hook for Mozilla audiences)
#
kevinmarks come to the open web party, bring your own website
#
kevinmarks was what I was trying to imply
#
ben_thatmustbeme This party is BYOW
#
ben_thatmustbeme hehe
#
snarfed i really do like just "bring your own web site"
#
bear Be the open web, bring your own website
#
snarfed i think the rest is clearly enough implied, and i think short and sweet, no comma is very valuable
tvn joined the channel
#
bear Be the open web, we can show you how
#
bear Be the open web, avoid silos
#
tantek snarfed: unfortunately the rest of is not clearly enough implied
#
tantek most people at Mozilla think "be the open web" means just by choice of browser
tvn joined the channel
#
tantek the key here is to connect "open web" and "your own website" - because most people still don't internalize that connection (and happily just use silos instead, despite claiming they "support the open web")
#
tantek bear - "avoid silos" is one of those "do nothing" to succeed answers - which is not enough.
#
bear I was just riff'ing on the Be the ..., ... pattern
#
tantek another iteration
#
tantek Join the open web with your own website.
catsup joined the channel
#
tantek which has the same problem I pointed out before, just trying to drop the comma
#
voxpelli "Your website makes the web open"
#
voxpelli I like the "bring" suggestions
#
bear Own your place in the open web
#
voxpelli "Your site is the open web"
#
kevinmarks some version of "a journey of a hundred miles starts with a single step"
#
tantek I liked the cadence of the Squarespace ad's conclusion
#
gRegorLove Be the website you want to see in the open web
#
tantek and that has a comma
#
tantek A better web, starts with your website.
#
gRegorLove a la "be the change you want to see in the world" :)
#
tantek good enough for a superbowl ad is probably good enough :P
#
kevinmarks that really doesn't need a comma, tantek
#
bear An open web starts with you
#
tantek interesting
#
bear The open web starts with you; find out how
#
tantek alright then
#
tantek Be a part of the open web with your own web site.
#
gRegorLove +1 for just "bring your own website"
#
voxpelli tantek: perhaps just "site" rather than "website" as that implies?
#
tantek not dropping "open web" because that's the essential hook for this audience
#
bear site is jargon IMO - most folks don't use that phrase for websites
#
voxpelli s/that implies/that's implied/
#
Loqi voxpelli meant to say: tantek: perhaps just "site" rather than "website" as that's implied?
#
tantek but "Bring your own website." is great for indieweb in general
#
kevinmarks it's also like "changing the world, one person at a time"
#
tantek bear, I agree that "site" by itself is jargon
#
tantek what is a site?
#
Loqi Web hosting can be the primary regular cost in maintaining an IndieWeb site; this page lists several options from free on up depending on your publishing needs, like a static, shared, private, or dedicated server https://indiewebcamp.com/site
#
kevinmarks opening the web, one website at a tiome
#
tantek lol
#
bear a lot of what we have said above are all great campaign type posters once a core phrase is locked in
#
tantek ok, sticking with "Be a part of the open web with your own website. " for today's meeting
#
tantek we'll see how it's read out loud
#
bear Be the open web
#
tantek sounds like too much responsibilty ;)
catsup joined the channel
#
voxpelli The slogan should probably emphasize that it's the independent sites that makes the open web
#
kevinmarks the people who are crazy enough to think they can open the web are the ones who do
#
tantek bear, speaking of, do you have an h-feed on your site?
#
voxpelli So rather than "Be part of the open web", maybe "Make the open web"?
#
bear it's a call to arms so implying responsibility will bring out the change makers
#
tantek in fact, all of you
#
tantek bear, voxpelli, kevinmarks do you have an h-feed on your site? add it: https://indiewebcamp.com/h-feed#IndieWeb_Examples
#
bear tantek - not yet, my site is adverse to a feed right now just because of it's structure
#
tantek bear, what's on your home page?
#
bear just contact page
#
tantek the rest of you have a homework assignment for Wednesday then, get an h-feed on your site (if you don't already) and add yourself to https://indiewebcamp.com/h-feed#IndieWeb_Examples
#
bear oh hey, I do have one - duh
#
voxpelli I have one, wonder since when
#
tantek aaronpk, snarfed, also surprised to see you're not in that list https://indiewebcamp.com/h-feed#IndieWeb_Examples
#
tantek and certainly ben_thatmustbeme !!
#
bear ah, when I went to new static
#
kevinmarks https://github.com/ampproject/amphtml/issues/657#issuecomment-149151284
#
aaronpk i don't have an h-feed!
eschnou joined the channel
#
GWG tantek, do I get college credit for this class?
#
ben_thatmustbeme aaronpk: you don't?
#
aaronpk no just a list of h-entrys
cleverdevil- joined the channel
#
kevinmarks aha, I can use blame to find when I added it
#
bear.im edited /h-feed (+169) "/* IndieWeb Examples */" (view diff)
#
voxpelli I wonder what order that list is in :P
#
voxpelli kevinmarks: just did that ;)
catsup joined the channel
#
bear yea, mine may be earlier but that is july of 2014 is the earliest I can see right now
#
ben.thatmustbe.me edited /h-feed (+248) "/* IndieWeb Examples */ add me" (view diff)
#
kevinmarks.com edited /h-feed (+216) "/* IndieWeb Examples */" (view diff)
#
ben_thatmustbeme i need tantek, i'm sure there are a lot of pages i don't have me listed on
#
ben_thatmustbeme under examples
#
tantek voxpelli: approximate date order
#
tantek like all IndieWeb Examples sections
#
kodfabrik.se edited /h-feed (+247) "/* IndieWeb Examples */" (view diff)
#
ben_thatmustbeme problem is i have no clue of date
#
voxpelli tantek: that order probably went out of the window about now :P
#
ben_thatmustbeme when i added it
#
tantek ben_thatmustbeme: make it a habit, as soon as you support a page of something on the wiki, add yourself to the examples!
#
tantek voxpelli, it's a wiki - fix it!
#
bear i'm fixing the order now
#
voxpelli will let everything cool down first
#
tantek ben_thatmustbeme: do you know the year?
#
voxpelli bear++
#
kevinmarks ben_thatmustbeme: https://waterpigs.co.uk/ is not in archive.org
#
Loqi bear has 77 karma
#
tantek then add YYYY-??-??
#
rhiaro.co.uk edited /h-feed (+304) "rhiaro example" (view diff)
#
ben_thatmustbeme kevinmarks: huh?
#
kevinmarks oh wait, http is not https
#
rhiaro heh I hit edit conflict with voxpelli, kevinmarks and ben_thatmustbeme
#
ben_thatmustbeme but i'm not waterpigs.co.uk
#
ben_thatmustbeme lol
#
voxpelli I wonder if I should make my "firehose" archive feed discoverable from the front page somehow... If someone wants to subscribe to _all_ of my content
#
ben_thatmustbeme yeah, i hit one with someone
#
bear.im edited /h-feed (+0) "date ordering" (view diff)
catsup joined the channel
#
kevinmarks derp
#
kevinmarks is binary searching archive.org to find a start date for waterpigs
#
rhiaro voxpelli: I'm thinking about that too, the 'firehose' type feed
#
ben_thatmustbeme tantek: i've had an h-feed on my site since i created it... i based it on a wordpress theme that had it already
#
rhiaro don't quite have a complete one yet though
#
ben_thatmustbeme so march 2014 i guess
#
tantek ben_thatmustbeme: then use the date that *you* implemented h-feed
j12t joined the channel
#
tantek and word it accordingly
#
voxpelli I have one, but I don't really intend any person to actually visit my firehose feed – it's just my last resort for archiving
#
tantek it's still useful for testing consuming code etc.
#
kevinmarks.com edited /h-feed (+0) "/* Barnaby Walters */" (view diff)
#
bear.im edited /h-feed (+0) "date ordering" (view diff)
#
voxpelli even my firehose feed doesn't print out articles in full either :/
#
kevinmarks my kevinmarks.com heed is titles only
#
kevinmarks s/heed/h-feed/
#
Loqi kevinmarks meant to say: my kevinmarks.com h-feed is titles only
#
voxpelli same on all of my blog post feeds
#
kevinmarks though the name/summary/featured pattern is interesting
#
bear yea, titles only is what i'm thinking my front page will have - just below the ego-self stuff :)
catsup joined the channel
#
kodfabrik.se edited /h-feed (+103) "/* Pelle Wessman */ Point out that my feeds have partial content" (view diff)
cleverdevil joined the channel
#
ben.thatmustbe.me edited /h-feed (+174) "/* Ben Roberts */" (view diff)
#
ben_thatmustbeme yay for git history
#
Loqi yay!
#
voxpelli would really like to experiment with feed discovery in indie reader and the subscription to just a part of a users posts
#
rhiaro yes, what voxpelli said
#
tantek bear, voxpelli interesting about different preferences for home page presentation
#
tantek perhaps something we could add to /homepage
Geng joined the channel
#
voxpelli feels like that's an experiment that was part of early day microblogs like Jaiku, but which has died of since
#
kevinmarks woodwind offers a choice of feeds
#
kevinmarks if it find more than 1
#
rhiaro find with rel=feed?
#
voxpelli yeah, I wonder how it finds the feeds
#
tantek rhiaro: huh? what's the use-case you're trying to solve?
#
voxpelli tantek: there's a section in the h-feed about partial content already
#
kevinmarks often that is a spurious 'atom vs rss' choice though
#
tantek but this is about "what you put on your home page"
#
kevinmarks rel="alternate" too I think
#
rhiaro doesn't alternate suggest a different versionof the same feed? eg. rss vs html
#
rhiaro but same content
#
voxpelli tantek: that's complicated for me as my "home page" is a blog and the social interactions and bookmarks features are pushed aside into sub pages, hmm
#
tantek.com edited /homepage (+119) "/* Stream of Updates */ add sub lists for partial posts and names/titles only" (view diff)
#
rhiaro tantek: I'd like to be able to be able to put a homepage url into a reader, and have the option to subscribe to just notes, or just articles, or a certain tag, etc, if they're provided
#
ben_thatmustbeme i'm thinking i'm just going to make any links to actual feeds be enclosed in an h-feed where u-url is the only item
#
ben_thatmustbeme well and a description
#
voxpelli kevinmarks: rel-alternate would be wrong in my case as the other feeds are not alternate presentations of the same content as is the case with rss/atom
#
rhiaro or any other collection of posts according to whatever criteria
#
tantek rhiaro: perhaps add that to /reader#Brainstorming to start with
#
ben_thatmustbeme the way i would want it to work is just get the list of all feeds on someone's site and I can pick which one i want
#
rhiaro eg. my /travel and /calendar feeds might be useful to subscribe to, and have some arbitrary rules to decide what ends up there
catsup joined the channel
#
ben_thatmustbeme which means i should put descriptions on them
#
tantek voxpelli, kevinmarks, add yourselves to https://indiewebcamp.com/homepage#Stream_of_Updates depending on what your home page has
#
rhiaro ben_thatmustbeme: yeah
#
bear.im edited /homepage (+138) "add example for homepage" (view diff)
#
rhiaro you could do rel="feed" title="Whatever"
#
ben_thatmustbeme basically use partial feeds with nothing in them as the way to find feeds on a site
#
voxpelli here's two old images from jaiku: http://4.bp.blogspot.com/_ZaGO7GjCqAI/SW8XexqbClI/AAAAAAAAOws/HXhYemRyR2Y/s640/jaiku.png
#
voxpelli and http://hannes.emotisys.de/wp-content/uploads/sites/1/2009/05/bild-3.png
#
voxpelli note the "Feeds" box
#
bear.im edited /homepage (+21) "add link to streams example entry" (view diff)
#
voxpelli so one could select if one wanted Jyri's delicious bookmarks or not – because unlike on Twitter you had the choice to include or not include the feeds they imported into Jaiku
#
ben_thatmustbeme actually i already have that, if you look at the h-feed on my main page it does have u-url for /feed
#
kevinmarks https://github.com/kylewm/woodwind/blob/5e334b8a056b3bb769f53d0a7b3aa83494362fd8/woodwind/views.py#L410
#
ben_thatmustbeme i never got the next/previous buttons on the UI, but i have them almost ready
#
tantek rhiaro: oh hey looks like we already have this documented: https://indiewebcamp.com/h-feed#canonical_feed_autodiscovery
#
tantek no need to make up "could do" :P
KartikPrabhu joined the channel
#
kevinmarks looks liek woodwind checks for rel="feed" for html and rel="alternate" for xml types only
#
tantek what is rel-feed?
#
Loqi It looks like we don't have a page for "rel-feed" yet. Would you like to create it? https://indiewebcamp.com/s/1055
#
tantek ^^^ kevinmarks, what to start that with a short definition and mention of Woodwind's implementation?
#
tantek s/what to/want to
#
Loqi tantek meant to say: ^^^ kevinmarks, want to start that with a short definition and mention of Woodwind's implementation?
#
kevinmarks rhiaro_: the convention is to use the title to distinguish the kind of feed in rel="alternate"
catsup joined the channel
#
voxpelli is adding such discovery to his site now
#
voxpelli Hmm, looks like parsers have started to dislike "<link rel="canonical" href="/" class="u-url" />" :( They just include an empty string for the "url" value
#
voxpelli Both the PHP and JS ones
#
kevinmarks rel-feed is a convention for linking to multiple alternative [[h-feed]]s from a site's homepage see [https://indiewebcamp.com/h-feed#rel_feed h-feed]
#
loqi.me created /rel-feed (+176) "prompted by tantek https://indiewebcamp.com/irc/2015-10-19/line/1445278876164 and dfn added by kevinmarks" (view diff)
#
Loqi ok
#
kevinmarks voxpelli: are you sure you're not missing a base there?
#
kevinmarks which page?
#
voxpelli kevinmarks: relative URL:s are valid there, aren't they? Although I could of course make it absolute, I should have all the data available
#
voxpelli kevinmarks: http://voxpelli.com/
#
kevinmarks realtive are valid, but if you don't give the parser a base url it can't resolve them
#
voxpelli kevinmarks: I give them the URL to the page, they should use that to resolve?
#
bear is there a place I can check the newly added rel="feed" autodiscovery link for my homepage?
#
kevinmarks woodwind?
#
voxpelli bear: http://reader.kylewm.com/ ?
#
voxpelli (that's Woodwind I think)
#
bear ah - reader - I keep forgetting that one
catsup joined the channel
#
kevinmarks woodwind subscribe, btu thta may be it finding the markup on the page
#
gRegorLove snarfed: Not seeing the profile + logout link on my bridgy page now. Odd.
#
aaronpk there's some brainstorming on subscribing to partial feeds already
#
snarfed gRegorLove: huh. same device/browser?
#
gRegorLove Yep
#
gRegorLove I'll re-auth
#
snarfed wait
#
gRegorLove Oops.
#
aaronpk hmm where'd it go
#
snarfed :(
#
snarfed ah well
#
gRegorLove Came back after re-auth
#
bear oof - seems that woodwind did auto-discover the right url but the name of the feed is all wrong
#
gRegorLove I'll let you know if it disappears again :)
#
snarfed good reminder though. PSA for bridgy users: if you re-login to one or more of your silos, bridgy will remember you and show a header links to your user pages
#
snarfed so you don't have to trawl through browser history or the horrible http://brid.gy/users
eschnou joined the channel
#
ben_thatmustbeme my browser just autocompleted all the time to those accounts
#
ben_thatmustbeme so i never really had an issue with this
[snarfed] joined the channel
#
[snarfed] browserhistory++
#
Loqi browserhistory has 1 karma
catsup joined the channel
#
kevinmarks brid.gy/users isn't bad when you realise you can hack the url
#
kevinmarks maybe add a 'find user by name' form that just does that for you
#
[snarfed] heh true kevinmarks
#
[snarfed] i do that myself sometimes
#
[snarfed] yeah, I've just avoided it because I'm not great at UX
#
voxpelli waits for GitHub Pages to rebuild his page – really needs to get a more premium static page host eventually
#
kevinmarks it is case sensitive, which is a bit tricky
#
rhiaro Yay finally figured out how / was permitted to book MIT rooms and booked for IWC MIT
#
aaronpk \o/
#
rhiaro Step 2: go find out where these rooms I've booked actually are
#
rhiaro this building is big and confusing
#
kevinmarks is it the Gehry one that looks like it is falling down?
#
rhiaro the Stata Centre
#
rhiaro.co.uk edited /2015/MIT (+20) "/* Venue */" (view diff)
catsup joined the channel
#
kodfabrik.se edited /homepage (+170) "/* Stream of Updates */" (view diff)
#
voxpelli aaronpk: you have a "separate feeds" example on that page – I added myself there as well, but realize adactio also have such feeds (Links) which is what I mimicked
#
voxpelli wondering if that should perhaps be extracted into a separate list or at least have adactio added to it as well
#
voxpelli kevinmarks: made the rel-canonical absolute, but still no success – weird that the parses finds the u-url's existence, but fails to add a value to it :/
catsup joined the channel
#
kevinmarks so the inside of the Stata Centre is as confusing as the outside?
#
rhiaro yup
#
kevinmarks unmung finds them http://www.unmung.com/mf2?url=http%3A%2F%2Fvoxpelli.com&html=&pretty=on
#
ben_thatmustbeme oh yeah
#
voxpelli kevinmarks: nope, check the url-propery of the hfee
#
ben_thatmustbeme there were parts of the first floor that i know i happened across once and couldn't find again
#
voxpelli s/hfee/h-feed/
#
Loqi voxpelli meant to say: kevinmarks: nope, check the url-propery of the h-feed
#
ben_thatmustbeme despite it looking like its just a straight hallway
#
kevinmarks huh. Uk universites are often like that (eg KCL) because they are multiple buildings attached togteher with different floor plans; desiging one from scratch like that is a bit daft
#
kevinmarks oh, I was looking at the rels
#
voxpelli I guess the adding of classes to link-elements within head is to be for problems, but very convenient when you want the canonical URL for an h-feed to actually mark up the rel-canonical as that url
#
kevinmarks mark up <header> <h1><a href="/">Pelle Wessman</a></h1>
#
voxpelli thing is that the header always points to the frontpage – I don't have an element that points to the actual feed on the feed page
#
kevinmarks I think mf2 deliberately ignores link
#
voxpelli wonder why it still adds an empty "url" property though?
#
kevinmarks http://microformats.org/wiki/microformats2-parsing#parsing_a_u-_property doesn't mention link
#
voxpelli oh, never realized the mf2-spec is so strict – thought it always looked at "href", then "src" etc no matter the tag
#
kevinmarks I suspect it is falling through to the last one there
#
voxpelli does that mean mf2 is incompatible with custom elements?
#
kevinmarks ie looking for the textcontent of link which is empty
#
kevinmarks no, custom elements will work for most things, but u- is strict
#
kevinmarks write it up at http://microformats.org/wiki/microformats2-parsing-issues
#
kevinmarks you have a concrete example and were confused by the parsing so it is worth documenting there
cleverdevil- joined the channel
#
voxpelli that will have to be for another day :P
#
kevinmarks just copy and paste from this chat
hs0ucy and Geng joined the channel
#
kevinmarks hm, I kinda want a 'everything except this tag' mode on woodwind
#
kevinmarks mainly to hide the fb feed tbh
_fran and yakker joined the channel
#
@benwerd Is your domain your online identity? You have just 7 days to literally get the T-shirt. https://teespring.com/my-url-is #indieweb #identity #iiw (twitter.com/_/status/656192428940984321)
#
aaronpk benwerd++ for literally using literally correctly
#
Loqi benwerd has 96 karma
#
ben_thatmustbeme heh
eschnou joined the channel
#
loqi.me created /would_be_nice (+259) "prompted by tantek and dfn added by tantek" (view diff)
#
@Comy1982 RT @benwerd: Is your domain your online identity? You have just 7 days to literally get the T-shirt. https://teespring.com/my-url-is #indieweb #ide… (twitter.com/_/status/656198756069744641)
#
tantek.com edited /would_be_nice (+329) "expand dfn, see also, antipattern, what to do instead" (view diff)
[jonnybarnes] joined the channel
#
[jonnybarnes] everyone appears to be a bot :simple_smile:
#
aaronpk at least everyone doesn't look like the same bot the way most IRC/Slack gateways work ;)
#
tantek hey jonnybarnes, do you support an h-feed stream on your home page?
nitot joined the channel
#
tantek what is rel=feed?
#
Loqi https://indiewebcamp.com/rel-feed
#
tantek heh
#
[jonnybarnes] erm, well, my home page doesnt have any kind of feed
#
[jonnybarnes] I think /notes has an h-feed
#
tantek ah - could you add that to https://indiewebcamp.com/h-feed#IndieWeb_Examples ?
#
loqi.me created /502 (+76) "prompted by tantek and dfn added by bear" (view diff)
#
jonnybarnes.uk edited /h-feed (+108) "/* IndieWeb Examples */" (view diff)
#
tantek.com edited /rel-feed (+109) "fix dfn, see also" (view diff)
#
@BackwardNC .@Jeter4NC92 Nope. Your misleading graph had 2010 & 2011 as GOP years. You can't have it both ways. #OwnYourData (twitter.com/_/status/656201696859836416)
#
[jonnybarnes] ok, i have a user page on the wiki but it appears people are now using templates
#
aaronparecki.com edited /post-type-discovery (+311) "/* Issues */ post types" (view diff)
#
[jonnybarnes] is there a handy guide on setting one up?
#
aaronpk i don't know why i made that comment "post types"
#
aaronpk the whole page is about post types
#
tantek rhiaro: you have rel=feed right? perhaps add yourself in a "IndieWeb Examples" section on /rel-feed ?
#
rhiaro tantek: not yet, plan to
#
tantek who else uses rel=feed on their home page?
#
KartikPrabhu raises hand
fkooman and eschnou joined the channel
#
cleverdevil also has rel=fee
#
cleverdevil *feed
#
[jonnybarnes] so http:// wiki links ask me to login (at least in the top menu), https:// wiki links I’m already logged in
snarfed joined the channel
#
aaronpk yes that's a weird thing
#
[jonnybarnes] I’m assuming if you log in on an https link the secure flag is set on the cookie
#
KartikPrabhu hmmm so rel-feed is not in the http://microformats.org/wiki/index.php?title=existing-rel-values list but I am sure I saw somewhere that rel-feed was recommended to point to feeds
#
aaronpk i want to redirect the whole wiki to https but i can't until we a) stop hotlinking avatars or b) I figure out mediawiki magic to be able to rewrite the img tag for those
#
aaronpk yes that is correct
#
aaronpk i could un-set the secure setting but that seems not ideal
#
tantek.com edited /rel-feed (+562) "why, how, indieweb examples" (view diff)
#
[jonnybarnes] nah its fine
#
tantek KartikPrabhu: could you update https://indiewebcamp.com/rel-feed#Kartik_Prabhu with details?
#
bear.im edited /404 (+112) (view diff)
#
tantek KartikPrabhu: huh? I see "feed" mentioned several times on the rel registry
#
[jonnybarnes] so if I use my homepage as an about page kind of, and I have an h-feed on `/notes`, could I put a rel=feed link to notes from my homepage?
#
bear.im edited /404 (-27) (view diff)
#
tantek jonnybarnes: see /rel-feed - I think it now answers your question :)
#
[jonnybarnes] particularly I already have a link to `/notes` in my top menu
#
[jonnybarnes] could I just whack rel=feed onto that?
#
kevinmarks could you make an image proxy to httpsify?
#
aaronpk i have the image proxy done
#
aaronpk it archives them too so images stick around once the original disappears too
#
aaronpk the problem is mediawiki now
#
KartikPrabhu tantek: oh sorry it is in the POSH usage list not the main one
#
snarfed ah, so that's why img rewriting would actually work
#
bear.im edited /404 (-7) (view diff)
#
aaronpk i even have some of the image rewriting working
#
snarfed hmm. could you just do a one time migration that rewrites all avatars to https on the image proxy?
#
snarfed ah, ok
#
kevinmarks I'm thinking about that with my card stuff - building a proxy on the appengine image store could be handy
#
bear.im edited /404 (+1) (view diff)
#
tantek KartikPrabhu: go ahead and add it to the rel registry then http://microformats.org/wiki/existing-rel-values#HTML5_link_type_extensions, allowing for Link or A, and linking to https://indiewebcamp.com/rel-feed as the spec
#
kevinmarks as their resizing and cahcing is very good
#
aaronpk the hook i'm using doesn't get fired with the way we do images in user templates
#
aaronpk since we have the <img> tag there directly, instead of using mediawiki's image feature
#
kartikprabhu.com edited /rel-feed (+9) "/* Kartik Prabhu */ rough date" (view diff)
#
bear.im edited /502 (+132) (view diff)
#
aaronparecki.com created /test (+27) "test image hotlinking" (view diff)
#
bear.im created /401 (+209) "Created page with "{{{stub}}} '''<dfn>401 is an HTTP/1.1 Status Code</dfn>''' returned from a webserver to signify that your are not authorized to make a request to the URI. == See Also == - http..."" (view diff)
#
aaronpk yeah check it out, I left it running oops https://indiewebcamp.com/test
#
loqi.me created /449 (+99) "prompted by ben_thatmustbeme and dfn added by ben_thatmustbeme" (view diff)
#
aaronpk i forgot this version of it is just a straight proxy, no storage
#
ben.thatmustbe.me edited /449 (+46) (view diff)
eschnou and nitot joined the channel
#
loqi.me created /webapp2 (+134) "prompted by tantek and dfn added by bear" (view diff)
eschnou and Geng joined the channel
#
loqi.me created /pip_install (+117) "prompted by tantek and dfn added by bear" (view diff)
#
tantek KartikPrabhu: just a friendly reminder (since you raised your hand as actually using rel=feed on your site) to https://indiewebcamp.com/irc/2015-10-19/line/1445286240691 - thanks much in advance!
#
loqi.me created /wsgiref (+104) "prompted by tantek and dfn added by bear" (view diff)
#
tantek.com edited /email (+553) "/* Criticism */ Ecosystem discriminates against indie servers" (view diff)
snarfed and Geng joined the channel
#
@roomthily what does an #indieweb look like for data publication. we have feels on metadata as public goods, cultures to support that. (twitter.com/_/status/656219790168559616)
rhiaro joined the channel
#
tantek ben_thatmustbeme: re: adding yourself to Examples sections on the wiki, could you take a look at the "types" of posts here: https://indiewebcamp.com/posts#Kinds_of_Posts - click on each on and if you support it on your site, add yourself to its "IndieWeb Examples" section?
nitot joined the channel
#
@FemometerInc Don’t get left behind. http://www.femometer.com/ #OwnYourData #Femometer https://twitter.com/FemometerInc/status/656225588923858945/photo/1 (twitter.com/_/status/656225588923858945)
#
tantek ben_thatmustbeme and if by doing so, you increase the number of examples to 3+, could you request in https://indiewebcamp.com/post-type-discovery#Feedback to add such "types" to that spec?
#
tantek hmm maybe I should make that a tell
#
tantek !tell ben_thatmustbeme re: adding yourself to Examples sections on the wiki, could you take a look at the "types" of posts here: https://indiewebcamp.com/posts#Kinds_of_Posts - click on each on and if you support it on your site, add yourself to its "IndieWeb Examples" section?
#
Loqi Ok, I'll tell them that when I see them next
#
tantek !tell ben_thatmustbeme and if by doing so, you increase the number of indieweb examples in any of those pages to 3+, could you request in https://indiewebcamp.com/post-type-discovery#Feedback to add such "types" to that spec?
#
Loqi Ok, I'll tell them that when I see them next
#
@nobantu RT @benwerd: Is your domain your online identity? You have just 7 days to literally get the T-shirt. https://teespring.com/my-url-is #indieweb #ide… (twitter.com/_/status/656228515050721280)
snarfed joined the channel
#
tantek.com edited /Federated_Social_Web_Summit (+205) "see also osfw3c, 2013 had no FSWS but did have osfw3c" (view diff)
#
loqi.me created /man_behind_the_curtain (+139) "prompted by tantek and dfn added by tantek" (view diff)
#
tantek.com edited /man_behind_the_curtain (+43) "The, see also" (view diff)
#
tantek.com moved /man_behind_the_curtain to /the_man_behind_the_curtain "the"
#
tantek.com edited /Firefox_Social_API (+264) "linky, see also" (view diff)
#
tantek.com edited /Firefox_Activations (+86) "linky, see also" (view diff)
#
tantek.com edited /Firefox_Share (+55) "linky" (view diff)
#
tantek.com created /Mozilla (+115) "stub" (view diff)
#
tantek.com edited /Firefox (+71) "more see alsos" (view diff)
#
loqi.me created /iOS (+93) "prompted by tantek and dfn added by aaronpk" (view diff)
#
aaronparecki.com edited /iOS (+80) "see also" (view diff)
nitot joined the channel
#
tantek.com created /Apple (+253) "stub with dfn, product, software, services links" (view diff)
#
tantek.com edited /Apple (+120) "tie-in to IndieWeb, UX, principles" (view diff)
camerongray joined the channel
#
gRegorLove On https://indiewebcamp.com/reply#Post_a_reply is the rel=in-reply-to for backwards compatibility?
#
gRegorLove (on permalinks only)
#
tantek gRegorLove: there are two possible views (even futures)
#
tantek one is that rel=in-reply-to is for backcompat only
#
tantek the other is the notion that it may be possible to depend on even less markup in a web where everything is assumed to be page-level
#
tantek that is, pages that each just represent one object and thus don't need explicit markup to express that they are an object (e.g. h-entry)
#
tantek a potentially even simpler / less markup indieweb as it were
#
gRegorLove Hm. Interesting.
#
tantek it's always possible that someone, or potential future someones, will come along and figure out how to do indieweb plumbing even more simply than we have so far
#
gRegorLove I only have u-in-reply-to currently was wondering if I should add the rel.
#
tantek what is rel-in-reply-to?
#
Loqi rel=in-reply-to is a rel value for indicating the destination of a link is an original post that the current page is a reply to http://indiewebcamp.com/rel-in-reply-to
#
gRegorLove I don't see much reason to add it, though.
#
tantek.com edited /rel-in-reply-to (+177) "add Why just to document current lack of a specific reason" (view diff)
#
tantek anyway, I haven't shared much about that potential future simpler indieweb vision because it didn't seem necessary
#
tantek I'm more acknowledging it as a future possibility
#
gRegorLove Surprisingly few (one) IndieWeb Examples on /reply.
#
tantek that someone(s) will come along, proclaim even what we are doing with h-entry, microformats2 etc. is more complex than necesssary, and the provide a working/workable brainstorm about how purely page-level semantics can solve 90% of the use-cases or something
#
tantek hmm
#
tantek ah the more link is buried
#
tantek.com edited /reply (+21) "/* IndieWeb Examples */ explicitly note more examples" (view diff)
KartikPrabhu joined the channel
#
tantek gRegorLove: see https://indiewebcamp.com/reply-context#IndieWeb_Examples
andicascadesf joined the channel
#
andicascadesf Hey guys!
#
tantek gRegorLove: see also https://indiewebcamp.com/reply-context-examples#Indieweb_Examples
#
andicascadesf I’m doing some research for an event on security and privacy and want to hear from some of my friends here at IWC.
#
andicascadesf https://blog.mozilla.org/privacy/2013/05/21/exploring-the-emotions-of-security-privacy-and-identity/
#
andicascadesf Sorry, here is the article I’m curious of.
#
andicascadesf *graphic
#
andicascadesf http://ffp4g1ylyit3jdyti1hqcvtb-wpengine.netdna-ssl.com/privacy/files/2013/05/Screen-Shot-2013-03-22-at-10.30.44-PM1.png
#
tantek.com edited /reply (-19) "/* More Examples */ link directly to where the most other examples are" (view diff)
#
andicascadesf Do you agree with this graphic about the mainstream users view of identity?
#
andicascadesf Where does Indie Web Camp fit in?
#
tantek what is privacy?
#
Loqi Privacy is “the ability of an individual or group to seclude themselves or information about themselves and thereby express themselves selectively” https://indiewebcamp.com/privacy
#
tantek what is security?
#
Loqi security in the context of the indieweb may refer to security concerns regarding personal domains, web hosting, https setup, private data, identity etc https://indiewebcamp.com/security
#
tantek what is identity?
#
Loqi A personal domain is a domain name that you personally own, control, and use to represent yourself on the internet https://indiewebcamp.com/identity
#
snarfed andicascadesf: i hate to say it, but i don't quite get that diagram
#
andicascadesf Neither do i!
#
snarfed ah ok :P
#
snarfed i get the venn diagram part, it makes sense, but not how the dotted line and other phrases connect
#
andicascadesf This isn’t 100% scentific, but would it be safe to say that there are 2 t ypes of users (and a big spectrum)
#
tantek andicascadesf: that ^^^ is how privacy, security, and identity fit into the Indie Web
#
andicascadesf *two types of users when it comes to those who care about protecting their online identity
#
tantek (also note that the blog post you cited that the diagram is from is >2 years old and a bit out of date)
#
andicascadesf @tantek
#
andicascadesf i agree
#
tantek for anything else it's probably worth doing a modern user-study to actually determine anything more qualitative
#
andicascadesf I’m just being general, as I don’ thave time to do a user study
#
tantek i.e. personal anecdotal experience is that there are as many views of privacy & security as people you talk to - no patterns of "two types"
#
tantek but if you have evidence for such "two types", definitely share!
#
tantek if you're just looking for general assertions then the above definitions of security, privacy, identity ought to do
#
andicascadesf well there’s always a spectrum.
#
andicascadesf http://uxnight26.eventbrite.com/
#
andicascadesf My goal is to teach designers that there are some people who care about their online privacy, and go through crazy measures to protect it, like always writing messages that are encrypted, cautious about which apps they use, use 2 factor authentication, careful about passwords.
ageis_ and snarfed joined the channel
#
andicascadesf but it seems like more web site sign-ups and registrations happen with convenience.
#
bear if it's a talk to a UX crowd then get them all to realize that passwords are an anti-pattern http://shapeshed.com/passwords-are-still-an-anti-pattern/
#
andicascadesf Thanks I’ll add it to the list of research.
#
andicascadesf @ageis can you scroll up?
#
ageis andicascadesf: no, I can't see anything from before I joined
#
bear this channel is logged - see the topic
#
ageis ok
#
KartikPrabhu ageis: https://indiewebcamp.com/irc/2015-10-19#bottom logs
#
bear or ... what KartikPrabhu++ just said
#
bear KartikPrabhu++
nitot joined the channel
#
Loqi KartikPrabhu has 121 karma
#
andicascadesf I’m conflicted between a few different topics.
#
andicascadesf Here are some of my goals:
#
andicascadesf 1. Encourage designers to create more tools that conveniently allow the mainstream user to protect their personal information, encrypt conversations.
#
andicascadesf 2. Educate the “mainstream / everyday” user on current tools that are available, and what works and what doesn’t work. Show the # of steps it takes to encrypt and decrypt email.
#
GWG Evening
#
andicascadesf 3. *this is debatable, but I feel very strongly about this*
#
andicascadesf 3. Show how Microsoft, AT&T and large corporations are secretly selling the personal information of the people we are designing these products for. They are not only betraying their customers – they are going behind the backs of their employees, who spend months / years / whatever creating these features. If designers & developers were more aware of what is happening, they will make more responsible decisions on
#
andicascadesf whom they choose to work for.
#
andicascadesf I’m not sure how much of #3 I want to talk about, but I think it’s a strong argument to drive change. Without designers + developers, there is no product.
#
andicascadesf Thoughts?
#
andicascadesf For me to stand up and share my political views is a personal choice, so I will make that decision according to what feels right.
#
snarfed consider text or IM instead of email for #2. they're both way ahead on UX
#
ageis snarfed: that's the point