#indiewebcamp 2014-04-17
2014-04-17 UTC
pauloppenheim, caseorganic, lukebrooker and cnxtrm joined the channel
# 
gRegor` If you have anchors on paragraph elements, webmentions could be sent to those. kylewm, KartikPrabhu
# gRegor` The mention is technically to the page, but you could handle the anchor in the target
# 
KartikPrabhu gregor`: true. I have starting adding anchors to section titles. might add mention support there
# KartikPrabhu gregor`: the more imp. problem would be how to expose the anchors to the reader
# KartikPrabhu UX/UI - wise
# snarfed yup, anchor in target is what people have tried so far, e.g. https://github.com/pfefferle/wordpress-webmention/issues/21
# gRegor` Dave Winer has paragraph symbols appear when hovering over, and as a link I believe.
# gRegor` Actually, guess he has a hash link now: http://scripting.com/2014/04/16/thePressIsntGettingHeartbleed.html
# KartikPrabhu gregor`: did see that. but not good for mobile due to hover
# gRegor` It's not hover now
# KartikPrabhu oh interesting. also another thing would be what to do with those comments. how to present them etc...
# gRegor` Just a link at the end of the line. Which I can't say looks very good... but it works :)
# gRegor` Heh, interesting
# KartikPrabhu yeah. not very robust but might work mostly
# gRegor` If it's smart, it checks for uniqueness and appends a number to avoid collision
# gRegor` Oh, it's javascript.
# gRegor` That seems... inefficient.
# KartikPrabhu hmm yeah
# KartikPrabhu but anyway. this seems like a good indie problem
# KartikPrabhu eh! i would. just this morning I got a bunch of very specific feedback from cweiske on IRC that could have been a frag-level comment
snarfed joined the channel
# 
KevinMarks2 The practical anchor is a text snippet
# KevinMarks2 I wonder if we can polyfill a way to do that
# 
JonathanNeal polyfill what?
paulcp joined the channel
# KevinMarks2 Straw man idea: link to page with appended text excerpt after ##
# KevinMarks2 To scroll to last bullet
# JonathanNeal KevinMarks2: elaborate on what js you’re looking for? I’d like to help if I can.
# KevinMarks2 Have a script that parses that client side
# JonathanNeal parses `that`, what is that? And what does it do with said thing?
pfenwick joined the channel
# KevinMarks2 If the fragment identifier has two # in, search the body text for the string and highlight
# JonathanNeal You want a script that interprets double hashes as a link to the first result of the string on the page and scrolls to that section?
# KartikPrabhu that sounds right
# JonathanNeal I could write that for you. You want it to be case insensitive?
# KevinMarks2 Yes
# KevinMarks2 I think so
# JonathanNeal Great. Have a sample page I can work with, or may I use the tantek page you mentioned?
# KevinMarks2 Then present it as both a script to include in your site, and a browser extension for other people's sites
# KevinMarks2 One of my transcript pages might be good
# JonathanNeal Send me a link. I’ll start on it right away.
# KevinMarks2 http://www.kevinmarks.com/w3cannotation.html send appropriate
# KevinMarks2 s/send/seems/
# 
Loqi KevinMarks2 meant to say: http://www.kevinmarks.com/w3cannotation.html seems appropriate
# JonathanNeal Great, I’ll get started on that right now. I should have something for you in the next 15 minutes.
# KartikPrabhu KevinMarks2: is that robust enough against edits?
# KartikPrabhu which seems to be something annotations folk are worried about
# JonathanNeal Elaborate?
# KevinMarks2 :D
# KartikPrabhu my page didn't scroll! ;)
j12t joined the channel
# KevinMarks2 Give Jonathan a chance
# KartikPrabhu KevinMarks2: confused about that line you linked to.
# KartikPrabhu what happens if you highlight something and then the author makes edits?
# JonathanNeal Should this search be aware of hidden text or stick to what is visible?
# KevinMarks2 Visible makes more sense to me by default
# KevinMarks2 Then the thing you linked to has changed, and the highlight will fall
# JonathanNeal You got it.
# JonathanNeal About done.
# KartikPrabhu I see. so if people are doing federated annotations, then the onus is on the highlighter to save a copy
# KartikPrabhu just like reply-contexts
# KartikPrabhu oh! in fact this is exactly like reply-contexts!
# KartikPrabhu surfs his brain wave
# KevinMarks2 That was what was going through my head in the annotations event
# KevinMarks2 Robust anchors
# KartikPrabhu cool! sounds easier than the whole xpath thing I've been reading about
# JonathanNeal Should it do anything else, like auto focus the element?
# KartikPrabhu JonathanNeal: highlight it?
# JonathanNeal Not sure what that would look like without getting extra hacky.
# JonathanNeal But the script is complete. Let me upload it.
# KevinMarks2 Robust anchors herbs towards keeping your own copy of the linked to object
# KevinMarks2 s/herbs/tend/
# JonathanNeal what do we call this, KevinMarks2?
# JonathanNeal search anchor?
# JonathanNeal dynamic / text / search anchor
# KartikPrabhu auto-anchor ?
# JonathanNeal experimentally updated to support Firefox and focus the node, kbs, KartikPrabhu, KevinMarks2
# KevinMarks2 Not working on Firefox android, should I retry?
# JonathanNeal Yes, refresh.
# KartikPrabhu JonathanNeal: how much does it highlight? the closest sentence?
# KartikPrabhu JonathanNeal++ for something so cool in under 15mins!
# JonathanNeal It highlights the deepest, first matching element.
# KartikPrabhu kbs kylewm: you guys need to learn to ++ better :P
# KevinMarks2 Excellent!
# KartikPrabhu JonathanNeal: on clicking away it un-highlights. Is that intended?
# JonathanNeal KartikPrabhu: that is the default behavior when clicking away from any focused element.
# KartikPrabhu JonathanNeal: I see, you're using focus. cool!
# KartikPrabhu lol! eh in due time kids
# KartikPrabhu ;P
# JonathanNeal How should autoanchors interpret special characters, like the plus sign in “lets+you+disclaim” or “webmention + pingback”?
# KartikPrabhu nevers claimed to be any good...
# KevinMarks2 Hmm. I was using + signs because I'm old and its hard to type %20 on a phone
# KartikPrabhu !tell snarfed: running activity-streams tests got this error " File "./alltests.py", line 22, in <module> filesystem.load_template_source = filesystem._loader.load_template_source AttributeError: 'module' object has no attribute '_loader '"
# JonathanNeal KevinMarks2: I was thinking similarly, except (quite fortunately in our usecase) we have text that happens to contain a + sign.
# JonathanNeal In that case, it would need to be escaped.
# KartikPrabhu JonathanNeal: that last one did not highlight in FF on Ubuntu
tantek joined the channel
# KevinMarks2 Not sure about indieWebcamp NY yet
# KevinMarks2 I just thought of a name : fragmentions
ngoldman joined the channel
hugoroyd joined the channel
# gRegor` Woo, got MySQL UNION query working. I should have a nice list of webmentions and comments interleaved together on my posts soon-ish.
# KartikPrabhu KevinMarks2: hmmm nice!
# KevinMarks2 Tantek is going
JonathanNeal joined the channel
# 
tantek Could you create http://indiewebcamp.com/2014/NYC/G uest_List by copying from http://indiewebcamp.com/2014/Cambridge/Guest_List and search/replace, and move our RSVPs to that page?
# 
tantek.com edited /2014/ (+75) "add UK and Cambridge as what you might be looking for" (view diff)
dybskiy joined the channel
# gRegor` Don't forget to use http://indiewebcamp.com/Template:Attendee
# 
david.shanske.com created /2014/NYC/Guest_List (+3600) "Created page with "== IndieWebCamp New York 2014 == * '''When:''' <span class="dt-start dtstart"><time class="value">2014-04-26</time> <time class="value">09:00</time></span> through <span class="d..."" (view diff)
# gRegor` Oh, misread
# david.shanske.com edited /2014/NYC/Guest_List (-86) "/* IndieWebCamp New York 2014 */" (view diff)
# tantek people are using "how indie web" as a way to rate/evaluate services: https://twitter.com/ShenoaLawrence/status/456594465914753024
# 
@ShenoaLawrence @erinanacker @higley I use Zoho. It’s been good so far, but I’m not sure how indie web they are. (twitter.com/_/status/456594465914753024)
scor joined the channel
dybskiy_ joined the channel
dybskiy joined the channel
# KevinMarks and it works for accents: http://sandbox.thewikies.com/autoanchor/##Çelik
# KevinMarks JonathanNeal: do you like the name fragmentions?
# JonathanNeal I do.
# KevinMarks I think we should write this up
# KevinMarks nice clean code: http://sandbox.thewikies.com/autoanchor/script.js
# KevinMarks the element is highlighted
# JonathanNeal tantek: browser?
# KevinMarks it was working on FF on android
# JonathanNeal http://sandbox.thewikies.com/autoanchor/ i put some example links at the top, those don’t take you to their respective locations?
# KevinMarks hm, case sensitive?
# tantek JonathanNeal - Selection object to actually select/highlight the text: https://developer.mozilla.org/en-US/docs/Web/API/Selection
j12t_ joined the channel
# JonathanNeal tantek: i can implement that if you like. i think it’s worth discussion whether fragmentions/autoanchors should work like anchors, focus, find, or some combination.
# JonathanNeal since you’ve all been discussing this for some time, what are the usecases / what do the people want?
v0 joined the channel
# KevinMarks2 Who's good at browser extension
# KevinMarks2 Could we wrap Jonathan's fragmentions code into one, so it works on arbitrary sites?
# JonathanNeal I may have it.
brianloveswords joined the channel
# KartikPrabhu JonathanNeal: re: use cases . i think the idea is to implement something like the annotations thing people have been talking about. a good first step seems to be able to link arbitrary parts of a document
# KartikPrabhu likes this ## notation with the accompanying JS!
# KartikPrabhu ## also has the English translation of "fragment of a fragment"
# KartikPrabhu which makes more senseto a human than xpath or something
# KartikPrabhu KevinMarks2: can ##a be used to mean fragment-id="#a" in HTML? If yes, would this not be a conflict?
# JonathanNeal Okay, now I know how extensions work. We just need an icon :)
# KartikPrabhu JonathanNeal: ## in a square work?
# KevinMarks2 Can fragments start with #?
# JonathanNeal KevinMarks2: yes.
# KartikPrabhu KevinMarks2: yes that was my question. I think they can
# KartikPrabhu but I don't know the spec very well... and couldn't read through it intelligently :P
# KevinMarks2 Reading the rfc I think they can't
# KevinMarks2 sub-delims = "!" / "$" / "&" / "'" / "(" / ")" / "*" / "+" / "," / ";" / "="
# KevinMarks2 No # in there
# KevinMarks2 From the abnf in appendix A of rfc 3896
# KevinMarks2 Hm. That may make this an invalid url
# KevinMarks2 But if it did, it can't be addressed by an excellent compliant url
# JonathanNeal Okay, the extension is finished. Not sure where to put it.
dybskiy joined the channel
# JonathanNeal Would someone provide a detailed description of fragmentions?
tantek joined the channel
# tantek the ## links don't work in Colloquy, it escapes the 2nd # into %23, e.g. KevinMarks, clicking on the link with my surname goes to: http://sandbox.thewikies.com/autoanchor/#%23%C3%87elik
# JonathanNeal over-escaping, hmm.
# JonathanNeal IMHO, hash anchors as hashbangs or storage are not really anchors, they’re hacks to avoid page reloads, whereas ## is actually a kind of anchor.
snarfed joined the channel
lukebrooker joined the channel
dybskiy joined the channel
# KartikPrabhu snarfed: yeah just running alltests.py gives that error
# KartikPrabhu hmm will do
# KartikPrabhu snarfed: np! chanign things now
# KartikPrabhu !tell snarfed: after commenting out those lines I get "ImportError: No module named mox
# KartikPrabhu KevinMarks2: too bad about the ##. I kind of liked them. It'll still be good if some other solution could be salvaged
snarfed joined the channel
melvster joined the channel
# KartikPrabhu oh ok
# KartikPrabhu snarfed: also appengine.api I am assuming that is google app engine?
# KartikPrabhu this is the one linked in the readme yes?
# KartikPrabhu cool! I'll work through those...
# KartikPrabhu oh installed mox through pip
# KartikPrabhu 51MB! holy cow that's huge
# JonathanNeal KartikPrabhu: what do you mean “too bad” did I miss something?
# KartikPrabhu JonathanNeal: I meant the fact that fragment ids can have id="#a" so a link to ##a would actually be a link to a fragment and your JS would hijack that
# KartikPrabhu IMO it would be best if the default browser functionality is not hijacked
# JonathanNeal That’s sad. Albeit I can’t find any examples of it on Google. Is this happening in the real world?
# JonathanNeal Also, the double hash is currently not a URL code point and generates an error in the validator.
# KevinMarks2 I say show off the demo and discuss it, we can come up with another syntax
# KevinMarks2 Or do a global crawl to see if #-preceded ids exist in the wild
# JonathanNeal KartikPrabhu: 1. relatively no one has used ## thus far, 2. existing ## would still work because i do not stop propagation, and 3. ## appears to be presently invalid, thus providing an opening.
# tantek aaronpk, btw - results of me asking about that pattern of overseeing/using of concepts like webintents, annotations, cards: https://twitter.com/t/status/456552583499243521
# 
@t name of illusion causing proponents of a concept to see most things as that concept? e.g. webintents,annotations,cards (ttk.me t4Vb1) (twitter.com/_/status/456552583499243521)
# JonathanNeal KevinMarks2: sure, would you like to throw up an example page somewhere? And when should we tweet this to get feedback?
# KartikPrabhu JonathanNeal: KevinMarks2: I support the throw it in the wild and test philosophy. I don't know of anyone who has seriously used id="#a" or if browsers use those correctly, anyway. (Though I have used it unintentionally)
# KartikPrabhu Bipedal walking simulations -- might be of interest/amusement to folks here: https://www.youtube.com/watch?v=rBHjx4dYxjA
# JonathanNeal so, throw a demo somewhere and tweet it tomorrow pt 10am?
# KartikPrabhu JonathanNeal: tweet/document it on indiewebcamp wiki ... all sound good
# JonathanNeal Awesome, is someone already doing that on iwc?
# JonathanNeal is not volunteering.
# KartikPrabhu lol!
# KartikPrabhu ok I'll make a /fragmention page
# JonathanNeal ++KartikPrabhu
# KartikPrabhu no the other way around ;)
# JonathanNeal KartikPrabhu++
# KartikPrabhu w00t!
# JonathanNeal Your karma points can now buy cigarettes in the states.
# JonathanNeal I will link to the iwc page on the github page. :)
# KartikPrabhu started a basic dumbass page...(never made a page from scratch before)
# KartikPrabhu KevinMarks2, JonathanNeal: please add to this: http://indiewebcamp.com/fragmention
# JonathanNeal Nice, KartikPrabhu.
grantmacken joined the channel
# JonathanNeal Another thing I can do is ignore ## when there is a matching, hash-lead ID that exists. Thoughts?
# KartikPrabhu JonathanNeal: that might work for your JS but will fail for a general linking idea
# JonathanNeal Yes, though I see it both ways. I think this article and the feedback will matter a great deal more than these workarounds.
# KartikPrabhu JonathanNeal: agreed!
scor joined the channel
# Loqi [mention] https://brid-gy.appspot.com/rsvp/facebook/11500459/685344088192195/698023475 linked to http://indiewebcamp.com/events/2014-04-23-homebrew-website-club (webmention)
cweiske, j12t and LauraJ joined the channel
# JonathanNeal All in favor of me dropping case-sensitivity?
# KevinMarks2 Yes, I think case insensitivity is a simple example of fuzzy matching
# KevinMarks2 Which was much discussed at the annotations thing
# KartikPrabhu KevinMarks2: JonathanNeal: is it bad to just keep it?
# JonathanNeal KevinMarks2: as in, fuzzy matching is good?
# KevinMarks2 The high level debate as I saw it was that identifying the thing being annotated was more than just linking
# KevinMarks2 Lots of examples of wanting annotations to map across editions of books, even translations
# KevinMarks2 And edits - versions of documents
# KartikPrabhu KevinMarks2: that's all good if annotaters are storing contexts locally. but still is there a need to drop case-sensitivity?
# JonathanNeal KartikPrabhu: I was writing documentation and came up with this “Matching case is strict, but makes it easier to target specific phrases. Ignoring case is cavalier, but makes it easier to target anything.” If you agree with that statement, then my judgement was that anything > specificity.
# KartikPrabhu JonathanNeal: isn't marking specific things the point though... ?
# JonathanNeal Do you think people expect ##call+the+dogs to not match “I told him to call The Dogs”?
# KartikPrabhu lets leave that to "test in the wild" shall we?
# JonathanNeal Agreed.
# KartikPrabhu yeah so if currently you have case-sensitivity leave it at that and see if people complain
# KartikPrabhu and vice-a-versa
# KartikPrabhu used 'vice-a-versa' incorrectly but no on complained!
# KevinMarks2 I think case sensitivity is harder to get right. What about white space sensitivity?
# KevinMarks2 Using more words is a better way to be specific than relying on case
# KevinMarks2 Will do. I will write this up in a bit.
# JonathanNeal KevinMarks2, KartikPrabhu, tantek, have tentatively added my thoughts to https://github.com/chapmanu/fragmentions may I add some of them to IWC page?
melvster joined the channel
# KartikPrabhu JonathanNeal: as tantek said... just add them someone will edit if they want to do so
eschnou joined the channel
# JonathanNeal Woot.
# KartikPrabhu nice!
# KartikPrabhu speaking of annotations has anyone played with the medium annotation on mobile yet.. for instance: https://medium.com/cool-code-pal/b21bfc1f63c7
# arcatan i've played with it in the sense that i always misclick and accidentally open the annotation column :(
# arcatan or mistouch
# KartikPrabhu lol! arctan... was just discussing the "too small target" issue with a friend
# KartikPrabhu also once opened it seems quite broken
# tantek KartikPrabhu: feel free to start http://indiewebcamp.com/annotation and add that medium annotation thing as a silo example ;)
# KartikPrabhu yes.. planning to play around with it a bunch later. will add observations
# tantek would be good to See Also [[fragmention]] too ;)
# KartikPrabhu on it!
# KartikPrabhu what's a good defn of annotation :P
# KevinMarks2 Heh. We spent a day on that with the w3c and I'm not sure we got a good answer
# KartikPrabhu haha! I am going to go with 'fragment level comments' which make more sense to me anyway
jsilvestre joined the channel
# KartikPrabhu annotation page stub! everyone add examples and screenshots!
# KartikPrabhu has anyone been POSSEing flickr photos? trying to convert a friend to indeweb ways...
Sebastien-L, friedcell and dybskiy joined the channel
# Garbee Hm. A better definition of an annotation is probably "A comment added to an arbitrary segment of content."
# Garbee For instance on the Web Platform Docs wiki ( http://docs.webplatform.org/wiki/Main_Page ) you can highlight any text, and then an annotation button will appear for you to add a comment about that specific part of content.
# Garbee Annotations are really much more granular than just fragments of a page (which is another kind of (deprecated) comment system WPD has.)
# KevinMarks that was the point of fragmentions
# Garbee just learned something for today.
# Garbee at just before 6AM I think that is pretty good. :D
# Garbee cweiske: yea, some system one sec and I'll go over the ML.
# Garbee https://hypothes.is/alpha/
# Garbee It is a *really* nice system.
# Garbee It is actually the kind of UX I was thinking of for a new bug reporting engine.
# Garbee So people could report issues without leaving the content and have the exact place of the bug tied to the report (compared to relying on either a page with a bunch of content given, or even just "I saw X, go fix it." with no link. :(
# Garbee Yea, for any kind of detailed writing, this hypothes.is is awesome.
# Garbee cweiske: If you want, you can jump into #webplatform and bugger renoirb about it if you want. He is the sys op for WPD and is the one that setup hypothes.is afaik.
# Garbee (he is probably asleep right now though, so wait until later if you don't get a response.)
adactio joined the channel
# KevinMarks part of the idea of fragmentions is to enable a webmention to target arbitrary text on the page
# Garbee Yea, I remember there was some way with Chrome dev tools to point something out, then generate a fragmented URL to it even though there was no ID or other easily targetable thing around it.
# Garbee I wonder how hypothes.is targets stuff and remembers...
# Garbee Because I think it has done a great job of annotation UX. Much more granular than the traditional way developers think about fragments (mostly just being headings of sections.)
# KevinMarks because it requires you to https://hypothes.is/alpha/##Create%20an%20account
# KevinMarks the do it with http://sandbox.thewikies.com/autoanchor/##multiple selectors
# KevinMarks :D this is fun
# Garbee :( I thought we had hypothes.is working off Wiki accounts (kinda like SSO, where it is all internal.)
# Garbee That is a let-down.
# Garbee KevinMarks: Do you know if http://www.w3.org/community/openannotation/ is the best way to keep up with this stuff? I'm not seeing much else from the W3C's site on Annotations (aside from the page from the workshop that just occured.)
# Garbee I should probably ask Doug too. He'd know what else is going on here. He is into them as well.
# KevinMarks that was who organised the event we went to
# Garbee Yea, saw that on this page: http://www.w3.org/2014/04/annotation/
# Garbee The whole URL changing thing is major though. If there is a solution to that, (with as little/no extra markup needed) then it would be epic.
# KevinMarks which thing?
# Garbee Tantek said " the anchoring problem is important - cool URLs don't change " in the segment of meeting notes here: http://sandbox.thewikies.com/autoanchor/
# Garbee But, URLs do change (especially with idiots like me switching blog systems every few months due to not being happy/whatever.
# Garbee So being able to somehow transition annotations across URL schemes would be killer.
# Garbee Also for things like WPD where we may move pages around in getting things organized, but people are annotating now with the current URL.
dybskiy joined the channel
# KevinMarks just added fragmention support to my blog
# KevinMarks my Blogger blog....
# KevinMarks thought that fails onthe mobile template
# KevinMarks maybe the name is a bet too clever for it's own good
# Garbee Whoops. I thought KevinMarks typo'd when he said fragmention.
# Garbee :P
# Garbee Not sure if too clever or I'm too stupid...
# KevinMarks naming things is hard.
# KevinMarks I should grab the domain tho'
# KevinMarks .org or .com?
# Garbee both just to be safe?
# Garbee I'd probably go with .com though initially. Since it is a piece of tech and not necessarily an organization.
pasevin joined the channel
# KevinMarks .io is hipster
# KevinMarks :D
# Garbee But like $90 a pop.
# Garbee :/
# KevinMarks yeah
# Garbee I have garbee.me couldn't be happier!
# Garbee jonathan@garbee.me for my email, and jonathan.garbee.me for my blog. So awesome to have those.
pfefferle joined the channel
# KevinMarks hm, it's 4:04 am will my links fail?
# Garbee If that was supposed to highlight the Larry Page quote, it worked fine.
# KevinMarks I thought it was an apposite one...
# KevinMarks I think that's cos my blog markup is a bit crappy
# KevinMarks it's an <em> with <br> in, so it's not got a <p> to highlight
# KevinMarks sorry, it was 11 years ago and I hadn't learned good semantic markup then
# KevinMarks wow, what a lot of crappy tlds .technology?
# KevinMarks .guitars?
dybskiy joined the channel
# KevinMarks though I am tempted to register foo.camp
# KevinMarks that's what they want you to do...
benprew joined the channel
scor joined the channel
# Garbee KevinMarks: Yea, Generic TLD's are going to make a mess.
adactio joined the channel
# Garbee .dev is being sought by Amazon (first in line) and then Google.
scor joined the channel
# Garbee I'd love for .dev and .local to be made reserved.
# Garbee Because a lot of devs use those two locally for testing/dev stuff. Like I route all .dev on my router to my local server box. That way I'm not changing the hosts file on all my boxes.
# KevinMarks .local is reserved
# KevinMarks at least, it's used for multicast DNS on the local network
# KevinMarks .localhost is explicitly reserved
# 
pdurbin in my /etc/hosts on linux I see localhost.localdomain
Sebastien-L, pfefferle, dybskiy_, benprew, jsilvestre and pasevin joined the channel
# KevinMarks anyone want to proof read my post?
# KevinMarks that does make me nervous too
# KevinMarks maybe #* would be better
# KevinMarks quote
# KevinMarks quite
# KevinMarks OTOH that would mean percent encoding covers the "id starts with #" case
# KevinMarks #*foo
# KevinMarks 'cos * means both footnote to normal humans and wildcard match to geeks
# KevinMarks do I tweet a link to my page, or to my blog?
fungoat joined the channel
# KevinMarks or to fragmention.com
# KevinMarks :D
# KevinMarks hm, 6am. I should go to bed
chloeweil, dybskiy, cnxtrm, tobiastom_, ireheart, brianloveswords, Sebastien-L, pasevin and gRegor` joined the channel
# Loqi [mention] http://www.marcus-povey.co.uk/2014/04/17/syndication-support-for-idno-email-posting/ linked to http://indiewebcamp.com/POSSE (pingback)
squeakytoy joined the channel
tantek joined the channel
_6a68 joined the channel
snarfed, dybskiy and kbs joined the channel
# 
aaronpk that's what I'm attempting to solve with this http://client.indieauth.com/signin
friedcell joined the channel
# tantek aaronpk - love that you developed it with the real world example of your pushups app rather than all the architecture astronomy* that went into all the OpenID abstractions. * http://www.joelonsoftware.com/articles/fog0000000018.html
# tantek nicely done on exploring fragmentions KevinMarks and JonathanNeal - especially the /fragmentions page!
iangreenleaf joined the channel
# KevinMarks2 http://www.kevinmarks.com/fragmentions.html post for the West coast morning people
arcatan joined the channel
# adactio KevinMarks2: I've linked to your post here: http://adactio.com/journal/6751/
# KevinMarks2 Maybe I should camelCase it. fragMentions
# KevinMarks2 Also I need to make my blogger template work on mobile
# KevinMarks2 Thanks adactio
# tantek Kevinmarks, you may be interested in: https://indiewebcamp.com/mobile#mobile_friendly :)
dybskiy joined the channel
# KevinMarks2 Aha. Kevinmarks.com is, just not epeus.blogspot.com
# KevinMarks2 Well, it uses blogger's mobile template, which does not seem as editable
benprew joined the channel
# JonathanNeal Thanks, tantek
# KevinMarks yes, first match
# KevinMarks the point is not to pick the minimal match for the link so much as to cite enough text to be unique
# KevinMarks I think ~10 words is globally unique (like the Stoppard quote)
# KevinMarks it highlights too, but I think it gets cropped by some layout issues
# KevinMarks looks like some margins are cropped?
# KevinMarks did you add the script, or was that for use in chrome?
indie-visitor, melvster and paulcp joined the channel
# V_S_C The links on IndieWebCamp comprehensively guide developers regarding different web-centric areas- silos, standards, annotations, offline apps. But I was looking to find coverage of comparative relation with extensiblewebmanifesto site promoted by Brendan Eich at Fluent2014
cweiske, snarfed, j12t and brianloveswords joined the channel
# JonathanNeal Good morning (from PT)!!
# JonathanNeal KevinMarks, tantek, are we ready for some in the wild testing of fragmentions?
# JonathanNeal aaronpk: ^ ? Really?
# JonathanNeal https://twitter.com/jon_neal/status/456839901003005952 would you (all) make a tweet yourself or retweet that? It’s start getting feedback.
# 
@jon_neal Fragmentions! A proposal/polyfill to use ## anchors as links to individual words or phrases in a document. https://github.com/chapmanu/fragmentions (twitter.com/_/status/456839901003005952)
jsilvestre joined the channel
# 
kbs kylewm: my version of erc - goes to: http://aaron.pk/a4VU1%23%23Find+the+dependencies
# JonathanNeal aaronpk: just tested your link sent at 10:00am. Worked great for me in Chrome.
# 
kylewm by it, I mean http://aaron.pk/a4VU1##Find+the+dependencies
# JonathanNeal Mike Smith pointed me to http://url.spec.whatwg.org/#url-code-points, presently valid characters include !, $, &, ', (, ), *, +, ,, -, ., /, :, ;, =, ?, @, _, and ~. That’s not to say we couldn’t try to update the ancient spec.
# kbs kylewm: yes - http://aaron.pk/a4VU1##Find+the+dependencies on FF and [an old safari fwiw] seems to strip out the second # after redirection
# JonathanNeal kylewm, aaronpk, worked for me in Chrome, but failed in FF. FF trimmed the second hash.
# kbs fwiw - http://aaron.pk/a4VU1#@Find+the+dependencies works ;-)
# kbs (as does http://aaron.pk/a4VU1#?Find+the+dependencies I guess)
# KevinMarks2 I was thinking about this while cycling.
# KevinMarks2 ## can't be a collision, as you'd have to escape the second one
# KevinMarks2 However, #* seems a reasonable alternative as * means both footnote and wildcard match
# JonathanNeal KevinMarks2: likewise, @ means mention.
# KevinMarks2 Similarly, an id can contain any character except a space
# JonathanNeal like @horsejs on social netweosk.
# JonathanNeal s/netweosk/networks/
# KevinMarks2 R
# KevinMarks2 So if there is a space in the fragment it can't be an id and must be a fragmention
# KevinMarks2 # is the signal for a text search though
# KevinMarks2 So if a fragmention is always more than one word, we may not need the ## at all, and could just overload #
# JonathanNeal Are we comfortable leaving ## in the wild for now or should I update it to use a valid convention?
# KevinMarks2 Let me get on the laptop, doing this on the phone is fiddly
# JonathanNeal yay
# KevinMarks2 Hmm the mac crashed when I shut the lid. Wonder if I lost anything
KevinMarks joined the channel
# KevinMarks interesting - CIRC shows upas an app in the dock now, even though it's a chrome app. wonder when that happened
# KevinMarks so, should we capture this validity discussion in /fragmention ?
# JonathanNeal KevinMarks: I plan to add notes about ##, #@, and #*, but I want to know if I should update the actual scripts to target a (presently) valid syntax that would also work in FF.
# JonathanNeal I’m looking for consensus.
# KevinMarks oh, I registered fragmention.com, fragmention.org and fragmention.net too...
# KevinMarks I think we are seeing strict URL parsers fail with ##
# KevinMarks so accepting #* makes sense to me
# KevinMarks they're currently redirected to my post, the indiewebcamp page and the github repository respectively
snarfed joined the channel
# KevinMarks JonathanNeal: I think your url is wrong on that page - you're http://johnathantneal.com, right?
krendil joined the channel
# KevinMarks s/johna/jona/
# Loqi KevinMarks meant to say: JonathanNeal: I think your url is wrong on that page - you're http://jonathantneal.com, right?
# KevinMarks (being tethered to a phone on the train means my wifi is spotty
# JonathanNeal I am jonathantneal, que?
# KevinMarks the link on the fragmention page was to jonathanneal.com
# KevinMarks have you logged into the wiki with that?
# JonathanNeal Whoops! I didn't write that. Haha.
# KevinMarks s/that/jonathantneal.com/
# JonathanNeal I am t
# JonathanNeal Yep
# KevinMarks 'cos your userpage is blank
cnxtrm joined the channel
# KevinMarks so, could we hook up fragmentions with webmentions and get medium-style comments on paragraphs?
# JonathanNeal KevinMarks: sowwy?
# KevinMarks this is a thought for aaronpk et al who have webmentions working
# KevinMarks another thing would be to make a purple numbers style thing that made fragmention URLs to copy per paragraph
# 
@Bjorn_W current thesis: if an online service is not sharing their software as #floss you may want to reconsider investing into it #hackpad #indieweb (twitter.com/_/status/456855035360448512)
# KevinMarks like tim bray has here: http://www.tbray.org/ongoing/When/201x/2014/04/13/Short_fragments
# KevinMarks hover over a para to see a ¶ sign, which include the id
j12t joined the channel
# KevinMarks or the # marks on http://scripting.com
# KevinMarks I think so. look for <p>, take first 5 words, make fragmention?
dybskiy joined the channel
# KevinMarks Is <p> not a suitable signal? though that would fail on my ancient crappily marked up <b> and <br> posts
# KevinMarks seems like a case for a css expression?
# KevinMarks for eg, I can see <article> being a signal for some people
tilgovi joined the channel
# JonathanNeal Catching up on the discussion. Interesting thoughts on the focus scope.
# KevinMarks hm, google hasn't indexed us yet, interesting
# 
@kevinmarks further discussions with @jon_neal about fragmentions in the #indiewebcamp irc. See http://www.kevinmarks.com/fragmentions.html and http://indiewebcamp.com/fragmention (twitter.com/_/status/456858470285721601)
# @jon_neal RT @kevinmarks: further discussions with @jon_neal about fragmentions in the #indiewebcamp irc. See http://www.kevinmarks.com/fragmentions.html and http://t.co/… (twitter.com/_/status/456858526933975040)
# KevinMarks yes, thats what set me off, and jonatahn ran with it
# KevinMarks JonathanNeal++
# KevinMarks literally ran, he had a polyfill done in 15 mins and on github in under an hour
# KevinMarks likes it when people instantiate my handwaving
ttepasse joined the channel
# JonathanNeal aaronpk: would you be willing to break down fragmention links to me? I don’t know what those are, or at least not know that term.
jedahan joined the channel
circ-user-fYlo5 joined the channel
# aaronpk the little # links here http://scripting.com/ or the ¶ links here http://www.tbray.org/ongoing/When/201x/2014/04/13/Short_fragments
# JonathanNeal is lost. sorry. would someone esplain or pm me to esplain?
# JonathanNeal thanks.
# JonathanNeal Thought #2, re: http://url.spec.whatwg.org/#url-code-points so a bunch of characters are invalid, does that limit fragmentions in general?
# KevinMarks no, because you can percent encode them
# KevinMarks "The URL units are URL code points and percent-encoded bytes"
# JonathanNeal KevinMarks: you’re right. Whew.
# KevinMarks already tried with tantek's surname
pauloppenheim joined the channel
# JonathanNeal aaronpk: breaking a bunch of articles or paragraphs into IDs seems bloated to me, but I like that it would let you match the second or third instance of a particular word or phrase.
# JonathanNeal Can this limitation in search fragmentions actually be improved to allow second, third, last matches?
# JonathanNeal KevinMarks: and that makes me happy, noted.
# KevinMarks we should test with emoji ;)
# JonathanNeal KevinMarks: the fact that that worked is awesome
# KevinMarks unicode FTW
# KevinMarks though the hotlink didn't see them
# KevinMarks that might be safari not escaping right, I think only firefox does that on copy
# JonathanNeal Would someone describe the rationale for using @ or *, or what those symbols represent in other contexts?
# KevinMarks @ is a mention, but of a person, not text - I think that's less clear (I know it's used in xml for attributes, but that is esoteric)
tantek joined the channel
# KevinMarks * is the universal footnote symbol, and also used as "wildcard match"
# KevinMarks downside is that this may imply regex to people, and we don't want 2 problems in our parsers
basal and KartikPrabhu joined the channel
# JonathanNeal Mike Taylor could not reproduce the second hash trimming in Firefox 28 and 31 nightly.
KevinMarks joined the channel
# 
@schnarfed @dewitt so true! the #indieweb people have been fighting this problem for years, and continue to this day. (https://snarfed.org/2014-04-17_twitter-dewitt-schnarfed-off-topic-but-bummer) (twitter.com/_/status/456870057025368064)
LauraJ joined the channel
# kylewm any objection to adding a 'See Also' section at the bottom of /Help:Editing that links to https://www.mediawiki.org/wiki/Help:Formatting ?
heath, dybskiy_, ttepasse and caseorganic joined the channel
# 
kylewm.com edited /posse-post-discovery (+1298) "/* Discover POSSE copies via rel-syndication */ details" (view diff)
# tantek again from the Architecture Astronaut article: "Those are all /sending/ operations, so our clever thinker invents a new, higher, broader abstraction called /messaging/, but now it's getting really vague and nobody really knows what they're talking about any more." - Wow - "all /sending/ operations" - totally predicted the "all /sharing/ operations" that Google/Android/Intents folks have gotten sucked into.
paulcp joined the channel
ttepasse and benprew joined the channel
# tantek GWG - http://david.shanske.com/ gives a 403 Forbidden and https uses a certfiicate for torahacademy.us
friedcell, _6a68, eschnou and brianloveswords joined the channel
# tantek.com edited /2014/NYC/Guest_List (+1257) "fix a bunch of ordering and section stuffs, add Jeremy as co-organizer" (view diff)
# tantek.com edited /2014/Cambridge/Guest_List (+79) "fix a bunch of ordering and section stuffs" (view diff)
# tantek brianloveswords: can you make it to at least a day of IndieWebCampNYC? or help out in anyway? Sign-up! http://indiewebcamp.com/2014/NYC/Guest_List
# tantek KevinMarks - sign-up and let's figure out flights etc.: http://indiewebcamp.com/2014/NYC/Guest_List
# brianloveswords tantek: Depending on when I get back from Boston during the day on Saturday, I can try to stop by for some of that, but I'm flying out to SFO on Sunday at 7:30am for the MoFo all hands, so I can't make any of that day.
benprew joined the channel
# Garbee haha, this is so true and hilariously written: http://waterpigs.co.uk/articles/remote-participation/
# brianloveswords tantek: I can help promote it though, I'll tweet it out and I'll mention it tonight at BrooklynJS
pasevin joined the channel
# tantek.com edited /2014/NYC (+833) "update look & feel similar to Cambridge, clean up a bit" (view diff)
# 
@brianloveswords If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456895878822125568)
# 
@dansinker RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456896013609877504)
# brianloveswords No problem tantek!
# 
@jonmarkgo RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456896126218948608)
# tantek brianloveswords: could you also create an event for it on Plancast.com and create http://lanyrd.com/2014/indiewebcampnyc/ as well?
# 
@dshaw RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456896261480656897)
# tantek brianloveswords: and then update the Lanyrd / Plancast links here: http://indiewebcamp.com/2014/NYC#RSVP please! :)
# 
@canuckistani RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456896492037746688)
# @schnarfed RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456896521641160705)
# brianloveswords tantek: I've never actually used either Plancast or Lanyrd, I'll try to figure 'em out though
paulcp and pauloppenheim joined the channel
# 
@indiewebcamp RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456897547521757184)
# 
@beaulebens RT @brianloveswords: If you're around NYC 4/26 and/or 4/27, you should go to NYC IndieWebCamp! http://indiewebcamp.com/2014/NYC/Guest_List (twitter.com/_/status/456898400190226432)
LauraJ joined the channel
# KartikPrabhu how can I upload mulitple files to the wiki
# KartikPrabhu oh noes!
# brianloveswords tantek: okay, Lanyrd and Plancast should be up: http://indiewebcamp.com/2014/NYC#RSVP, let me know if I messed anything up there.
eschnou joined the channel
# tantek awesome! Can you add those same links to the equivalent spot on the /2014/NYC/Guest_List page? near the top
ttepasse and warden joined the channel
# 
kartikprabhu.com uploaded /File:medium-annotation-icons.png.jpg "screenshot displaying annotation icons in Medium"
# kartikprabhu.com uploaded /File:medium-annotation-sidebar.png.jpg "annotation display in Medium"
# kartikprabhu.com uploaded /File:medium-annotation-sidebar-med.png.jpg "annotation display on medium-sized viewport"
# kartikprabhu.com uploaded /File:medium-annotation-overlay.png.jpg "annotation display on overlay in Medium.com"
paulcp and ttepasse joined the channel
brianloveswords joined the channel
# KartikPrabhu added some screenshots of Mediums annotation UI
dracos joined the channel
# dracos Have added fragmentions to e.g. http://federal-reserve.sayit.mysociety.org/meeting-16-september-2008##any+of+it+right or http://leveson.sayit.mysociety.org/hearing-24-november-2011/ms-joanne-rowling##certain+creatures - thanks for the idea :)
# JonathanNeal dracos++: that is a really cool usage
# snarfed GWG: consider SNI? http://en.wikipedia.org/wiki/Server_Name_Indication
# dracos JonathanNeal: I’d been idly wondering about how to link to bits within speeches that would work if, say, a speech was updated/moved (as Hansard on TheyWorkForYou sometimes does) that didn’t involve IDs, this certainly seems like one such way :)
# KartikPrabhu good to see fragmentions already being used!
# KartikPrabhu yeah that is weird :)
_6a68 joined the channel
# JonathanNeal aaronpk: i had the idea of firing an event, but the onfocus event already fires.
# JonathanNeal also, hashchange would have fired
ireheart joined the channel
# dracos aaronpk: you could do something with element.className where the focus is (or classList with slightly less browser support). element.className = element.className ? element.className + ‘ fragmention’ : ‘fragmention’
# dracos aaronpk: yes, you’d do just that once you added that line
# JonathanNeal aaronpk: good idea, hmm.
# dracos that line would add the class to the element, then you could style it separately however you wanted
# dracos you’d have to also have a line to remove the class when the hash changed to de-highlight (or whatever it was doing)
# JonathanNeal aaronpk: okay, want to see my idea?
# JonathanNeal nvm then. I had the idea of writing the fragmention to the location object.
# JonathanNeal window.location.fragmention = decoded fragmention
# KartikPrabhu aaronpk: but then that CSS would be for everything that is in focus. It would be good to add a .fragmention class to the element to keep things separate
# KartikPrabhu only 1. but do you want to have the same background colour to indicate general focus and fragmention focus?
# KartikPrabhu if so then that is fine
# KartikPrabhu yes... that is what I was saying. Focus is more general than what happens with fragmentions, and it would be good to keep them separate in CSS too
# KartikPrabhu sure thing
# JonathanNeal KartikPrabhu: 1. onfocusin you would check the URL for ##, or 2. i could expose fragmentation on the location object, or 3. I *could* also add an onfragmention event.
# KartikPrabhu JonathanNeal: As I see this situation, using JS to style something is always overkill! I would just add a class="fragmention" to the element using JS and then have whatever styles using CSS
# JonathanNeal I’m just discussing ways I can expose that to you.
# KartikPrabhu oh I see...
# KartikPrabhu JonathanNeal: oh so you are saying that fragmention.js should not add the class but trigger some event and the site can do the rest...
# JonathanNeal Yes.
# KartikPrabhu aaronpk: and I used tab and the highlight on the fragmention was gone even though the URl still has the fragmention
# KartikPrabhu that i think is counter-intuitive... but that might be just me
# KartikPrabhu even a stray click removes the focus and hence the highlight
# dracos KartikPrabhu: no, it is to me too. I’d just add a class as per my comment above, I’ll probably make my fork do that
# KartikPrabhu aaronpk: yeah. tie-ing it to focus might not be best. But maybe we should wait for more feedback on this
# 
@CaptainCalliope @ownCloud @fkarlitschek There's an IndieWebCamp event in NYC next weekend. Interested in joining a Boston delegation? http://indiewebcamp.com/2014/NYC (twitter.com/_/status/456922280275476480)
# 
aaronparecki.com edited /fragmention (+52) "/* indieweb implementations */ note custom CSS" (view diff)
# KartikPrabhu aaronpk: yeah...
# KartikPrabhu and that way readers can also go on read other things and then come back
# dracos aaronpk: have put class additition/removal in my add-a-class branch: https://github.com/dracos/fragmentions/compare/add-a-class . Just fixing it to work when your client goes to …#%23… which my IRC seems to do
pfenwick joined the channel
# JonathanNeal aaronpk: :fragmention would be a separate discussion, yea.
KevinMarks joined the channel
kbs joined the channel
# KevinMarks Right, off to get bandwidth and power before caseorganic talk
# dracos aaronpk: SayIt now works with encoded # in the fragment, and is adding a ‘fragmention’ class that highlights: http://leveson.sayit.mysociety.org/hearing-24-november-2011/ms-joanne-rowling?c#%23certain+creatures
# dracos (not removing it on hashchange, just debugging that!)
# dracos kbs: Thanks :)
# KartikPrabhu dracos: nice work! I didnt anticipate the remove class on fragment change probelme
# JonathanNeal dracos: let me affirm how well you are keeping my code conventions. Some really good pull requests will change unnecessary things and your script is just as I would have written it.
# dracos KartikPrabhu: I had the code, but I’d forgotten className isn’t self-modifying ;) Is working now.
# KartikPrabhu on sayit?
# dracos KartikPrabhu: yeah
benprew joined the channel
# dracos JonathanNeal: Do as you would be done by, I always say :) Even if I have to do a horrible space->tab replacement each time ;)
# KartikPrabhu dracos: doesn't seem to be removing the class
# dracos KartikPrabhu: the HTML is heavily cached (an hour by default IIRC), so you’re probably seeing that, change the query parameter to check
# JonathanNeal Yes, if you had polluted my code with double spaces I would have cried.
# dracos KartikPrabhu: sorry, I didn’t think it worth bringing out the varnishadm command :)
# KartikPrabhu dracos: working confirmed! nice
# KartikPrabhu dracos++
# JonathanNeal dracos: at first, I was like “Why is he wrapping the .+, ugh, I’ll have to change this” then I saw you use it and was like “Nice optimization” and then I thought about why you would need to use it, and I was like “Thank you, dracos”.
# gRegor` Nice choice of Jenn Schiffer's article for /annotation page, KartikPrabhu. Haha
# gRegor` She's hilarious
# dracos JonathanNeal: :-) (and change requested made, flattery will get you everywhere)
# KartikPrabhu gRegor`: yeah! it is also funny how most people who comment don't get the sarcasm!
ttepasse joined the channel
jedahan joined the channel
# gRegor` Aha. wiki page titles have text-transform: capitalize
# gRegor` I was wondering why it showed up like that even though the page was lowercase
KevinMarks_, pfenwick and _6a68 joined the channel
# gRegor` kbs: I remembered BitMessage today and was going to ask your thoughts on it, if you were familiar. https://bitmessage.org/wiki/Main_Page
# KevinMarks_ reads logs
# gRegor` I think I actually installed it and ran a test with it once, months ago. Interesting, though not sure it's very feasible.
# gRegor` Figured I'd mention it at least.
# KevinMarks_ dracos++
# KevinMarks_ you added fragmentions to shakespeare! http://shakespeare.sayit.mysociety.org/much-ado-about-nothing/act-iv/scene-i-a-church##link+my+dear+friend
# dracos KevinMarks_: nice quote :)
# KevinMarks_ I already loved sayit for making it easy to grep shakespeare, but linking to a line in scene context is better
jedahan joined the channel
# dracos KevinMarks_: It would have to do a bit more work to be cross-line (and then it might as well treat whitespace as “ “), and it still highlights the whole speech, as there aren’t any paragraphs, but yeah, more robust than most things I can think of
# KevinMarks_ highlighting the speech is good for actors... (my son did Tempest and Twelfth Night since September)
_6a68 joined the channel
# KevinMarks_ does the whitespace collapse not already happen? I thought that searching the element tree did that
caseorganic joined the channel
# gRegor` kbs: One of the things I do like about it is that the sender doesn't have to have the recipients public key, just their hash (BitMessage ID). It's still not easily readable, but it seems a step in the right direction
# dracos KevinMarks_: oh you’re right, it works across a <br>. Doesn’t appear to across a <p> though.
# KevinMarks_ that seems to be the case. is this a block vs inline thing?
# JonathanNeal dracos: thanks for the change, I’m having the team review it (but it will be brought in)
# dracos KevinMarks_: It’s looking at textContent, so it’s what’s actually there, e.g. http://leveson.sayit.mysociety.org/hearing-13-march-2012/mr-richard-fedorcio?asd##actually.%0a%0aWe+can works because it has two newlines
# JonathanNeal I started working at a University that is actually making really cutting edge stuff. Only they haven’t had a lot of experience working with the larger internet community, so I’m encouraging them to get involved by letting them review the projects. Thanks for your patience.
# KevinMarks_ hm. that seems non-ideal - collapsing whitespace to single spaces and matching would be better
# gRegor` kbs: True, but as we've discussed, the verification out of channel is a pain point there.
# gRegor` Some of those key servers aren't even ssl
# dracos KevinMarks_: yeah
# gRegor` Not that I think a blockchain including public keys is very scalable, either, heh
# KevinMarks_ the tricky part of using the public keys in the blockchain for communication is that you are using the same key to read messages as you do to send money
# KevinMarks_ JonathanNeal: I did wonder why you used the university github and not your own
basal joined the channel
# KevinMarks_ dracos: I think the second # being encoded is someone taking the URL RFC seriously, and a marker for why we might want to switch to #*
# KevinMarks_ !tell adactio your justify every pixel link should be http://www.digital-web.com/articles/joshua_davis/##justify+every+pixel not http://www.digital-web.com/articles/joshua_davis/##justify-every-pixel
# KevinMarks_ weird, Adactio's Twitter card gets the title wrong in https://twitter.com/elroyjetson/status/456937359435517953 DRY failure?
# 
@elroyjetson The idea of Fragmentions is brilliant - http://adactio.com/journal/6751/ /cc @adactio @kevinmarks (twitter.com/_/status/456937359435517953)
# 
gregorlove.com created /User:Gregorlove.com/sandbox-row (+324) "Created page with "<includeonly> <tr> <th scope="row" class="navbox-group" style="padding: 0.25em 1em" (view diff)