#indiewebcamp 2015-05-23

2015-05-23 UTC
#
tantek a+blockquote means a blockquote that is the immediate following sibling of an a
#
KartikPrabhu ah judas priest!
#
KevinMarks_ so the a always wins?
#
tantek bad markup = unpredictable results
#
KartikPrabhu oh poo! My webmention reciever thinks that KevinMarks_ post does not link to mine!! weird!
#
KartikPrabhu debugging time
#
KevinMarks_ so how should I markup that if I want the whole blockquote clickable? wrapper div?
#
Loqi gives KevinMarks_ the whole blockquote clickable
#
@alexskunz @chuq @mrussellphoto @rwongphoto pull in all the social comments via IndieWeb and show them on your site. On my to-do list. (twitter.com/_/status/601901377736978432)
craftycorvid joined the channel
#
KevinMarks_ reads spec - phrasing content is the problem?
#
KevinMarks_ hm, twitter uses <div href="" to make it linkable
#
KevinMarks_ that seems wrong in a different way
#
tantek no that doesn't work either
#
tantek you're likely seeing odd JS effects
#
tantek to make the whole blockquote clickable you can do this
#
tantek <blockquote style="padding:0"><a href="" style="display:block;margin:0;border:0">…</a></blockquote>
#
tantek with styling extracted out to style rules of course
#
KevinMarks_ oh, that's weird. OK
#
gRegorLove I was noticing that in Chrome dev tools when I embed a blockquote in an a href, it actually reverses it, a inside blockquote.
#
KevinMarks_ is the cite the same problem?
#
KartikPrabhu gRegorLove: yeah. browsers handle "bad markup" differently
frzn joined the channel
#
KevinMarks_ realises why no-one uses semantic elements
#
KartikPrabhu semantics is ok, but this rigid rule of what goes inside what is annoying
#
KevinMarks_ especially as the rule is not expressed anywhere clearly
callovarne joined the channel
#
KevinMarks_ I just read http://www.w3.org/html/wg/drafts/html/master/dom.html#paragraphs and it wasn't any clearer
#
gRegorLove I think it's generally that inline elements shouldn't appear inside block-level elements.
#
gRegorLove Er, strike that, reverse it.
#
tantek ;)
#
KartikPrabhu gRegorLove: the other way round
#
KartikPrabhu :P
#
gRegorLove It was fine in my brain, ok!
#
KevinMarks_ is cite block level?
#
gRegorLove Fingers just betrayed me.
#
tantek no cite is inline
#
KevinMarks_ ah, it's 'normal' not 'none'
#
KevinMarks_ should I be putting h-cite on that?
#
tantek no text-decoration:none is what you want
#
KevinMarks_ yes, I did thta on the a; i was trying to stop the cite being italic too
#
tantek font-style:normal
#
KevinMarks_ yep
#
KevinMarks_ how's that?
#
KevinMarks_ now wondering if h-cite etc makes more sense that auto-mention
#
tantek auto-mention?
#
KevinMarks_ by analogy with class="auto_link"
#
KevinMarks_ but this actually creating an h-cite isn't it
#
tantek a quote
#
tantek really it's just another auto-embed - of the text
#
tantek rather than an img
#
KevinMarks_ yes, that was what i was thinking
#
tantek KevinMarks_: re: " will display the tweet inline in your tweet," - not quite
interactivist joined the channel
#
tantek will display it *after* your tweet, nested
#
KevinMarks_ ah right
#
KevinMarks_ yes, inline was my wished for thing
#
tantek we can do it inline in cassis
#
KevinMarks_ yes
#
KevinMarks_ but in effect it's embedding a reply context
#
tantek except it's not, it's a mention context
#
KevinMarks_ so, we shouldn't d any u-in-reply-to stuff as that is unknown
#
KevinMarks_ but it is an h-cite, right?
#
tantek it can be marked up as one yes, however it's not yet clear what the value is to consuming code
#
tantek KevinMarks_: the sad thing about the Twitter display of the tweet you mentioned, is that it removes the information about WHERE you mentioned the tweet permalink in your tweet
#
KevinMarks_ yes, exactly
#
tantek e.g. you say "We should debate this on TWiG next week. See also …" - so where in that text did you actualy link to Jarvis's tweet?
#
tantek it makes less sense the way twitter displays it :/
#
KartikPrabhu i don't know why my webmention receiver is breaking on KevinMarks_'s post http://known.kevinmarks.com/2015/in-the-case-of-digital-content-the-artifact-once-created
#
KevinMarks_ hm.
#
KartikPrabhu it says source does not have link to target, but it of course does!
#
KevinMarks_ the trickiness of h-cite si that it isn't quite right; p-content isn't what this is, it's more of a p-summary
#
KevinMarks_ and the p-name is unknown
#
KevinMarks_ and the author is approximate (
#
KevinMarks_ because it's just the domain
#
KevinMarks_ so may not be worth doing
#
KevinMarks_ eg http://www.unmung.com/?html=%3Cblockquote+class%3D%22auto-mention+h-cite%22%3E%3Ca+class%3D%22auto-link%22+href%3D%22https%3A%2F%2Fkartikprabhu.com%2Farticle%2Fmarginalia%23In%2520the%2520case%2520of%2520digital%2520content%2C%2520the%2520artifact%2C%2520once%2520created%2520and%2520published%2C%2520is%2520not%2520static%22%3E%3Ccite+class%3D%22p-author%22%3Ekartikprabhu.com%3C%2Fcite%3E%3Cp+class%3D%22p-summary%22%3EIn+
#
KevinMarks_ name is going wrong
#
@LibertyBundle RT : Hey, ! Come check out our awesome, supportive group! https://twitter.com/W_Angels_Wings/status/601900442428116992 W_Angels_Wings: Hey, #Indieauth… https://twitter.com/W_Angels_Wings/status/589113223216963584/photo/1 (twitter.com/_/status/601914822561800193)
#
KartikPrabhu KevinMarks_ conveniently BillSeitz reply showed up as marginalia :P https://twitter.com/BillSeitz/status/601911984834019328
#
@BillSeitz .@kevinmarks @kartik_prabhu MoonReader EBook app makes it easy to attach notes at any point, or highlight. (And email/share both.) (twitter.com/_/status/601911984834019328)
#
KevinMarks_ I can send a ebmention from this post too
#
KartikPrabhu I want to see why my webmention receiver is not working with your original post. It does work with all the same bridgy backfeeds!!
#
KevinMarks_ according to ben, the webmention is causing a curl error, so probably is not being sent
#
KevinMarks_ I can resend with indiewebify.me ?
KartikPrabhu joined the channel
#
KartikPrabhu but I am using my form on my site and it says "source does not link to target" which is.... oh!! of course my form only looks for the post URL not a fragmention... stupid me!
#
KartikPrabhu smacks his head
#
KartikPrabhu KevinMarks_ can I send a webmention to all links on your post page
#
KevinMarks_ sure, but i'm juste editing it, give me second
#
KartikPrabhu I am sure it will only show up as a mention
#
KevinMarks_ wondering if we should put u-highlight on it if we mean it as a highlight?
#
KartikPrabhu I am thinking a u-bookmark-of is a highlight
#
KartikPrabhu I wonder if my webmention processor looks for bookmark-of yet?
#
KartikPrabhu hmm mf2util does not know about bookmark-of so might need more work
#
KartikPrabhu kylewm ^
benwerd joined the channel
#
KartikPrabhu what is bookmark?
#
kylewm KartikPrabhu: would gladly accept that change
#
Loqi A bookmark (or linkblog) is a post that is primarily comprised of a URL, often title text from that URL, sometimes optional text describing, tagging, or quoting from its contents https://indiewebcamp.com/bookmark
#
kylewm I've been thinking I should probably re-do mf2util as more generic utilities rather than being aware of specific properties, more like barnaby's utils package
#
KartikPrabhu hmm
#
tantek KartikPrabhu: that's an interesting connection - highlight being a form of bookmark
#
KartikPrabhu highlights are strange to classify
#
tantek in the literal sense - it certainly is a bookmark of a sort
#
tantek one of the ways you put a mark in a book!
#
KartikPrabhu highlights = fragmentioned bookmark or = fragmentioned quote
#
kylewm lol, marginalia is another way to mark a book :P
#
tantek true
#
KartikPrabhu tantek: yes. The reason I highlight or underline in my case is to mark things I want to come back to in a book
#
KartikPrabhu hence fragmentioned bookmarks = highlights for me
#
tantek yes - which is in meaning the same as the act of bookmarking
#
KartikPrabhu yup
#
tantek highlights are just a method of presenting what is a semantic bookmark of a range of text
#
tantek whereas quoting is something else
#
KartikPrabhu agrees
#
tantek what is a highlight?
#
Loqi It looks like we don't have a page for "highlight" yet. Would you like to create it? https://indiewebcamp.com/wiki/index.php?action=edit&title=highlight&summary=prompted+by+tantek+https%3A%2F%2Findiewebcamp.com%2Firc%2F2015-05-22%2Fline%2F1432344742842
#
KartikPrabhu so I'll go with the assumption that a fragmentioned bookmark is meant as a highlight
#
Vendan wouldn't marginalia be basically a fragmention with a note?
#
tantek A highlight is a /bookmark of a /fragmention
#
loqi.me created /highlight (+69) "prompted by tantek https://indiewebcamp.com/irc/2015-05-22/line/1432344742842 and dfn added by tantek" (view diff)
#
Loqi ok
#
KartikPrabhu Vendan: that is what it is :)
#
tantek Vendan - not quite
#
Vendan so like a note of a highlight
#
tantek no - a reply
#
tantek marginalia is basically a reply (comment) where the in-reply-to URL is a fragmention
#
KartikPrabhu Vendan: see https://kartikprabhu.com/article/marginalia for examples :)
#
Vendan yeah, and what's a reply? a note with a in-reply-to url
#
KartikPrabhu yes
#
tantek vendan, what is a reply
#
Loqi A comment is a kind of post that is in reply to some other post, that makes little or no sense without reading or at least knowing the context of the source post https://indiewebcamp.com/reply
#
Vendan it's just interesting how all the different stuff ties together
#
tantek it is likely a note but may be an article
#
tantek Vendan, e.g. on medium, they added responses which are *article* replies to articles
#
KartikPrabhu that is also something I should do. If an article is used to reply to a fragmention then marginalia should deal with it nicely instead of showing the whole post in margins :P
#
tantek right
#
KartikPrabhu Medium's "recommend" is just a like-of me thinks
#
tantek yes
#
KartikPrabhu all in all I think Medium.com is just re-naming stuff and appropriating it to themselves
#
KartikPrabhu <shrug>
#
KartikPrabhu I wonder how granular indieweb highlights should be
#
KartikPrabhu certainly not <p> level
#
tantek.com edited /Medium (+31) "equiv of like" (view diff)
#
KartikPrabhu and certainly not letter level
#
tantek KartikPrabhu: can you start by documenting Medium's granularity of highlights?
#
tantek e.g. grab screenshots, or hotlink to theirs
#
KartikPrabhu ugh then I'll have to use Medium :P
#
tantek from /highlight#Silo_Examples
#
KartikPrabhu but yes should try it at least
#
tantek you may not need to use it
#
KartikPrabhu Yeah I'll experiment on KevinMarks_'s Medium posts :P
#
tantek KartikPrabhu: hotlink images from ev's post e.g. https://d262ilb51hltx0.cloudfront.net/max/800/1*fdWsCNAM8wJHPXqp88dI3A.png
#
tantek from https://medium.com/the-story/medium-is-not-a-publishing-tool-4c3c63fa41d2#b387-4ecc289e3d0d
#
KartikPrabhu tantek: I do know that when you highlight you can select as much/as little as you want. But If 2 people highlight the same thing but differ by one letter in the end (maybe selection error) does Medium take them as the same phrase highlighted by 2 people?
#
tantek KartikPrabhu: good question - see if you can find examples in KevinMarks's medium posts
#
KartikPrabhu yeah
#
KartikPrabhu it would be annoying for presentation if every different letter counts as a separate highlight
#
KartikPrabhu I don't even know how to markup overlapping highlights
#
KevinMarks_ could also look at rapgenius
#
tantek.com created /recommend (+119) "stub with dfn" (view diff)
#
tantek KartikPrabhu: " how to markup overlapping highlights" - you may have to experiment with Medium's UI to see how they do that.
#
tantek I have a solution in mind, but it's not elegant, and I don't want to taint your research.
#
tantek e.g. you may have to use multiple accounts to create overlapping highlights on a Medium article and see what happens
#
KevinMarks_ they seem to reply on over a lot
#
KevinMarks_ s/over/hover/
#
Loqi KevinMarks_ meant to say: they seem to reply on hover a lot
#
KartikPrabhu sure. will investigate.
craftycorvid joined the channel
#
KevinMarks_ s/reply/rely/
#
KartikPrabhu wow! it is really hard to see how to reveal something when a child of it is linked to
#
KartikPrabhu use case: if someone links to a particular marginalia on a page, the page should automatically unhide it
#
KevinMarks_ looks like the outer span wins in these overlapping highlights
#
KartikPrabhu KevinMarks_ what if there is only an intersection :P
#
KartikPrabhu oh my god who thought of naming the opposite of focus() method as blur() ! https://developer.mozilla.org/en-US/docs/Web/API/HTMLElement/blur
#
KartikPrabhu sheesh
#
kylewm hahaha
#
KevinMarks_ hm, so medium can write responses to anything
#
KevinMarks_ https://medium.com/the-story/write-about-anything-on-the-web-with-the-medium-bookmarklet-86661f9002fe
KartikPrabhu joined the channel
#
KartikPrabhu oh if you link an element with a fragment the browser activates the focus on that element! interesting
#
KartikPrabhu nope ignore that
#
KevinMarks_ https://html.spec.whatwg.org/multipage/browsers.html#scroll-to-the-fragment-identifier
#
KartikPrabhu KevinMarks_ yup :target is the correct selector
#
KevinMarks_ it does say that it should focus
#
KartikPrabhu I am wondering how to show a particular marginalia when it has been linked from the URL via a fragment
#
KartikPrabhu KevinMarks_ at in Firefox the :focus selector in CSS does not trigger when you use a fragment
#
KartikPrabhu I think that is supposed to be user-focused
#
KevinMarks_ the definition of 'focusable element' is tricky
#
KartikPrabhu yes. that is what my minor research tells me too
#
KevinMarks_ https://html.spec.whatwg.org/multipage/interaction.html#focusable-area
#
KartikPrabhu I really wan to avoid overloading marginalia.js to do all the heavy lifting
#
KartikPrabhu some ids and classes and CSS should be able to do the trick
#
KevinMarks_ an iframe is one, so you could inject one and force focus to that
#
ben_thatmustbeme i think of focus as anything you can usually tab to
#
ben_thatmustbeme s/focus/focusable/
#
Loqi ben_thatmustbeme meant to say: i think of focusable as anything you can usually tab to
#
KartikPrabhu ben_thatmustbeme: yes. one can set tabindex using JS but I want to avoid too much JS reliance
#
KevinMarks_ whetehr you can tab to links or not is complicated
#
ben_thatmustbeme scrolls back to get the full context of exactly what you are trying to do
#
KartikPrabhu here is what I am thinking. each marginalia group gets an id and the "activation button" simply links to it. The using the CSS :target selector one can show/hide the marginalia group
#
KartikPrabhu This much is easy to do. But if the fragment actually links to a particular marginalia comment, I still want to make the group visible with scroll to the linked one
#
KartikPrabhu make sense?
#
KevinMarks_ details/summary is a way too
#
KartikPrabhu yeah but they don't do this UI automatically
#
ben_thatmustbeme trys to picture this in his head as he has not followed marginalia in some time
#
KartikPrabhu I'll put up a prototype online soon so people can see what I mean
#
KartikPrabhu now just brainstorming/braindumping
#
KevinMarks_ hm, the way details/sumamry works doesn't interatc well with id's either
#
KartikPrabhu yeah
#
KartikPrabhu now if only I can select any <a> that links to current fragment from the URL I can make a prototype...
callovarne joined the channel
#
KartikPrabhu hmm seems like that does not exist ^ :(
craftycorvid and wolftune joined the channel
#
KartikPrabhu boo! this is hard. I guess I'll write-up a bunch of thoughts on the "if it worked like magic scenario" and hopefully clever people with chime-in
callovarne joined the channel
#
GWG sneezes
#
ben_thatmustbeme coughs
#
GWG ben_thatmustbeme: I was just talking about you
#
Vendan well, glennjones' new microformats test runner is pretty slick
#
GWG Vendan: Oh?
#
Vendan http://testrunner-47055.onmodulus.net/
#
Vendan he's still cleaning it up
#
Vendan in most cases, there's about 50% chance the test is out of date if the parser is failing
#
Vendan but the interface is just slick
#
kylewm I really love glennjones's style
KartikPrabhu joined the channel
#
GWG Am I reading that most parsers seem to be failing 66% of tests?
KevinMarks__ joined the channel
#
Vendan yeah
gRegorLove joined the channel
#
Vendan a lot of it is out of date tests, really
dariusdunlap, KevinMarks and KevinMarks__ joined the channel
#
KevinMarks__ Ah, I broke all the tests by adding rel-urls?
#
KevinMarks_ should I send a pull request that adds in a rel-urls key to allt he json?
#
KevinMarks_ http://microformats.org/wiki/microformats2-parsing-brainstorming#more_information_for_rel-based_formats
#
KevinMarks_ also, should I edit http://microformats.org/wiki/microformats2-parsing to include rel-urls now?
#
KevinMarks_ do we have consensus enough to change the parsing rules?
loic_m, glennjones, csarven and endi joined the channel
#
@dsearls @ilparone Some of us aren't giving up. e.g. @indiewebcamp @indie @vrm @berkmancenter @stanfordcis ... among many others. (twitter.com/_/status/602001676174626816)
wolftune and j12t joined the channel
#
KevinMarks_ autolink test tool: http://www.unmung.com/?rawtext=https%3A%2F%2Fkartikprabhu.com%2Farticle%2Fmarginalia%23In%2520the%2520case%2520of%2520digital%2520content%2C%2520the%2520artifact%2C%2520once%2520created%2520and%2520published%2C%2520is%2520not%2520static.%0D%0Ahttp%3A%2F%2Fwww.kevinmarks.com%2Fjeffquote.png%0D%0Ahttp%3A%2F%2Fwww.kevinmarks.com%2Fmentionquote.html%23%23we%2Bpotentially%2Bhave%2Ba%2Bquote%2Bfrom%2Bthe%2Bsou
#
KevinMarks_ today in evil things to do with URLs, I'm your host
j12t, eschnou, LauraJ, wont_on, tilgovi, KevinMarks, KartikPrabhu, endi, simoncook, interactivist, ttepasse, squeakytoy, sivoais, frzn, mlncn, tantek and glennjones joined the channel
#
tantek analyzing snarfed's checkins, he's got quite a few interesting variations
#
GWG He hand codes them, doesn't he?
#
tantek GWG, possibly, still analyzing
#
tantek !tell snarfed your checkin https://snarfed.org/2014-11-13_11514 has an error in its attempt to use class="u-featured" as the same element has class="shadow". Multiple class attributes disallowed in HTML. Class names must be combined: class="u-featured shadow"
#
Loqi Ok, I'll tell them that when I see them next
#
ben_thatmustbeme hmm looking at the php-mf2 parser it claims errors because the labal includes line breaks (and doesn't remove extra spaces). are those requirements of microformats parsing?
#
tantek the what inculdes line breaks?
#
ben_thatmustbeme http://testrunner-47055.onmodulus.net/run/php-mf2/all/
#
ben_thatmustbeme h-adr geo
#
ben_thatmustbeme is the one i'm currently looking at
eschnou joined the channel
#
tantek nothing in the HTML should cause errors in the php-mf2
#
tantek maybe not the result you're expecting (e.g. bad HTML) but nothing should cause errors in the parser
#
ben_thatmustbeme okay now that i have a free hand again.. its expecting "3 Charlotte Road, City of London, EC2A 3PE, UK", but the parser returns "3 Charlotte Road, \n City of London, \n EC2A 3PE, \n UK"
mlncn joined the channel
#
ben_thatmustbeme because the html souce had new lines between
#
ben_thatmustbeme so i'm guessing that should be stripped with the html
#
glennjones ben_thatmustbeme I always collapse white-space/returns as the returns and spacing left as no relation to text once you remove tags
#
ben_thatmustbeme but does the rules for mf parsing state that? or is it considered implied by stripping out html?
#
ben_thatmustbeme just trying to be clear. i'm going to try to update php-mf2 to fix that either way
#
@LukasRosenstock Check out my two-part blog series on phone numbers and the #IndieWeb: - Part 1: http://lukasrosenstock.net/post/59674472?utm_content=buffer456f9&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer - Part 2: http://lukasrosenstock.net/post/60002190?utm_content=bufferd7257&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer (twitter.com/_/status/602115538207252480)
#
ben_thatmustbeme also, need to get the ruby parser up to date and tested in here
#
ben_thatmustbeme s/here/there/e
#
Loqi ben_thatmustbeme meant to say: also, need to get the ruby parser up to date and tested in there
#
glennjones the parsing page refers to returning textContent ie the DOM Level 3 specification method which does get rid of whitespace
#
ben_thatmustbeme (working on stuff difficult with a baby and a dog on you)
#
ben_thatmustbeme ahh, okay
#
Vendan wooo, thanks for merging that baseurl update, glennjones
#
Vendan I'm up to 65.1% now :D
#
Vendan at least on the v2 stuff
#
glennjones vendan I am still working my way through the test clean up errors, but your parsers look good with v2 stuff
#
Vendan thanks
#
ben_thatmustbeme hmm, there are quite a few issues filed on php-mf2.
#
Vendan still have issues with datetime and value-title stuff
#
glennjones ben_thatmustbeme: All the parsers are doing well against the common stuff, it just the edge case that are been flaged, and some errors in the tests
raretrack and j12t joined the channel
#
ben_thatmustbeme i'm talking about https://github.com/indieweb/php-mf2/issues
#
Vendan I do have to say, that's a very very nice interface to the test runner
#
tantek.com edited /checkin (+1626) "/* IndieWeb Examples */ add Ryan Barrett, with analysis of different kinds of checkins (with photo(s), person(s))" (view diff)
#
glennjones Thanks I hope we can use the runner to nail a few small difference in the parsers. ie help us take our code forward
#
glennjones sorry guys got have guests who just turned up, but we should talk more about this
#
raretrack.uk edited /IRC_People (+52) "/* Nicknames */ added me (raretrack)" (view diff)
#
Vendan hrm, looking at the DOM level 3, it says no whitespace normalization is allowed on textContent
#
tantek.com edited /checkin (+486) "brainstorming, plain text design, cite snarfed's example" (view diff)
#
tantek welcome raretrack!
#
ben_thatmustbeme hmmm... interesting. Vendan, I see that too
#
Vendan and I just checked it out in a jsfiddle, textContent does contain newlines and spaces and junk
#
raretrack thanks tantek
#
Vendan so this seems like a case where the test is wrong, but honestly, looks better then the right result
#
ben_thatmustbeme Vendan, either that should be a part of the mf2 spec then or it should be just assumed the consumer can do whitespace normalization
#
Vendan yeah, that's my thoughts
#
ben_thatmustbeme actually, it could have been in a pre tag, which would make it important, no?
#
Vendan with the current specs, it could be done using class="value" for all the parts, and <data class="value" value=", "> </data> for the comma portions
#
Vendan or really, just wrap the existing commas in <span class="value">
#
tantek ben_thatmustbeme: it's deliberate that it's supposed to contain newlines and spaces etc.
#
tantek the use-case is whitespace formatting in notes
#
Vendan so, test is wrong
#
tantek http://indiewebcamp.com/note#Whitespace
#
Vendan there, fixed, and ouch, down to 49% of tests
eschnou and GWG joined the channel
#
@msgbi .@guenterhack @TeraEuro @Lambo folge mal @kevinmarks #indieweb (twitter.com/_/status/602130680982700032)
#
@jansauer RT @btconf: Group photo of the #IndieWebCamp Germany during #btconf in Düsseldorf. https://twitter.com/btconf/status/600591662813941760/photo/1 (twitter.com/_/status/602131184899006464)
KartikPrabhu, tantek, KevinMarks__, dietrich and wolftune joined the channel
#
benborges kylewm, remember how i changed recursively all the Uploads folder to www-data user and www-data group ?
#
kylewm yes
#
benborges now i lost ALL my images
#
benborges http://1-m.cc/profile/benborges
#
benborges this is weird right ?
#
benborges not sure how i'm going to go back, forgot to backup before touching that
#
benborges i thought it would not hurt anything, but it seems it did, i just didn't noticed right away
#
benborges also lost the dropdown menu, not sure if connected
#
kylewm how is your uploads folder arranged? Mine looks like Uploads/known.kylewm.com/[0-9a-f]/[0-9a-f]/[0-9a-f]/[0-9a-f]/...
#
benborges exactly like yours then
#
benborges inside Uploads i have my domain 1-m.cc
#
benborges then all these folders too
#
kylewm hmm very frustrating
#
kylewm it is like the php script doesn't have access to that data at all somehow
#
kylewm are you running Ubuntu?
#
aaronpk benborges: it's possible that php-fpm is running under a different user than www-data
#
aaronpk try `ps auxhf | grep fpm` and you should see what user it's running as
#
benborges trying now
#
benborges i'm running debian jessie
#
benborges root 498 0.0 3.5 401052 27560 ? Ss May18 0:33 php-fpm: master process (/etc/php5/fpm/php-fpm.conf)
#
benborges www-data 536 0.0 3.7 414256 28888 ? S May18 0:49 \_ php-fpm: pool www
#
benborges www-data 1697 0.0 3.6 413792 28552 ? S May18 0:42 \_ php-fpm: pool www
#
benborges www-data 29959 0.0 3.6 413300 27972 ? S May22 0:17 \_ php-fpm: pool www
#
benborges root 13366 0.0 0.2 4536 2168 pts/0 S+ 16:41 0:00 \_ grep fpm
#
benborges php master process seem to be in root indeed
#
benborges while the rest is www-data
#
benborges what should i do ?
#
aaronpk hm okay that seems correct then
#
benborges it's running has user = www-data
#
benborges group = www-data
#
benborges in /etc/php5/fpm/pool.d/www.conf
#
kylewm did you double check that the permissions are correct after running chmod -R ?
#
benborges u mean the Upload folder ?
#
kylewm you can run 'ls -lR' on the Uploads folder to check all its subfolders
#
kylewm I'm pretty sure I've accidentally done chmod -r instead of chmod -R before and removed read permissions
#
benborges ls -lR show all folder are owned by www-data and www-data group
#
benborges sometimes i dont see the group
#
benborges just www-data
#
benborges some files are also root and www-data
GWG joined the channel
#
benborges is there a magic command to turn all my known folder www-data ? is that a good idea ?
#
benborges permission i see are : drw-rwSrw- kylewm
nloadholtes joined the channel
#
kylewm huh, I mean yeah you could run chown -R www-data:www-data .
#
benborges this : chown -R www-data:www-data 1-m.cc/
#
benborges this is what i gave to the domain folder inside the Upload folder
#
benborges yesterday
#
kylewm oh, wait maybe the problem is that you don't have x permissions on those folders, doesn't that mean you can't browse the folder?
#
benborges followed by sudo chmod -R 666 1-m.cc/
#
kylewm my folders are like "drwxr-xr-x 3 www-data www-data 4096 May 23 09:32 known.dev"
#
benborges i dunno enough about perms
#
kylewm individual files are like "-rw-r--r-- 1 www-data www-data 105 May 23 09:32 dd8a8eee78a44418025a28820e2c7c1d.data"
#
benborges my individual files are -rw-rw-rw- 1 www-data www-data 112 May 5 21:37 02ca6cb26c2fa04d94ad3d36e5c097b5.data
#
kylewm benborges: I think do this -- http://superuser.com/a/91938
#
kylewm the first two lines
#
kylewm it will set directories to rwxr-xr-x and files to rw-r--r--
#
benborges u means these 2 lines ?
#
benborges chmod 755 $(find /path/to/base/dir -type d)
#
benborges chmod 644 $(find /path/to/base/dir -type f)
#
kylewm yeah, they list three different variations, any of them will work
#
kylewm but obviously change /path/to/base/dir to your Uploads folder
#
benborges and images are back :)
#
benborges http://1-m.cc/profile/benborges
#
benborges i did this one : chmod -R u+rwX,go+rX,go-w /Uploads
#
benborges no my only remaining problem (that started when i added a page)
#
benborges is http://1-m.cc/ dropdown menu don't work, while both have sub pages to show
#
benborges image upload is back too !!
#
kylewm wow, I had no idea about u+rwX... nice to learn something!
wolftune joined the channel
#
benborges i still can't tweet an image, not sure why or where to look why
#
benborges KevinMarks_, was saying uploaded images should be tweeted along with title etc..
#
benborges but i'm still unable to do this
#
benborges it' s like if image post type never get POSSEed
#
benborges http://1-m.cc/2015/image-upload-test-to-twitter-from-known-indieweb
#
benborges even though i POSSED to twitter
#
benborges there is no trace of it
#
benborges and logs tells this
#
benborges PHP message: Known (1-m.cc): INFO - Nothing was posted to Twitter: stdClass::__set_state(array(
#
benborges 'errors' =>
#
benborges array (
#
benborges 0 =>
#
benborges stdClass::__set_state(array(
#
benborges 'code' => 189,
#
benborges 'message' => 'Error creating status.',
#
benborges )),
#
kylewm let's take it to #known
#
benborges indeed
#
kylewm sorry, #knownchat
wolftune, nloadholtes, KevinMarks, KevinMarks__, eschnou, j12t and j12t_ joined the channel
#
brucewang.net created /User:Brucewang.net (+36) "Created page with "(this page is under construction :P)"" (view diff)
LCyrin joined the channel
#
brucewang.net edited /IRC_People (+61) "/* Nicknames */ added __number5__ (sorry didn't want to take top slot but someone else registered number5 on freenode)" (view diff)
#
petermolnar.eu edited /Facebook (+117) "/* Criticism */" (view diff)
petermolnar and eschnou joined the channel
#
petermolnar slightly offtopic: had anyone here used Tox for communication yet?
#
Loqi petermolnar: tantek left you a message on 5/22 at 2:22pm: what are the "explore", "discover", "recommended" & similar tabs that you're referring to? Is this on Medium? Could you make/upload a screenshot to /Medium if so? http://indiewebcamp.com/irc/2015-05-22/line/1432329759893
#
Loqi petermolnar: tantek left you a message on 5/22 at 2:26pm: The general idea is, if you have silo-feature-envy: 1. take explanatory screenshots of the feature, 2. upload to the wiki, 3. document it in a /silo_name#Features section - that way we have something concrete (in time and URL space) to refer to to start mocking up indieweb versions. http://indiewebcamp.com/irc/2015-05-22/line/1432330011135
dns53, Guerillero|BNC and snarfed joined the channel
#
KevinMarks https://github.com/microformats/tests/issues/16#issuecomment-104855221
#
KevinMarks Are people happy if I add rel-urls to the main microformats 2 parsing page?
#
KevinMarks Hm, suppose I should ask in microformats
#
@dkreuz Level 1 of #Indieweb achieved. I encourage my students to follow. http://indiewebify.me (twitter.com/_/status/602191986439880704)
KevinMarks__, craftycorvid, nloadholtes, KevinMarks, KevinMarks___, Guest83752, glennjones, KartikPrabhu, mlncn, tilgovi and interactivist joined the channel
#
GWG I've just rewritten something I've rewritten before and will rewrite again. And the output looks the same.
wolftune, callovarne, craftycorvid, KevinMarks, KevinMarks___, KevinMarks__, gRegorLove and dariusdunlap joined the channel
#
KartikPrabhu baah annoying! StartSSL is "offline for maintenance" on weekend night GMT
#
KartikPrabhu what is https?
#
Loqi HTTPS is an abbreviation for Hypertext Transfer Protocol Secure, a protocol for secure communication, supported by web servers (like Apache & nginx) and browsers https://indiewebcamp.com/HTTPS
KevinMarks, KevinMarks__, callovarne, emmak, mlncn, KartikPrabhu, craftycorvid, cmhobbs, tilgovi, aaronpk and aaronpk_ joined the channel