#indiewebcamp 2013-07-07

2013-07-07 UTC
Guest57324, tantek, bnvk, poppy, b0bg0d, xtof, sandeepshetty, Guest48798, benwerd, erikmaarten and saurik joined the channel
#
@benwerd
@t Envious of the #indieweb hacking time! Also, I really enjoyed working there. Also, this is an automatic #indieweb multi-reply.
earplugs, hr3f and jMCg_ joined the channel
#
benwerd
(No it isn't, Ben.)
bnvk, seyz and xtof joined the channel
#
werd.io
edited /idno (-209) "/* Itches */"
(view diff)
#
@benwerd
@bret @julien51 I'm late saying it, but thank you both for really great #indieweb posts. So glad we're all a part of this community.
#
werd.io
edited /multiple-reply (+47) "/* Indieweb Examples of Multiple Replies */"
(view diff)
xtof, b0bg0d, bnvk, catsup and pfefferle joined the channel
#
Loqi
pfefferle: barnabywalters left you a message on 7/6 at 12:55pm: aaronpk I added markup examples to /comment-presentation and /reply-context — does this adequately explain everything we were talking about earlier? http://indiewebcamp.com/reply-context#Markup http://indiewebcamp.com/comments-presentation#How_to_markup
#
pfefferle
!tell barnabywalters perfect! thanks! and also thanks for pointing me to the <body /> or <html /> as h-entry thing… never thought about that and it makes a lot of things a lot easier… working on my theme to support the comments markup and try to implement the p-in-reply-to into the web mention plugin.
#
Loqi
Ok, I'll tell him that when I see him next
b0bg0d and erikmaarten joined the channel
#
@Auroch
Mozilla Persona - Votre Identification Web Libre https://sebastien.bollingh.eu/2013/07/07/mozilla-persona-solution-didentification-web-open-source/ #identité #indieweb #sécurité
barnabywalters joined the channel
#
Loqi
barnabywalters: aaronpk left you a message 12 hours, 11 minutes ago: here's the date formatting library I was telling you about. would be curious to hear your thoughts on it! https://github.com/indieweb/date-formatter-php
#
Loqi
barnabywalters: pfefferle left you a message 1 hour, 15 minutes ago: perfect! thanks! and also thanks for pointing me to the <body /> or <html /> as h-entry thing… never thought about that and it makes a lot of things a lot easier… working on my theme to support the comments markup and try to implement the p-in-reply-to into the web mention plugin.
#
barnabywalters
aaronpk: looking good!
b0bg0d joined the channel
#
waterpigs.co.uk
edited /comments-presentation (+156) "/* Current Indiewebcamp Practices */ added myself + Taproot"
(view diff)
#
waterpigs.co.uk
edited /comments-presentation (+115) "/* Current Indiewebcamp Practices */ added werd.io"
(view diff)
bnvk, erikmaarten and b0bg0d joined the channel
#
waterpigs.co.uk
edited /export (+345) "stubbed sections"
(view diff)
#
waterpigs.co.uk
edited /export (-5) "de-headered"
(view diff)
erikmaarten, b0bg0d, tantek, xtof, bnvk, gjones, bnvk_ and seyz_ joined the channel
#
@obra
@julien51 - The subtome link on http://blog.superfeedr.com/indieweb/ links to tumblr.
barnabywalters, gjones, earplugs, bnvk, tantek and fmarier joined the channel
#
barnabywalters
aaronpk: okay if I merge my pull request to mf2-shim?
#
barnabywalters
twitter tests are all passing and I added some useful stuff
#
barnabywalters
also I should add it to packagist
#
barnabywalters
okay, I’m going to merge
#
barnabywalters
!tell aaronpk I merged my mf2-shim changes, pushed to https://packagist.org/packages/mf2/shim — let me know if you want to be added as a maintainer
#
Loqi
Ok, I'll tell them that when I see them next
#
waterpigs.co.uk
edited /Taproot (+559) "/* Taproot */ added a bunch of links to libraries I’ve open sourced"
(view diff)
bnvk joined the channel
#
aaronpk
good morning!
#
barnabywalters
morning aaronpk
#
Loqi
aaronpk: barnabywalters left you a message 28 minutes ago: I merged my mf2-shim changes, pushed to https://packagist.org/packages/mf2/shim — let me know if you want to be added as a maintainer
#
barnabywalters
just working on twitter mention shimming
#
aaronpk
barnabywalters: there's something wrong with the mf2 parser reading eschnou's page, see http://aaronparecki.com/replies/2013/04/19/2/indieweb
#
barnabywalters
hm strange I am not seeing any problems
#
barnabywalters
on wpc/php-mf2
#
barnabywalters
you’re all up to date, right?
#
aaronpk
i believe so
#
aaronpk
what exactly is that webignition library doing? I want toget rid of it since that seems to be where all the problems i've had come from
#
barnabywalters
it’s resolving relative URLs
#
barnabywalters
and yes it needs to go, we just need to write a URL resolving function/extract it from the webignition library
#
barnabywalters
aaronpk: actually I think the problem here is that you’re showing PHP notices in a production environment
#
aaronpk
well yea
#
aaronpk
but that shouldn't be throwing an error in the first place
#
tantek
barnabywalters, nah, showing PHP notices in a production environment is just part of http://indiewebcamp.com/selfdogfood#testing_your_code_in_production
bnvk joined the channel
#
barnabywalters
tantek: it’s a HUGE security risk and a silly thing to do
#
tantek
and with that, off to get a late lunch at food carts
#
aaronpk
I don't show php errors on actual production stuff, but I don't mind as much here
#
barnabywalters
aaronpk: the easiest quick fix would just be to use @$parser->parse();
#
tantek
barnabywalters - perhaps you can add a section on advice / how-to NOT show PHP notices when selfdogfooding in a new section here: http://indiewebcamp.com/selfdogfood
#
barnabywalters
muting any notices from the parser
#
barnabywalters
tantek: will do
#
tantek
thanks!
#
tantek
human needs food, badly.
#
aaronpk
silly humans
#
barnabywalters
this human has had slightly too much food today
bnvk joined the channel
#
aaronpk
hmm maybe I can just hide notices on my site but leave warnings/errors
#
barnabywalters
aaronpk: another approach which I take is to only show debugging information when I’m logged in
#
barnabywalters
aaronpk: easiest thing is just to mute warnings/errors/notices for the ::parse() call using the @ operator
#
aaronpk
that's a good idea. turning on display errors only when you're logged in?
bnvk_ joined the channel
#
aaronpk
I don't really want to mute errors on it, I'd rather know when something is wrong
#
barnabywalters
aaronpk: do you not log stuff anywhere?
#
barnabywalters
one thing on my todo list is to build a really nice log viewer to ease debugging
#
@osiziqol
Google Reader is dead. The answer is not another SPOF, but integrated indie aggregation. Crazy? Join us: #indiewebcamp (ttk.me t4Qo1) #dead
#
aaronpk
I do, but I also don't want to only see it when I think of looking at the log files :)
#
aaronpk
maybe I should just set up errbit for my site, heh
#
aaronpk
you've seen that, yea? probably exactly what you're looking for
#
barnabywalters
looking now
bnvk__ joined the channel
#
barnabywalters
mm, looks nice. I wonder which would be more straightforward, writing a log viewer or setting up ruby, rails and mongo ;)
#
aaronpk
i'd be happy to let you use my install of it when I set it up :)
#
aaronpk
it's also pretty easy to push to heroku
#
barnabywalters
ah, that’s handy
#
aaronpk
weird, I also don't remember it requiring mongodb
#
waterpigs.co.uk
edited /selfdogfood (+911) "/* testing your code in production */ added notes about debugging notices in production, some PHP code samples"
(view diff)
#
barnabywalters
goodnight all
#
Loqi
night night
b0bg0d joined the channel
#
aaronpk
!tell barnabywalters thx for the idea of only showing errors if you're logged in, just implemented that in like 3 minutes
#
Loqi
Ok, I'll tell him that when I see him next
#
@RyanGarner
RT @Cisco_Mobile: Here's how Facebook plans to use Instagram to track people http://techcrunch.com/2013/07/07/instagram-is-listening/ #privacy #ownyourdata
#
@t
at #WDS2013 @ProBlogger Academy workshop, listening to @darrenrowse speak to a room packed with #indieweb potentials. (ttk.me t4Qt1)
tantek joined the channel
#
tantek
greetings indiewebcamp - been out for a bit attending WDS2013
#
tantek
will go read the logs for the past two days. any particular hot spots?
#
pdurbin
could use a highlight reel ;)
#
pdurbin
it's great stuff. I just can't keep up
#
tantek.com
edited /export (+89) "linky linky - see specific silo pages for how to export from each"
(view diff)
#
tantek.com
edited /comments-presentation (+6) "/* Current Indiewebcamp Practices */ Werdmuller"
(view diff)
#
tantek
starts with Recent Changes
#
aaronpk
another idno install!
#
tantek
one of us. one of us. :)
#
tantek
ooh a third indieweb mult-reply implementation. very nice benwerd.
#
aaronpk
oh? where?
#
aaronpk
hm, my server does not recognize s.ly's SSL cert authority
#
tantek
aaronpk, looks like idno is now tied with p3k for # of users
#
tantek
editing projects...
#
aaronpk
I suspect it will surpass p3k very shortly :)
#
tantek
it's a tradeoff though - with each new user, user-support takes a greater % of your time than feature development
#
aaronpk
indeed
#
aaronpk
that's why p3k is sticking at 2 for a while :)
#
tantek.com
created /User:Werd.io (+69) "stub with an h-card"
(view diff)
#
aaronpk
hmm updating my system's CA bundle still doesn't recognize s.ly
#
aaronpk
unfortunately this means I won't be able to send webmentions to him or pull out h-entry info
#
aaronpk
this brings up an interesting technical use case for indieweb "friending"
#
aaronpk
at the point that I "friend" s.ly, I can download the appropriate certs for his domain
earplugs and gjones joined the channel
#
tantek
aaronpk - some more material for the business models page - freshly blogged from #WDS2013: http://tantek.com/2013/188/t4/wds2013-problogger-challenge-monetizing-list-11
#
aaronpk
awesome!
earplugs joined the channel
#
aaronparecki.com
edited /p3k (+1181) "itching: smaller/de-emphasized articles"
(view diff)
#
glennjones.net
edited /2013/UK (-1) "Correction"
(view diff)
sandeepshetty joined the channel
#
tantek.com
edited /projects (+161) "/* experimental */ bump idno up to just below p3k, since now it too has 2 indie users."
(view diff)
#
sandeepshetty
The idea of displaying p-summary by default/first for comments has been bothering me for a while.. (http://indiewebcamp.com/comments-presentation#How_to_display)
#
sandeepshetty
Seems like we do it because it is plain text as opposed to e-content which is HTML... is there another reason for it?
#
sandeepshetty
the reason it bothers me is because if it were local comment, I would show the whole comments so treating remote comments differently doesn't feel right
#
tantek
sandeepshetty - another reason is when people write REALLY long comments and it might screw up your comment or reply-context display
#
tantek
aaronpk had a recent example of this
#
sandeepshetty
that should be an explicit rule/guideline that applies to both remote/local comments: truncate any comment greater than x and show a read more link,
#
tantek.com
edited /FAQ (+849) "sort questions by likelihood of relevance, fix/updated everyone going to want to run their own website question (of course they are, like personal computers, cell phone numbers etc.), add What is new q and a"
(view diff)
#
aaronpk
(if you choose to because it would make your UI better)
#
aaronpk
yea I replied to a really long note from barnaby and currently I don't truncate it and it kind of messed up my UI
b0bg0d joined the channel
#
sandeepshetty
and trusting p-summary to be "short" is also not a good idea so that rule could be added explicitly... and shouldn't be a reason to show p-summary by default
#
pdurbin
aaronpk: 106 words... 689 characters... not *that* long :)
#
aaronpk
yea, but long enough to screw up my UI
#
sandeepshetty
it feels like a hack to get a plain text version of the comment (which works for what most indiewebers call "notes" but not for "articles")
#
sandeepshetty
and that too because php-mf2 was stripping tags from p-* and giving plain text (which is another issue I want to take up after this)
#
aaronpk
actually that brings up an interesting point. should we even include a p-name property for plaintext notes that don't have a title?
#
aaronpk
like if the p-name is the same as the p-content, do we need p-name?
#
tantek
sandeepshetty - it's not about "trusting" in any absolute form
#
tantek
p-summary is *likely* to be shorter, so you can try using it if the e-content is "too long" (for your UI)
#
tantek
and if even the p-summary is too long, then abbreviate
#
tantek
thought I wrote this up on the comments-presentation page
#
aaronpk
I guess if I grab the e-content and think it's too long, I might as well just truncate it myself. why would I need the p-summary?
#
aaronpk
I could see using p-summary for articles though
#
sandeepshetty
aaronpk: I tried my atom feed on feedly and I have p-name = p-summary = p-content and it looks like sh** :)
#
tantek
aaronpk - because if there is a p-summary and it is shorter, it may be more meaningfully truncated than you might be able to automatically truncate
#
aaronpk
heh yea, notes don't really have a name. for syndicating to facebook I made up a title like "Note posted on July 6 2013" or something
#
tantek
p-summary is likely explicitly crafted by the author
#
tantek
with even the expectation that that's what syndicated copies might use
#
sandeepshetty
aaronpk: Thats similar to what I had in mind as a fix
#
aaronpk
tantek: ok so that implies the author is doing intelligent summarization, so that should be documented as a recommendation too
#
tantek
on the publishing side yes
#
sandeepshetty
IIRC, nobody dies this
#
sandeepshetty
(craft a summary)
#
aaronpk
I certainly don't for notes, but I do for articles
#
sandeepshetty
(on the indieweb)
#
aaronpk
I either automatically create the summary from the first 2 paragraphs plus first photo, or I manually type up the summary for articles
#
sandeepshetty
the summary concept is also dying out on drupal, wordpress, etc..
#
tantek
aaronpk - I've got some (pseudo?)code crafted to put the first whole sentence or two into a p-summary
#
sandeepshetty
especially since most blogs show the entire content on the list pages
#
tantek
so that notes can be even more precisely truncated than what people's truncation code does
#
tantek
sandeepshetty - who cares about what "is also dying out on drupal, wordpress" ?
#
aaronpk
sandeepshetty: I don't think that's necessarily true. especially with inline previews on twitter and facebook, summaries of articles is even more important now
#
tantek
seriously - none of those are bastions of innovation
#
tantek
so looking at what's dying out there has very low value
#
tantek
there's almost no vision in drupal or wordpress - too much is focused on commercial CMS use-cases
#
tantek
not indieweb
#
tantek
sandeepshetty - if you don't see value in explicit summary markup - don't bother with it
#
tantek
others do and will try it out
#
tantek
if you find that by syndicating their p-summary content you see better results, you might consider it too
#
tantek
I think that's true about indieweb stuff in general
#
tantek
if you don't see any value in something, then just ignore it and work on something else
#
tantek
no need to tell everyone not to do it
#
tantek
or make it sound worthless in the absolute
#
tantek
it may just not make sense for your personal site - and that's totally ok!
#
tantek
we all have different itches and scratches and that's part of what makes this community awesome
#
aaronpk
last 5 lines -- great for the wiki
#
tantek
as we organically cluster and implement similar things, we can start making stronger statements about community support of those things
#
tantek
aaronpk - where should that go - /principles?
#
aaronpk
yea totally
#
sandeepshetty
the problem is telling everyone to do the opposite (explicitly) - shouldn't it apply both ways?
#
sandeepshetty
and I'm not saying don't have a p-summary.. just saying i don't like the "rule" of showing p-summary by defualt
#
sandeepshetty
(for comments)
#
sandeepshetty
let the implemenation/person decide
#
sandeepshetty
and I'm coming from the perspective that a comment is a comment is a comment.. remote or local..
#
sandeepshetty
same rules should apply for both
#
tantek
sandeepshetty - that's certainly true about a comment *appearing* as a comment
#
tantek
yet the real world case that aaronpk pointed out cannot be ignored
#
tantek
sometimes comments get too long
#
tantek
every major social network auto-ellipses long comments with view more… links
#
tantek
so this is the new convention
#
tantek
long comments get ellipsed
#
tantek
p-summary just helps give the author some control over how that happens
#
aaronpk
I agree with sandeepshetty on this one though, it shouldn't be a "rule" to show the p-summary, it should be recommended that you truncate text to fit your UI, and suggested that the p-summary may have a better truncated version
#
tantek
author control = a good thing
#
tantek
aaronpk - really? is it that much work to look for the p-summary?
#
tantek
or are you saying we should prefer full content first?
#
tantek.com
edited /comments-presentation (+417) "start FAQ with Why use p-summary"
(view diff)
#
aaronpk
it is a bit of extra work to first check p-summary if it's not something you care about
#
tantek
aha ok
#
tantek
that makes sense
#
tantek
ONLY IF you feel a need to truncate should you check the p-summary
#
tantek
is that what you're saying?
#
tantek
otherwise just always use the e-content in full?
#
aaronpk
that would work
#
tantek
ok that I agree with
#
tantek
preferring full content is better for the indieweb
#
sandeepshetty
and if p-summary is also long.. truncate that as well
#
aaronpk
right, truncate to whatever length you want based on your design, but take into account the fact that the p-summary may have a shorter/better version for you to start with if it's present
#
tantek
right
#
tantek
or p-name even
#
aaronpk
ok so p-name... what do you think about my comment on that earlier?
#
aaronpk
should the p-name be present if it's the exact same content as the p-content?
#
sandeepshetty
aaronpk: for context like feed readers I would rather decide on what shows up as the title than have every feed reader decide
#
sandeepshetty
so I would make up a p-name like you are doing already
#
tantek
aaronpk explicit p-name is necessary unless you want the default microformats2 implied p-name
#
tantek
every microformats2 object has a p-name after being parsed
#
tantek
thus you choose how to control that
#
sandeepshetty
which brings us to the other problem of displaying html securely.. think we should change this from "you may want to filter some HTML" to "you should filter unsafe HTML"
#
tantek.com
edited /comments-presentation (+465) "/* How to display */ prefer full content, use p-summary instead of truncating yourself, truncated only if that's what you need per your own site UI/design. also truncate the p-summary or p-name if you have to"
(view diff)
#
tantek
ok updated - take a look and see if that reflects what we discussed:
#
sandeepshetty
if e-content is dedault
#
tantek
hey Loqi - no diff?
#
aaronpk
it scrolled by real quick
#
sandeepshetty
Do we need this: "for longer posts where only a part of it is the comment". I would markup this up (or rather not have any markup) as a mention
#
sandeepshetty
s/this up/this
#
Loqi
sandeepshetty meant to say: Do we need this: "for longer posts where only a part of it is the comment". I would markup this (or rather not have any markup) as a mention
#
sandeepshetty
(as a publisher)
b0bg0d joined the channel
#
www.sandeep.io
edited /comments-presentation (-1) "/* How to display */ emphasized filtering e-content"
(view diff)
#
sandeepshetty
Should I remove "for longer posts where only a part of it is the comment" in favor of marking up posts like that as a mention instead of in-reply-to ?
tantek joined the channel
#
tantek
JonathanNeal - this one's for you :) http://indiewebcamp.com/FAQ#What_is_new
bnvk joined the channel
#
tantek
sandeepshetty - nah - I've seen comments where people start commenting on something and end up writing more extensively (perhaps on tangents)