#indiewebcamp 2013-10-29

2013-10-29 UTC
caseorganic, smus_, bnvk, paulcp and mikeal joined the channel
#
tantek.com
edited /silo-quits (+321) "add jwz quitting Dropbox"
(view diff)
#
tantek.com
created /silos-quits (+24) "redir a typo I made"
(view diff)
smus, mikeal, caseorganic, smus_, lukebrooker and shaners joined the channel
#
iamshane.com
edited /2013/Hollywood () "(-637) IWC:H is *two* days"
(view diff)
smus, ryana, skinny, ozten, caseorganic and paulcp joined the channel
#
shaners
aaronpk: just catching up on a few days of logs. and noticed your flights and "trips" (or whatever) plans. i like it.
#
shaners
you were searching for a good collective name for all kinds of "trips" that wasn't trips.
#
shaners
i like "travels", plural. or "journeys", plural.
lukebrooker, tpinto, mikeal and tantek joined the channel
#
aaronpk
shaners: thanks! yeah I've settled on "travel" now even though it's not a plural noun
#
shaners
travels is a totally legit plural noun
#
shaners
like, The Many Travels of Bilbo Baggins
#
aaronpk
but you wouldn't say "this is one travel"
#
shaners
sure you could
#
aaronpk
you could, but you would be wrong
#
shaners
travels (noun): she amassed great wealth during her travels: journeys, expeditions, trips, tours, excursions, voyages, treks, safaris, explorations, wanderings, odysseys, pilgrimages, jaunts, junkets
#
shaners
traveling, touring, sightseeing, backpacking, globe-trotting, gallivanting;
#
aaronpk
I like "jaunts"
#
shaners
yeah. it's cute and all. but to me "jaunt" is short.
#
aaronpk
the pattern I'm trying to follow with my top-level path segments is a plural noun which when singular describes a single item
#
aaronpk
notes -> note
#
aaronpk
articles -> article
#
aaronpk
replies -> reply
#
aaronpk
presentations -> presentation
#
aaronpk
events -> event
#
shaners
me too
#
shaners
you're not using singular nouns in your URLs, right?
#
aaronpk
correct
#
shaners
what's your plan for "audios"
#
aaronpk
what kind of audio?
#
shaners
let's say you write and record a song. or a voice memo. or "this crazy noise from the construction across the street"
jschweinsberg joined the channel
#
aaronpk
i think those are all very different things
#
aaronpk
I wouldn't put them all in the same category
#
shaners
oh yeah?
#
aaronpk
I have a couple dozen songs I should probably turn into "posts". right now they are all on one page.
#
aaronpk
I don't currently have any voice memos I would publish.
#
shaners
ok. the case of "this crazy noise from the construction across the street"
#
aaronpk
I ended up posting it as a note with a URL to the audio
#
shaners
no canonical url for the audio file itself?
#
aaronpk
same way I post images embedded in notes
#
aaronpk
well in this case it's hosted on soundcloud, so not ideal. but I imagine I wouldn't necessarily give the audio its own canonical URL even if it's hosted on my site.
#
tantek
scrolls up
#
aaronpk
i'm not sure, it's still something I'm trying to figure out. so I'm just doing whatever in the mean time and I'll see what sticks.
#
tantek
aaronpk - nice to see you expanding the boundaries with new post type experiments
caseorganic joined the channel
#
shaners
aaronpk: when you're here for fhc/iwc, can i get a few minutes of your time to look at my flickr archivr setup, please? i'm doing something wrong and would like an extra set of eyes on it.
ryana joined the channel
#
aaronpk
shaners: sure thing
indiewebcamp-vis joined the channel
#
aaronpk
I'll be in town all Sunday. Driving east early monday morning.
izto and snarfed joined the channel
#
shaners
any saturday?
#
aaronpk
yep of course
#
aaronpk
I'll be getting in Friday
#
shaners
k.. thought so.
#
aaronpk
this is why I needed to save the "trips" slug for my URLs
#
aaronpk
I want to send you a URL with my trip itinerary that shows my flight in, my hotel info, the farmhouseconf and indiewebcamp events, and my departure
#
shaners
trips is a kind of meta collection of "jaunts"?
#
aaronpk
it would basically look like my tag pages, showing all the various posts for that trip
#
Loqi
SHOW ALL THE VARIOUS http://loqi.me/89u
#
shaners
Loqi: shhh
#
aaronpk
a trip is a collection of travel, events, etc
#
aaronpk
might even include photos and notes (probably carefully curated, since not all notes posted during a trip are relevant to that trip)
tpinto and skinny joined the channel
#
tantek
has anyone here taken a look at getradar.in?
skinny joined the channel
#
tantek
appears to be a Dopplr replacement - they support importing Dopplr data supposedly
#
aaronpk
wow it looks like it's trying to completely replicate dopplr
#
aaronpk
oh he just made it?
#
tantek
indeed
#
tantek
so the friends re-use from Twitter is not something I can work with
#
tantek
sharing your trips is *very* different than following someone on Twitter
#
tantek
sure there is some overlap, however there are plenty of (most?) people that the opposites apply
#
aaronpk
agreed. currently I have 3 trips published on my site publicly. however, I'm waiting to make sure I've implemented authenticated trips properly before adding more flights.
#
aaronpk
also I plan on showing various levels of detail for each trip and flight depending on who you are
#
tantek
aaronpk, I could also see wanting to change the "public"ness of trip information based on future / past
#
aaronpk
the main purpose I'm using my travel permalinks for already is telling people when I'm arriving or when I'll be in town
#
aaronpk
tantek: interesting point! hadn't considered that yet.
#
tantek
aaronpk - agreed - that's what Dopplr was really good for
#
tantek
I'll likely make most of my *future* trip info *private*, and then as I return home, make most of my trip info *public*.
#
aaronpk
it's funny that already in the day and a half these links have been live I've already shared them with 3 different people and avoided sending them emails or super long text messages
ryana joined the channel
#
tantek
lower security threat model for past trips than for future trips
#
tantek
aaronpk - that's great!
#
aaronpk
yeah, I can totally see hiding flights/trips until after they're over. also the flight number is not really relevant after the trip is over either.
thatryana joined the channel
#
aaronpk
and showing my seat is really only relevant to people i'm traveling with, so nobody else should see it even though it's not a "security" or "privacy" thing
#
tantek
aaronpk - seat and flight #s are a security thing because of griefing potential
snarfed joined the channel
#
aaronpk
wow you'd have to be pretty hardcore to track down my flight # and seat and buy a ticket next to me to grief me
#
tantek
aaronpk - unfortunately there are other attack vectors.
#
aaronpk
well said
#
KevinMarks
importing friends from foursquare would make more sense
#
aaronpk
KevinMarks: true! at least there's some similar levels of trust in that group of people
caseorganic joined the channel
#
tantek
KevinMarks - agreed.
#
KevinMarks
is there an indieweb pattern here of delegating friends lists?
#
shaners
KevinMarks: tantek and I talked through an idea while in pdx for xoxo, but built nothing (or documented the idea). so... not really?
#
KevinMarks
grrr. GoDaddy (which I have things registered at for legacy reasons) won't let me use nameservers with hyphens in
#
tantek
KevinMarks - what's the use case?
#
KevinMarks
which Amazon have
#
tantek
(for delegating friends lists) ?
#
KevinMarks
the use case is co-opting an existing friend graph (at twitter, foursquare etc) rather than having to maintain rel-friend links
#
KevinMarks
like IndieAuth does
#
tantek
IndieAuth hides a backend auth provider though, and conditionally so, with fallbacks.
#
tantek
so that's a bit different
#
KevinMarks
challenge is that the friend graph may not have urls
#
tantek
the closest we have to re-using existing friend lists (it's not a graph, it's just a list) e.g. at Twitter is some of the spam brainstorming
#
KevinMarks
true about list
#
tantek
KevinMarks - stop calling it graph. That was misframing propagated by Zuck & co.
#
tantek
(since like 2008)
#
KevinMarks
OpenSocial we had to make the ID not be a URL in the friend list
#
KevinMarks
well, FB has a graph - in principle you could ask for 2nd degree connections
#
KevinMarks
in practice they don't do that
#
KevinMarks
I did say list to start with.
ryana and crossdiver joined the channel
#
KevinMarks
rel-me is a graph
#
KevinMarks
rel-friend (etc) is
#
KevinMarks
a list of friends implicitly has multiple urls per friend due to rel-me
#
tantek
KevinMarks - rel-me itself may form the basis of a graph, but that result is still a single node when you're talking about "people", which "friends" is a special case thereof.
#
tantek
thus a friend list is still a list, not a graph, for any/all computational purposes.
tpinto joined the channel
#
KevinMarks
the challenge is what the unambiguous representation of the friend is
#
KevinMarks
if you want to do set logic on the lists (friends in common etc)
#
tantek
sounds like you're making it harder than it needs to be
#
tantek
I think when you focus on a specific use-case, the problem is greatly simplified
#
tantek
so rather than solving anything in abstract, e.g. "friends in common", just code what you need for your next important use-case for yourself and see what you figure out.
#
KevinMarks
I'm still fighting my own domain DNS crap :(
#
KevinMarks
also playing the 'where the heck did I register that domain?' game
tilgovi, hupili, smus, tpinto, LauraJ, tobiastom, cweiske and andreypopp joined the channel
#
tantek
!tell benwerd I noticed you're using Disqus on your previous personal URL benwerd.com, e.g. http://benwerd.com/2013/03/14/silos-the-open-web-and-selfdogfooding/ - beware recent Disqus ad sneakery controversy: https://twitter.com/fdevillamil/status/394368809597149184
#
Loqi
Ok, I'll tell them that when I see them next
smus and Jihaisse joined the channel
tpinto and melvster joined the channel
#
@DatenundGesundh
Lectures from the Data and Health Symposium @ETH are now online http://www.datenundgesundheit.ch/2013/10/29/vortrage-des-symposiums-data-and-health/?lang=en #persmed #genomics #ownyourdata #ethics #bigdata
(twitter.com/_/status/395102349859237888)
smus and pfefferle joined the channel
#
KevinMarks
yay - http://kevinmarks.com now is me again
#
Loqi
giggles
#
neuro`
KevinMarks: congrats.
#
neuro`
KevinMarks: DNS setup works like a charm here.
#
KevinMarks
this was a long and tangled path, eventually leading to a godaddy account I'd forgotten about
#
Jihaisse
what was the problem ?
tpinto and smus joined the channel
#
KevinMarks
I registered the domain ages ago
#
KevinMarks
I had the email admin set to a domain I don't have control over any more;
#
KevinMarks
it was very complicated
LauraJ joined the channel
dpk, earplugs, smus, bnvk, andreypopp and tpinto joined the channel
#
KevinMarks
pondering Ghost and whether giving them the full indieweb markup makeover is worth it
#
KevinMarks
it's a cute little blogging platform, but markdown bugs me
#
pdurbin
KevinMarks: what do you prefer to markdown?
#
cweiske
I prefer rST, but don't its link markup
#
cweiske
I like it because you have so much more possibilities of marking up things
#
cweiske
they have a defined structure for extensions
#
cweiske
we use it to write docs for our software @work, convert it automatically to wiki formats (e.g. confluence) and deploy it automatically to our customer's wikis
#
cweiske
markdown doesn't even have definition lists
#
cweiske
or tables
bnvk, barnabywalters and smus joined the channel
#
barnabywalters
KevinMarks: RE ghost + mf2, I sent a PR and they’re not interested at the moment: https://github.com/TryGhost/Casper/pull/29
#
barnabywalters
but feel free to use my fork of Casper as the basis of an indiewebified theme if it’s any help
tpinto joined the channel
#
pdurbin
I always use autolink in markdown if it's supported
bnvk joined the channel
#
tommorris
barnabywalters: some kind of mashup of Ghost Caspar and Wordpress' P2 would be interesting ;)
#
pdurbin
cweiske: we just moved our docs to sphinx so I'm trying to get into rST a bit. Markdown seems so much more simple. I like Markdown's "plain text email" philosophy.
#
Jihaisse
why markdown is better than wysiwyg ?
josephboyle, bnvk, tpinto and ttepasse joined the channel
#
pdurbin
Jihaisse: I'm not sure what you mean. I can edit markdown in vim
#
@bru
RT @t: Updated: How To Export Your Data From Dopplr http://tantek.com/2013/294/b1/export-your-data-from-dopplr @dopplr shutdown in 4 days. #ownyourdata #indieweb (ttk.me t4Sm1)
(twitter.com/_/status/395163153514119168)
#
@abragad
RT @t: Updated: How To Export Your Data From Dopplr http://tantek.com/2013/294/b1/export-your-data-from-dopplr @dopplr shutdown in 4 days. #ownyourdata #indieweb (ttk.me t4Sm1)
(twitter.com/_/status/395163204277792768)
#
Jihaisse
pdurbin: in a blog plateform, why using a markdown is better than using a wysiwyg editor ?
#
pdurbin
Jihaisse: it's probably worse, not better. but this is how I like to blog (i.e. not with wordpress): http://tom.preston-werner.com/2008/11/17/blogging-like-a-hacker.html
#
cweiske
Jihaisse, I don't like to use a mouse when writing text
#
tommorris
WYSIWYG doesn't deal with semantics well.
#
tommorris
how does one represent an abbreviation in WYSIWYG?
#
neuro`
Most WYSIWYG editors are actually WYGIWYD
#
neuro`
What You Get Is What You Deserve
#
tommorris
WYGIVRTWYT - What you get is vaguely related to what you typed.
#
neuro`
Something like this yeah
#
Jihaisse
yeah, but for lambda people, it's not really easy
tantek joined the channel
#
tantek
KevinMarks - congrats on getting KevinMarks.com back!
#
tantek
is there anything about that process you can document publicly to help out other indieweb folks?
brianloveswords joined the channel
#
pdurbin
easier to spell than "Epeus' epigone" ;)
#
tantek.com
edited /Disqus (+1510) "document recent Disqus ad sneakery criticism with citations and embedded screenshots"
(view diff)
#
tantek
pdurbin :)
#
tantek
!tell benwerd: Ok I wrote up the disqus controversy on the wiki page to make it easier to find, since searching for "disqus" in Google (even Google news) just brings up every site that *uses* Disqus: http://indiewebcamp.com/Disqus#Displays_ads_on_site_to_public_but_not_owner
#
Loqi
Ok, I'll tell them that when I see them next
#
tantek.com
edited /Disqus (+12) "/* See Also */ silos"
(view diff)
bnvk joined the channel
#
tantek
hey bnvk - how's your post / reply-context markup going?
smus joined the channel
#
bnvk
tantek: I haven't worked on it since yesterday, I'd like to get to it over the weekend but it's unlikely since there is a huge music festival taking over Reykjavik and I have friends from US in town for it :/
#
tantek
bnvk - should take like 5 minutes!
#
bnvk
hehe, ok
#
barnabywalters
bnvk: I’m always around to help with mf2 stuff too if you need it :)
#
tantek
yes :)
#
bnvk
cool, thanks!
#
bnvk
you guys rule
#
barnabywalters
pop into the office sometime :)
#
bnvk
I should / will do that!
#
Loqi
I agree
#
tantek
bnvk - say hi to the folks in Iceland from the SF Embassy for me ;)
#
bnvk
SF Embassy?
#
barnabywalters
oh, they’re actually coming here for a movie screening on thurs
#
barnabywalters
(here being the office)
#
bnvk
who?
#
barnabywalters
some of the people from SF Embassy
#
barnabywalters
brian knows them
#
tantek
bnvk - seriously - with barnabywalters and me here, it should take you mere minutes to update your microformats2 markup on your permalinks
#
bnvk
whoa, this is the first i'm hearing of this week
#
bnvk
there's so much going on in lil Reykjavik
#
barnabywalters
now with sample code, black boxes and small-caps
#
bnvk
ok, lookin into it all
#
bnvk
that's coming along really nice
josephboyle, sfarthing and dybskiy joined the channel
#
bnvk
tantek: barnabywalters: is there any strictness about nesting of containers with MF or does it look recursively
#
tantek
bnvk - they work just like nested HTML elements
#
barnabywalters
child microformats don’t have to be on child elements of the parent though
#
tantek
do you have a specific question?
#
tantek
barnabywalters - that's a *very* confusing statement.
_6a68 joined the channel
#
tantek
please restate it in a positive way
#
bnvk
right right, but I mean- can I do the "p-author p-name" x number of children deep inside of the "h-entry" ?
#
barnabywalters
child microformat root classnames can be on any descendant HTML element of the parent root classname
#
barnabywalters
e.g. div.h-entry > div > div.p-author.h-card is fine
#
bnvk
"can be on" or rather "can be inside of" ?
#
bnvk
ok, that's what i'm asking :)
#
tantek
bnvk you can put microformats property class names on any descendant inside a microformats root class name element, WITHOUT any intervening microformats root class names in the descendent chain
#
tantek
bnvk "can be on" can be confusing
#
bnvk
yah, exactly what I mean
#
barnabywalters
http://waterpigs.co.uk/php-mf2/ is a good sandbox for this kind of experimentation
#
tantek
however you CANNOT put a microformats property class name for its root class name on the *same* element, e.g. this is wrong: <span class="h-card p-name"> <-- won't work.
#
barnabywalters
it’s not yet updated to v0.2.0 yet though
#
tantek
when you put a microformats property and root class name on the same element, it embeds that root microformat as a value of that property class name, which then is a part of whatever root class name is above it (nearest ancestor root class name)
#
barnabywalters
oops, looks like a bug :)
#
bnvk
tantek: yah, I get that part about multiple MF on same element
#
tantek
bnvk - it's not about multiple microformats on the same element
#
tantek
so that makes me thing there's still confusion
#
bnvk
well, MF with a hierarchy is more what I mean
#
tantek
bnvk - from that pin13 URL - you're putting h-cite and h-entry on the same element, which is not what your post means
#
bnvk
I get that multiple MF which are different "levels" can exist on the same element
#
tantek
bnvk - they can, but the uses for that are quite rare
#
tantek
if you're doing that, you're likely making a mistake
#
bnvk
ah, should be just "h-cite" right?
#
tantek
e.g. this is wrong: class="note-reply-context p-in-reply-to h-entry h-cite"
#
tantek
it's making the error I pointed out above about <span class="h-card p-name"> <-- won't work.
#
tantek
you're putting "p-in-reply-to h-entry" on the same element which is wrong
#
tantek
p-in-reply-to is supposed to be a property of the h-entry
#
tantek
which means it must go on an element *inside* the h-entry
#
tantek
or rather, the h-entry must be *outside* the element with p-in-reply-to
#
bnvk
hrm, I was copying that "p-in-reply-to h-entry" on the same element from barnabywalters implementation
#
bnvk
doh, wait, that's h-cute
#
bnvk
cite
#
tantek
bnvk - sounds like a misreading of barnaby's implementation
#
tantek
look closely at the first two lines of code here: http://indiewebcamp.com/reply-context#Markup
#
tantek
note that the h-entry is OUTSIDE the p-in-reply-to
#
tommorris
Ooh. indiewebify.waterpigs.co.uk? that's nifty.
#
tommorris
barnabywalters++
#
Loqi
barnabywalters has 18 karma
ozten joined the channel
#
bnvk
I think the confusion (for me) stems from a seeming incongruity with the following: "h-entry > h-cite > p-author" then the author of the actual post is just "h-entry > p-author"
#
bnvk
rather than having it's own more granular container
#
tantek
each *does* have its own granular container
#
tantek
h-cite > p-author is the author you are citing, not you
#
bnvk
yes yes
#
barnabywalters
this post is in reply to a post by this author
#
barnabywalters
as opposed to
#
barnabywalters
this post is by this author
#
tantek
the property applies to the nearest root ancestor. where's the incongruity?
#
bnvk
but, the reply context is 3 levels deep to get to the p-author where as for the normal (non reply entry) it's only 2 levels of "required" MF tags
#
tantek
in both cases there are only two levels. the root class name element, and the property class name element.
#
tantek
any intervening elements don't matter (as long as they themselves don't have root class names)
#
bnvk
no, the from the "root" being h-entry in the reply context p-author is 3 levels?
#
tantek
bnvk - no. only element with microformats class names on them add to "levels"
mikeal joined the channel
#
bnvk
yes, that IS what I am saying
#
tantek
the reply-context h-cite's p-author has NOTHING to do with the h-entry above
#
bnvk
I get that
#
tantek
from the h-entry's perspective, it just has an embedded h-cite, it knows/cares NOTHING about what's inside it
#
bnvk
I think i'm just thinking about this wrong
#
tantek
each h-* class wrap all the p-* inside it so any further up the chain h-* parents don't see them
#
tommorris
would something like <a rel="author" href="#me"> work?
#
tantek
each h-* class is a wrapper
#
tantek
tommorris - what would do or mean?
#
tommorris
I have a footer on the page with my details. I want to link to it rather than have an embedded hCard in each post
#
tantek
tommorris do you mean home page?
#
tantek
in that case maybe you mean: <a rel="author" href="/#me">
#
tantek
(you were missing the "/" before the #me)
#
tommorris
tantek: no, on every page, at #me there's a footer element which is an hCard
#
tantek
I don't think we support doing anything with any fragments in the current authorship algorithm
#
barnabywalters
okay, http://indiewebify.waterpigs.co.uk/validate-h-entry/?url=http%3A%2F%2Fbrennannovak.com%2Fnotes%2F338 should give more useful feedback upon finding posts where the author is just a string
#
barnabywalters
instead of failing, which is what was happening before
#
bnvk
yah, i think i've got the structure about right
#
tantek
barnabywalters - hey that's pretty slick
#
bnvk
trying to fix my datetime value and add a url
#
barnabywalters
tantek: that’s the plan :)
#
tantek
barnabywalters - I'd dump the rel=tag suggestion - we have evidence that mixing rel and class confuses authors, and they're unlikely to get the URL structure right for a rel-tag anyway.
#
barnabywalters
bnvk: woah, looking way better now!
#
tantek
just use p-category
#
barnabywalters
okay, noted
#
barnabywalters
btw there is a GH repo for this so you can make copy edits via the web UI
#
barnabywalters
at some point we’re either going to have to make editing it easier or link off to the wiki everywhere
#
tantek
bnvk - you should use "p-in-reply-to h-cite" instead of "p-in-reply-to h-entry"
#
tantek
it's not an h-entry because it's not your content. it's you citing someone else's content so that's why it should be an h-cite
#
bnvk
dah, yah... forgot that element
#
tantek
barnabywalters - would be cool to add reply-context checking to the valid-h-entry too!
#
tantek
if it finds an in-reply-to property, it can suggest marking it up as an h-cite
#
barnabywalters
tantek: it’s next on the list! gets a bit complex as it’s basically the same thing nested
#
aaronpk
that validator is sweet
#
tantek
and if it finds it improperly marked up as a p-in-reply-to h-entry, then it can flag that as an error!
#
tantek
aaronpk - totally
#
aaronpk
love the prompts for making it more complete!!
#
tantek
barnabywalters please add these to microformats.org/wiki/validators
#
tantek
those are amazing
andreypopp joined the channel
#
barnabywalters
should these go in a new section for vocab-specific validation UIs?
#
barnabywalters
oh, there’s already one
#
barnabywalters
will read the whole page before asking questions from now on
#
tantek
in other news - was just seriously badly bitten by Gmails new inbox sorting
#
tantek
Primary | Social | Promotions
#
tantek
turned it off and finding all kinds of critical stuff I missed because it got auto-filed in Promotions
#
tantek
yeah :(
#
aaronpk
wonders what kind of critical emails tantek receives given his strong dislike for email
#
@t
#Gmail automatic inbox sort
#
aaronpk
ah yeah
mikeal and benprew joined the channel
#
Jihaisse
tantek: there is a missing / in your minified url : https://twitter.com/t/status/395214891692605440 (ttk.me t4Sn2)
#
@t
#Gmail automatic inbox sort
#
tommorris
tantek: fastmail is a big improvement IMHO
#
tommorris
and my utter hatred for email isn't far off yours. ;)
#
@brennannovak
@t I hear some people are working on a better email client... might even have some #indieweb features ;)
(twitter.com/_/status/395216226802077696)
#
aaronpk
bnvk: can you make my email less like email?
#
aaronpk
excellent
#
bnvk
I'm working on a blog post outlining that very thing
smus and benprew joined the channel
#
bnvk
barnabywalters: hrm, I can't seem to get the validator happy with publication datetime
#
bnvk
I'll come by the office in a few minutes
#
barnabywalters
it’s being parsed as a property of your h-card
#
bnvk
ah, so that's "wrong" hierarchy
bnvk joined the channel
#
barnabywalters
tommorris: interesting! not sure what’s up with author there
#
tommorris
I don't have an hCard in the h-entry
#
barnabywalters
ah, I’m checking something wrong
#
barnabywalters
yeah, it should prompt you to add one
#
barnabywalters
there are lots of nested ifs in PHP template code at that point
#
tommorris
I really want to reference the one on the page already. ;)
smus, ryana, _6a681 and npdoty joined the channel
#
tantek
jihaisse - it's a permashortcitation, not a minified URL, by design: http://indiewebcamp.com/Twitter#POSSE_entire_note_to_twitter
#
tommorris
barnabywalters: handy. but we really need to sort out microformats-2 include pattern - itemref or object or something
#
tommorris
having an hCard in every entry when there's already a page-level author seems DRY violating to me
#
barnabywalters
tommorris: my biggest issue with the include pattern is recursion, and JSON’s inability to do recursion
#
barnabywalters
so we’d have to limit it to one level
#
barnabywalters
tommorris: there’s always h-feed > p-author, with h-entries as siblings
#
tommorris
I could plonk h-feed on <body> ;)
#
barnabywalters
that only makes sense if it is actually a feed
#
barnabywalters
for h-entries, put h-entry on the body
#
barnabywalters
then the author being outside the .h-entry element is no longer a problem
#
tantek
barnabywalters - would adding fragment support to the authorship algorithm be too hard? we already have fallback to rel-author
#
barnabywalters
tantek: I’m meh about mf fragment parsing. fragments are intangible, tricky to understand
#
barnabywalters
widely misunderstood
#
tantek
agreed, my worry too.
#
tantek
barnabywalters - yeah, I tend to agree that <body class="h-entry"> is the right answer to permalink pages
#
tantek
good call
_6a68 joined the channel
#
barnabywalters
I like the mf approach of “a page is a thing” as opposed to the more Semweb approach of “each thing on a page with an @id or @resource is a thing”
#
barnabywalters
pages are tangible, ids/“resources” not so
#
barnabywalters
and putting root classnames on the body element just strengthens that approach
#
tantek
also ids/resource are more fragile
#
tantek
barnabywalters - agreed about strengthens approach.
#
barnabywalters
@resource is *completely invisible*
#
tantek
also the <body class="h-*"> pattern is a path out of the OGP metacrap mess.
#
Jihaisse
tantek: oh, ok… didn't know that
bnvk joined the channel
#
tantek.com
edited /Twitter (+212) "/* POSSE entire note to twitter */ add explicit subhead to link to why permashortcitation"
(view diff)
ozten, npdoty and glennjones joined the channel
#
glennjones
hi all, looking coding up checkins pages on my site, is there any standard format i.e. h-event ? Ben is using a h-entry with a geo in the e-content http://werd.io/2013/checked-into-papalote-mexican-grill
dybskiy joined the channel
#
tantek
glennjones - had a look at http://indiewebcamp.com/checkin ?
#
tantek
glennjones - do you have mockups for how you want your checkins to look?
#
tantek
the UX of a checkin post is more important than the markup
#
glennjones
tantek it simple geo(as a map), date time venue name(optional) and comment(optional) - feels like an event
#
tantek
except that it's in the past
#
tantek
and not something you invite people to
#
tantek
not something you RSVP to
#
tantek
a checkin can be a post that there *was* an event, but a checkin is no more an event than a Tweet is an event
cweiske joined the channel
#
tantek.com
edited /posts (+47) "add explicit subhead Inferring post kinds from properties so we can discuss it in particular"
(view diff)
#
tantek
a checkin is merely a *note* with a *location* that's an explicit *venue*
#
tantek
at least last time aaronpk and I brainstormed about this
#
tantek
don't know if he's changed his opinion since
#
tantek
also - how is tommorris marking up his checkins?
#
tantek
interesting, tommorris, consider adding p-location to this: <div class="h-card checkin">, e.g. <div class="p-location h-card checkin">
#
tantek
to make it explicit that the post happened at that location
#
tantek
glennjones - can you add the Ben Werdmuller h-entry checkin example you found to this section in the wiki? http://indiewebcamp.com/checkin#People_actively_publishing_check-in_data
#
glennjones
A checkin does has a strong like between the date time and location, I just think that h-event expresses that. The two other unsaid things is that I was there and its in the past
#
tantek
barnabywalters - you were looking into checkins at some point too
#
tantek
it looks like we have 3-4 implementations of indieweb checkins now, but no guidance on markup on : http://indiewebcamp.com/checkin
#
glennjones
Yes I will add the examples onto the wiki for discussion
#
tantek
do you know if we put checkin markup / guidance brainstorming somewhere else?
#
tantek
thank you glennjones
#
tantek
it's good to capture multiple perspectives
glennjones joined the channel
#
barnabywalters
http://waterpigs.co.uk/notes/?tagged=location also has non-checkin posts with location data attached
#
barnabywalters
looks like I’ve got h-entry > p-location.h-adr.h-geo
smus joined the channel
#
tantek.com
edited /Falcon (-489) "/* Working On */ got this working a few days ago: pre-fill in-reply-to field with p-in-reply-to URLs from h-entry of the current post. not really a feature, just one fewer inconvenience."
(view diff)
benprew, mikeal, tobiastom, josephboyle and smus joined the channel
#
tantek.com
edited /micropub (+654) "Form markup first approach"
(view diff)
#
bnvk
okie, that's enough MF tinkering for today
#
bnvk
"it'll just take 5 minutes" he said... ;)
#
tantek
it would have - but there was pre-existing confusion we had to overcome ;)
#
barnabywalters
(yoda voice) you must unlearn what you have learned
#
tantek
barnabywalters++
#
Loqi
barnabywalters has 19 karma
paulcp joined the channel
#
tantek
barnabywalters - what are the tools/code you have that does original-post-disocvery and rel-syndication discovery?
#
barnabywalters
OPD code not all currently open source, PHP version available here: http://waterpigs.co.uk/services/original-post
#
barnabywalters
rel-syndication discovery and validation is in indiebify.me but it’d be trivial to move to a new package
#
barnabywalters
it’s only in there for convenience during development
#
tantek.com
edited /Falcon (+559) "/* Working On */ reclarify original post discovery and rel-syndication implementation tasks, reprioritize them higher since they should be easier to implement, and immediately improve posting interface for me (reduce steps & stress)"
(view diff)
#
barnabywalters
which reminds me, I need to convert the pseudocode algorithms on the mf wiki into prose algorithms people can actually recreate
#
tantek
rel-syndication is what I was looking for
#
barnabywalters
mainly starting at line 94
crossdiver joined the channel
#
barnabywalters
tantek: original post discovery? at the moment my code is horribly messy and non-spec-compliant
#
barnabywalters
working only for twitter
#
barnabywalters
at the mo I’d recommend using my /services/original-post endpoint
#
barnabywalters
then as we incrementally improve it I’ll release it as a package
#
tantek
doing replies right is a lot of work
#
tantek
lots of details to get right, especially with POSSE involved
#
tantek
making more sense why no one has previously figured out real workable blog to blog commenting before.
crossdiver joined the channel
#
tantek
barnabywalters, rel-syndication discovery should be easier
#
tantek
(easier than original post discovery)
#
tantek
since the indieweb person controls the source side of the rel-syndication parsing challenge
glennjones joined the channel
#
tantek
original-post-discovery is hard because you're trying to discern information from a silo permalink which we have no control over the markup thereof
ShaneHudson joined the channel
#
tantek.com
edited /Falcon (+369) "/* Working On */ subheads since working on so many things, move rel-syndication to top, a few minor edits"
(view diff)
#
tantek
also I'm trying to document the reasoning for each step of implementing replies so that the next person/people that come along will hopefully both understand it better, and feel that the work is justified by actual needs / uses
#
tantek
writing all this out makes me even more impressed with all the stuff you guys (aaronpk, barnabywalters, tommorris, etc.) have gotten working
benprew, tilgovi, jschweinsberg and tobiastom joined the channel
#
_6a68
tantek: wow, that's a really good point. how do you even go about doing that? post in the forum with a permalink back to your site?
#
_6a68
i'd never thought about POSSE forum/stackoverflow/etc posts, but it's a great idea
#
tantek
bingo
#
tantek
aaronpk is POSSEing to github already
#
_6a68
whoa rly
#
_6a68
i love the indieweb worldview <3
#
_6a68
zomg, aaronpk's site is just so pretty
#
_6a68
thanks for the link tantek :-D
smus joined the channel
bnvk joined the channel
#
_6a68
tantek: related question: if you POSSE to forums, what about IRC?
#
tantek
IRC is more like a commons
#
tantek
forums are more like silos
#
tantek
at least freenode IRC and #indiewebcamp in particular
#
_6a68
ah, I see what you mean
#
tantek
logs are archived publicly so anyone can get them
#
tantek
but yes - a corporate IRC channel / server would make sense as a POSSE destination as well
#
_6a68
right, there's no canonical link and no central control of the content in the case of freenode
#
_6a68
cool, thanks!
#
KevinMarks
now I have my domain back, I can get to self hosting, which is why I looked at Ghost.
#
tantek
!tell benwerd,barnabywalters - here's an original-post-discovery challenge - how do we find benwerd's original posts from his FB POSSE copies e.g.: https://www.facebook.com/ben.werdmuller/posts/10100631599571559
#
Loqi
Ok, I'll tell them that when I see them next
#
Loqi
Ok, I'll tell them that when I see them next
smus joined the channel
#
tantek.com
edited /rel-syndication (+30) "/* See Also */ original-post-discovery"
(view diff)
#
tommorris
wonders if indiewebsters might be interested in http://tommorris.org/posts/8619 ;)
bnvk joined the channel
#
tantek
tommorris - is it open source that people are running on their own domain?
mikeal joined the channel
#
tantek
otherwise, eh
#
tantek
bnvk - been looking at your note/reply markup on https://brennannovak.com/notes/338 - nice job!
#
bnvk
tantek: it's coming along :)
#
tommorris
tantek: any open source. I'm documenting a vim plugin at the moment, but documenting our own stuff is good too
#
tantek
I'd say documenting the stuff you live & breathe is far more important than any other open source
#
tantek
bnvk - couple of suggested improvements for your h-cite
#
tantek
the <a class="p-summary p-name p-content" href="http://blogs.telegraph.co.uk...
#
KevinMarks
a checkin is an event for as long as you're still there - this is the foursquare case. If you have an endtime, RSVPs make sense until you leave
#
tantek
should be <a class="p-summary p-name p-content u-url" href="http://blogs.telegraph.co.uk
#
KevinMarks
or the neo-dopplr one where it indicates arrival in a city for a few days
#
tantek
bnvk - and the publisher reference: <a class="u-url p-author p-name" href="https://brennannovak.com/people/thetelegraph" target="_blank">The Telegraph</a>
#
tantek
bnvk should be: <a class="p-publisher h-card" href="https://brennannovak.com/people/thetelegraph" target="_blank">The Telegraph</a>
#
tantek
bnvk though I noticed that that link just goes to your home page which isn't the best experience, so this would be better in that case: <span class="p-publisher h-card">The Telegraph</span>
#
tantek
bnvk - hope that helps
#
bnvk
cool, i'll update :)
#
tantek
there are some duplicates in there you may want to take a look at
#
tantek
but most consuming applications I think will handle that np
#
aaronpk
bnvk: I also noticed you're still using myopenid.com which is going to shut down soon! Luckily there's an easy replacement :) http://indiewebcamp.com/indieauth#Use_IndieAuth_for_your_OpenID
glennjones, bnvk and snarfed joined the channel
#
KevinMarks
aha, thanks aaron. I need an openid replacement
smus joined the channel
#
mikeal
!tell caseorga_ i can't make it to LA this weekend for FarmhouseConf, the logistics and cost ended up being too much
#
Loqi
Ok, I'll tell them that when I see them next
#
KevinMarks
now what still wants an openid login so I can test it...
eschnou joined the channel
#
KevinMarks
I was thinking livejournal
#
aaronpk
ohhhh yeah
#
tantek
heads-up KevinMarks - benwerd and I are conspiring on an indieweb dinner in SF on Friday night.
#
KevinMarks
"kevinmarks, your current position in the Top Journals is: 2,786,781"
#
crossdiver
Would writing a PESOS-style backup for Disqus make it all better?
#
crossdiver
trying to do comments on my site, but want non-indieweb folks to still be able to comment easily
#
KevinMarks
"Comment Posted Success Your comment has been added. According to this journal's settings, it was marked as spam."
#
bnvk
tantek: cool, thanks for the protips :D we should update IndieWebifyMe to denote duplicate entries
#
tantek
this is all a good learning experience for learning about indieweb/microformats itself
#
bnvk
aaronpk: yah, I definitely will get around to indieauth, one of these days :)
#
aaronpk
get around to? doesn't it already work?
#
KevinMarks
so LJ successfully lets me login using openID with IndieAuth, but decides I'm a spammer anyway
#
tantek
screenshot?
#
KevinMarks
I had to change prefs to turn off spam filtering, then it 'screened' the comment with a 3 step 'are you sure' process to let it though
#
bnvk
aaronpk: I mean adding it to my site so I or others can site in with IndieAuth
#
aaronpk
oh, whatever
#
tantek
bnvk - why not take the 30 seconds it takes to swap out myopenid for indieauth for your delegation at least?
#
aaronpk
you can change your openid tags in like 30 seconds and get off myopenid
#
aaronpk
that's what I meant
#
bnvk
oh, ok. I was unaware it was that easy
#
KevinMarks
yep, it was way faster than actually getting the comment to post.
#
bnvk
additionally indieauth continues to be really fickle for me signing in anywhere else due to the fact I'm using https and it's caching or something... that persistent bug
#
aaronpk
I thought I fixed that in the last batch of changes?
#
bnvk
nope
#
KevinMarks
now I need to fix my external accounts to point to kevinmarks.com instead of kevin-marks.com
#
bnvk
my only URL that keeps showing a green is Twitter
#
bnvk
and it still throws the ruby error
#
KevinMarks
my google+ one works because that has loads of rel-me's out so I can auth with any account
#
bnvk
I always test signing in to IndieWebcamp
#
tantek
exactly
#
aaronpk
shouldn't matter what you're signing in to
#
bnvk
hrmm
#
bnvk
that works just fine with Twitter on indieauth.co,
#
bnvk
.com
#
bnvk
but on indiewebcamp.com it's still totally broken for me
#
bnvk
i'll happily screen share / help debug this
#
aaronpk
you'll have to enter explicitly https://brennannovak.com in the "sign-in" form until https://github.com/aaronpk/IndieAuth/issues/28 is fixed
tilgovi joined the channel
#
bnvk
yes, that is what I enter "https://brennannovak.com/"
#
aaronpk
ah, the wiki is removing https!
#
aaronpk
slaps the wiki
#
bnvk
hehe
benwerd joined the channel
#
Loqi
benwerd: tantek left you a message 12 hours, 18 minutes ago: I noticed you're using Disqus on your previous personal URL benwerd.com, e.g. http://benwerd.com/2013/03/14/silos-the-open-web-and-selfdogfooding/ - beware recent Disqus ad sneakery controversy: https://twitter.com/fdevillamil/status/394368809597149184
#
Loqi
benwerd: tantek left you a message 5 hours, 40 minutes ago: Ok I wrote up the disqus controversy on the wiki page to make it easier to find, since searching for "disqus" in Google (even Google news) just brings up every site that *uses* Disqus: http://indiewebcamp.com/Disqus#Displays_ads_on_site_to_public_but_not_owner
#
Loqi
benwerd: tantek left you a message 1 hour, 16 minutes ago: - here's an original-post-discovery challenge - how do we find benwerd's original posts from his FB POSSE copies e.g.: https://www.facebook.com/ben.werdmuller/posts/10100631599571559
#
benwerd
!tell tantek Thanks for the Disqus note - I haven't altered my old WP install at all. I should probably deactivate it (all Disqus comments are copied to my local WP db)
#
Loqi
Ok, I'll tell him that when I see him next
#
@kevinmarks
@dsifry @tomcoates @anildash webmention is the reinvention of pingback, based on what we know now
(twitter.com/_/status/395277757409529856)
#
@benwerd
Did you know? Disqus displays ads on your comments but hides them from you. Just disabled everywhere. http://indiewebcamp.com/Disqus#Displays_ads_on_site_to_public_but_not_owner
(twitter.com/_/status/395278370323181568)
#
@anildash
Seems like a Technorati-circa-2004 service that we ping with our tagged blog posts would be infinitely cheaper to build today. Kickstarter?
loleg joined the channel
#
benwerd
Would love tantek's feedback on that
#
aaronpk
you know how people make shiny things like http://humanstxt.org/ and get everyone excited about adding it to their site really quick?
LauraJ and smus joined the channel
#
aaronpk
I wonder what it would take and/or if it would be a good idea to try that with a really simple indieweb thing, like indieauth, or an h-card on your home page
#
bret
afternoon iwc
smus_ joined the channel
#
bret
aaronpk: I would wager that would be a good tactic. When I was learning html5 half the fun was looking at all the shiny websites
tpinto, caseorganic and _6a68 joined the channel
#
benwerd
aaronpk: I like the idea of an h-card in particular. Something like an indie-technorati like in anil's post could be a great hook
#
aaronpk
a couple years ago someone made this in portland: http://prtlnd.com/
#
aaronpk
really simple directory of people, you added a photo, your name, two tags, and a link to your site
#
aaronpk
it got picked up a ton in portland. started out with just "designers" and "developers" and they quickly had to add tags for artists, musicians, etc
#
aaronpk
i'm thinking it would be fun to do that again but pick up all the name/photo/tags from the URL you enter. so the only thing you'd type in is your website.
scor joined the channel
#
aaronpk
the key is it has to a) look really good and b) needs to be part of some community, either geographic or topical, so people want to add themselves to the list
squeakytoy, caseorganic, caseorga_, bnvk, bnvk_ and fmarier joined the channel
#
benwerd
aaronpk: I like that very much. A simple "we are independent" directory might be a nice step, along the lines of that site
#
aaronpk
that's a great tagline
smus, jontyw, KevinMarks, othiym23, brixen and 31NAAM6I1 joined the channel
#
benwerd
weareindependent.com: ours at an affordable $2195. I guess I'm saved from another random domain buy.
smus joined the channel
#
aaronpk
how about indpndnt.com
#
aaronpk
or, .org
#
benwerd
weareind.ie?
#
@aral
Not that it needs saying, but I hope you guys know @aaronpk, @caseorganic, @t, @BarnabyWalters et. al. are doing awesome things w #indieweb
(twitter.com/_/status/395298623820677120)
#
aaronpk
the TLD as part of the word is cute, but rather hard to say
#
aaronpk
aw thanks aral!
#
@aral
…and IndieWeb, like Open Data, is so important for Indie Data that I can’t even imagine how much harder Prometheus would be to do w/out it.
(twitter.com/_/status/395299023273623552)
tantek joined the channel
#
Loqi
tantek: benwerd left you a message 1 hour, 26 minutes ago: Thanks for the Disqus note - I haven't altered my old WP install at all. I should probably deactivate it (all Disqus comments are copied to my local WP db)
#
benwerd
(And then I did, and spread the word)
#
benwerd
tempted to buy weareindie.tv and use it as a challenge to implement indie-video
#
_6a68
benwerd++ so good
#
Loqi
benwerd has 9 karma
#
benwerd
Man, I need to sort out that karma score. Must log into the channel more. (I finally finished my project that's been taking me away from idno this weekend!)
#
tantek
yay benwerd is back in the channel!
#
Loqi
giggles
#
benwerd
hoping to actually participate and build stuff again. been a while.
#
tantek
and dangit here I thought maybe I had a chance of catching up ;)
dybskiy joined the channel
#
benwerd
I've been following you all from the get-go ;)
#
tantek
hey aaronpk - any idea why FB keeps putting the OSBridge logo as the image preview for indiewebcamp URLs? e.g. https://www.facebook.com/ben.werdmuller/posts/10100632242049029
#
tantek
what's weird is that you even have the <meta property="og:image" content="http://indiewebcamp.com/wiki/skins/indieweb/images/indiewebcamp_logo_color.png" /> markup in there and it doesn't seem to help
#
tantek
so much for OGP metacrap
#
benwerd
apparently the og image needs to be at least 200x200
bnvk joined the channel
#
benwerd
so I guess they just arbitrarily ignore it otherwise
#
benwerd
while picking another image that is also smaller than that
#
benwerd
Oh! No. The OSB logo *is* bigger than that.
#
benwerd
(269x209.)
smus joined the channel
#
aaronpk
whoa that's really weird
#
@giyom
Imagine the #indieweb equivalent to banking: #indiebank i.e. ppl pooling deposits w/ open source software to drive resilient pymt innovation
(twitter.com/_/status/395304854488223744)
#
@tek_fin
RT @giyom: Imagine the #indieweb equivalent to banking: #indiebank i.e. ppl pooling deposits w/ open source software to drive resilient pym…
(twitter.com/_/status/395306469501190144)
paulcp, smus, loleg, lukebrooker, caseorganic, fmarier, glennjones, brianloveswords and LauraJ joined the channel
#
@NurtureGirl
RT @giyom: Imagine #indieweb = to banking: #indiebank i.e. ppl pooling deposits w/ open source software to drive resilient pymt innovation
(twitter.com/_/status/395325616192778240)
#
@kevinmarks
RT @NurtureGirl: RT @giyom: Imagine #indieweb = to banking: #indiebank i.e. ppl pooling deposits w/ open source software to drive resilient…
(twitter.com/_/status/395329092838375424)
#
peat
hah!
npdoty and smus joined the channel
#
@kevinmarks
"@morbital: @edbott somebody needs to start an "OpenTwitter"" #indieweb
(twitter.com/_/status/395334013637763072)
tantek joined the channel
#
@kevinmarks
"@mathewi: RT @kevinmarks: "@morbital: @edbott somebody needs to start an "OpenTwitter"" #India Web /cc @davewiner" er #indieweb #DYAC
(twitter.com/_/status/395335676893229056)
erikmaarten joined the channel
#
KevinMarks
though IndiaWeb may be worth campaigning for too
benward_ joined the channel
#
KevinMarks
@benwerd indie video has a chance via js
#
benwerd
via js?
#
KevinMarks
Monty joining Brendan at Moz is encouraging
paulcp joined the channel
#
KevinMarks
asm.js + WebGL = dyanamic codec download
lukebrooker_ joined the channel
#
benwerd
oh oh oh
#
benwerd
a JS codec
#
benwerd
the state of video on the web otherwise makes me weep internally on a daily basis
#
KevinMarks
that animated GIF is the de facto format depresses the crap out fo me
#
benwerd
the trick is still that point of upload - but you can possibly mitigate it if you *capture* the video right there using web technologies
#
benwerd
uploading a pre-existing video is likely still to be very hard
brianloveswords_ joined the channel
#
benwerd
animated GIFs were really cool in 1994 when I was making them, I swear
bretolius joined the channel
#
KevinMarks
we had quicktime fast-start then already
#
KevinMarks
by '98 we had it running on everything available
jgraham909 joined the channel
#
KevinMarks
netsplit? not seen that fro a while
#
_6a68
hey, animated GIFs are still cool. in fact, giphy has a whole series on GIF artists: http://giphy.com/artists/