#indiewebcamp 2013-04-01

2013-04-01 UTC
bnvk joined the channel
#
aaronpk
tantek: i think you're going to like this reply i'm working on :)
bnvk, jancborchardt, yarns1, yarns and bnvk_ joined the channel
#
Loqi
Ok, I'll tell him that when I see him next
#
Loqi
Ok, I'll tell him that when I see him next
tilgovi, hadleybeeman and tantek joined the channel
#
Loqi
Ok, I'll tell him that when I see him next
ddahl__, ianloic and tantek joined the channel
#
Loqi
tantek: aaronpk left you a message 56 minutes ago: here's another one for ya! http://aaronparecki.com/notes/2013/03/31/2/this-is-what-email-would-have-looked-like
#
@cstanhope
RT @t: conversations on email do not require creating new accounts. neither should conversations on the web. #ADN #indieweb (ttk.me t4PF2)
tilgovi joined the channel
#
tantek
aaronpk - awesome
#
tantek
well written follow-up
#
@benatkin
#freeculture #indieweb #ownyourdata
lmorchard, andreypopp, seyz, eschnou, andreypopp_, friedcell, portenkirchner, mxuribe, yarns and scor joined the channel
#
christopheducamp.com
created /pingback.me-fr (+1954) "[fr: translated from pingback]"
(view diff)
#
christopheducamp.com
created /pingback-fr (+1159) "[fr: translated from pingback]"
(view diff)
#
christopheducamp.com
created /webmention-fr (+361) "[fr: translated from webmention]"
(view diff)
andreypopp joined the channel
#
christopheducamp.com
edited /pingback-fr (+2) "/* Voir aussi */ [[Categorie:building-blocks]] changed in [[Category:building-blocks-fr]] ... is that ok ?"
(view diff)
scor, jedahan, andreypopp, friedcell, jzb, yarns and tantek joined the channel
#
christopheducamp.com
created /building-blocks-fr (+1719) "[fr: translated from building-blocks]"
(view diff)
#
christopheducamp.com
created /posts-fr (+7409) "[fr: translated from posts]"
(view diff)
#
christopheducamp.com
created /note-fr (+1369) "[fr: translated from note]"
(view diff)
#
tommorris
woo. go ChristopheDucamp!
#
christopheducamp.com
created /post-fr (+13) "{{:posts-fr}}"
(view diff)
#
christopheducamp.com
created /article-fr (+442) "[fr:translated from article]"
(view diff)
yarns joined the channel
#
christopheducamp.com
created /notes-fr (+12) "{{:note-fr}}"
(view diff)
#
christopheducamp.com
created /articles-fr (+15) "{{:article-fr}}"
(view diff)
#
christopheducamp.com
created /Semantics_Of_Article-Note_Distinction-fr (+2279) "[fr: translated from original]"
(view diff)
#
christopheducamp.com
created /photos-fr (+653) "[fr: translated from photos]"
(view diff)
#
christopheducamp.com
created /posts-elsewhere-fr (+2194) "[fr:translated from posts-elsewhere]"
(view diff)
#
christopheducamp.com
created /elsewhere-fr (+1040) "[fr: translated from "elsewhere"]"
(view diff)
seyz and barnabywalters joined the channel
#
christopheducamp.com
created /identity-elsewhere-fr (+513) "[fr: translation identity-elsewhere]"
(view diff)
#
christopheducamp.com
created /salmon-fr (+44) "[fr: translated from salmon - tbc]"
(view diff)
yarns joined the channel
#
christopheducamp.com
created /webactions-fr (+10930) "[first draft translation from webactions - á relire]"
(view diff)
#
barnabywalters
aaronpk: okay, I sent the PSR-0 pull request. ping me if there are any problems :)
#
aaronpk
woo! awesome
#
aaronpk
I saw you run the tests and send a pingback about github linking to indiewebcamp!
#
barnabywalters
I assume mentionclient is designed to be queued?
#
aaronpk
I don't assume that, but yes that's probably best
#
aaronpk
I'm only using it via a command line script right now, haven't built a queuing thing for my posting UI yet
#
barnabywalters
my pingback code isn’t, and is quite slow even though I make all the requests I can in parallel
#
aaronpk
yea, it will always be slow, definitely should be run in the background somehow
#
aaronpk
oh I should take the PK namespace
#
aaronpk
omg the saga continues. everything below here is new: https://alpha.app.net/brennannovak/post/4368142
#
barnabywalters
wow, alpha.app.net’s linking-to-individual-posts is really broken
#
aaronpk
yea it's super annoying
#
aaronpk
i see what they're trying to do...
#
barnabywalters
uh, what are they trying to do?
#
aaronpk
show the full conversation context
tantek joined the channel
#
barnabywalters
bah, I want the resource, not it’s context! :) And it scrolls (in safari at least) to *below* the targeted posts
#
Loqi
gives barnabywalters the resource
#
aaronpk
oh yea, I noticed it does that if you're not logged in
#
aaronpk
tantek: more follow-up on this thread! https://alpha.app.net/brennannovak/post/4368142
#
barnabywalters
it must be another marketing tactic!
#
barnabywalters
“sign up or we won’t scroll the page properly”
#
tantek
barnabywalters - lol
#
tantek
I'm seeing it in FF too - the "Want to join the conversation? Create an account >" banner covers the content
#
aaronpk
yea, I don't think they often look at their site when logged out :)
#
aaronpk
uh... "https://alpha.app.net/ryantharp/post/4369735" "I actually followed the instructions of his contact page, listing that IRC is his preferred method of communication and so far he refuses to engage me there." ... it was 1:30am when he posted this, I was asleep
#
aaronpk
also I don't see his message in any of my scrollback
#
christopheducamp.com
created /link-preview-fr (+3797) "[fr: draft translation of link-preview]"
(view diff)
#
aaronpk
wow I really pissed off a bunch of app.net users
#
tantek
anybody here from app.net?
#
barnabywalters
hah yeah — but the root point “dalton hasn’t made good on his promises” remains valid whatever they say
#
aaronpk
also people really seem to want a "country club" social network
#
tantek
aaronpk - it's interesting that in that follow-up thread the replies changed from defending openness to admitting app.net is a closed silo and being apologists for it
#
tantek
re: spam etc.
#
donpdonp
aaronpk: whats the expected route to comment on your post
#
aaronpk
donpdonp: i'm happy to talk about it in IRC as a conversation, otherwise it should be short or long notes on your own site :)
#
tantek
"minicloud" - lol. it's a server. nothing cloud-like about it.
#
aaronpk
I feel like I need a disclaimer, something to the effect of "yes I know setting up a blog and using pingbacks is hard right now, everyone at indiewebcamp is working to make it easier"
#
donpdonp
aaronpk: how about even an explanation of how it might work.
#
tommorris
I have an app.net account, but only because my stuff is syndicated in there
#
tantek
donpdonp - write a post (note/article/comment) on your own site, linking to the original with rel="in-reply-to", and have your webserver send aaronpk's server a pingback/webmention accordingly.
#
donpdonp
aaronpk: from looking at your post, id assume comments were unwelcome
christopheducamp joined the channel
#
aaronpk
donpdonp: hmm I should try to fix that
#
barnabywalters
funny thing is, setting up a blog and sending pingbacks takes about 30 seconds on wordpress.com
#
barnabywalters
not sure how long adding a custom domain to that wordpress.com account takes
#
aaronpk
donpdonp: the last sentence invites replies...
#
donpdonp
tantek: thx
#
tommorris
christopheducamp: thanks for all the FR translations on the wiki.
#
aaronpk
did you not see that or is that the thing that is unwelcome?
#
donpdonp
aaronpk: 'I would also like to personally challenge ...' ?
#
aaronpk
oh sorry, the follow-up post has that, not the original!
caseorganic joined the channel
#
christopheducamp
tommorris thx. rager to test pingback.me
#
donpdonp
aaronpk: ah.
#
christopheducamp
oops. eager… :)
#
donpdonp
aaronpk: much better :)
#
aaronpk
i should add that to the footer of all my posts under a "Comments" header :)
#
donpdonp
aaronpk: yes indeed
#
tommorris
when people say "federation", I have absolutely no clue what they are supposed to mean
#
barnabywalters
aaronpk: it’s interesting how many people think you’re suggesting a federated social network
#
barnabywalters
and then bang on about spam and friends not being on there
#
barnabywalters
might be worth pointing to the why-not-federated page on the wiki
#
barnabywalters
donpdonp: sandeepshetty
#
barnabywalters
he wrote the webemention spec
#
donpdonp
barnabywalters: kthx
#
aaronpk
did I really come off like this? "I want this and it's not there yet. Therefor I declare ADN a failure."
#
tantek.com
edited /comment (+442) "add How To section"
(view diff)
#
barnabywalters
aaronpk: I certainly didn’t read it like that, but twitter/adn are hotbeds for that sort of snarky nonsense
#
aaronpk
thank you, someone else gets it "I disagree. I don't see it as him wanting ADN to serve his purpose. I see it as wanting to enact a purposeful change. The ability of not needing an ADN account to interact is great. The email approach to a social network would be awesome" (@aj)
#
barnabywalters
quick poll: how many people here publish XFN, and if so: where, and how detailed?
#
tommorris
barnabywalters: no, but I should. ;)
#
barnabywalters
ftr I do, to a fair level of detail: http://waterpigs.co.uk/contacts
#
barnabywalters
I want to build a simple XFN crawler and service to (hopefully) prove to people that “federated” spam can be eliminated fairly easily
#
barnabywalters
but need some actual data :(
#
Loqi
barnabywalters meant to say: but need some actual data :)
#
tantek.com
edited /comment (+412) "/* How To */ your server should send pingbacks/webmentions automatically, an IndieWeb Reply button would help with automating linking to originals"
(view diff)
#
tommorris
barnabywalters: in RDFland, we used to have it so if you were within two-hops of TimBL, you were trusted on the wiki. ;)
#
tantek
barnabywalters - I used to keep much more detailed XFN. Now I use it sometimes in people mentions in blog posts.
#
donpdonp
i have rel="me" links. thats as useful as I found XFN to be.
#
aaronpk
barnabywalters: I have no plans to make a contacts list like you, but I would use XFN in individual blog posts. still haven't yet
#
tantek
donpdonp - here is a URL answer to your question about how to comment: http://indiewebcamp.com/comment#How_To
#
donpdonp
aaronpk: i like the 'people buttons' on your post
#
tantek.com
edited /comment (+9) "/* How To */ replying"
(view diff)
#
donpdonp
tantek: got it. thx
#
barnabywalters
tommorris: oh brilliant! that’s exactly what I want to implement — great to know this has been done before. Why is it no longer in use — were there any problems with it?
#
tommorris
see, the problem I have is that I really don't like "blogrolls". they kind of had some stink around them because of don't repeat yourself
#
tommorris
I'd much rather plonk all that on a separate page and link to it
#
tantek
barnabywalters, XFN was designed for blogrolls and blog post mentions of people
#
tommorris
rather than have the same 20-30K of links on every single post
#
tantek
overtime blogrolls became culturally less popular
#
barnabywalters
tommorris: yeah, I’m not a fan of blogrolls on the homepage/every page
#
barnabywalters
so I have my big full contacts list, and then XFN and h-card on mentions of people in my notes
#
tantek
tommorris - blogrolls didn't have a DRY problem. FOAF did (invisible meta file duplicate of already visible data).
#
tantek
blogrolls were supposed to reflect current state
#
tommorris
tantek: by DRY in this case, I mean that every single post had the same sidebar. the reader was downloading the same blob of HTML over and over again
#
tantek
and be only on the home page, not on every archive page
#
tommorris
and that leads to page bloat
#
tantek
tommorris - that's just a bad template
#
tommorris
also, there's a bit of a mismatch. blogrolls are "these people are worth reading", which is different set than "people who I know/trust/am related to/work with/would fuck/would vote for/whatever"
#
barnabywalters
which in itself has many interesting uses — e.g. XFN on notes + feedreader which accepts XFN as input = feed from people I’ve communicated with in the last few days
#
tantek
tommorris - not true - people would publish lists of friends, relative etc.
#
barnabywalters
tommorris: that is my problem with the “blogroll” terminology
#
tantek
before XFN
#
tantek
XFN simply modeled their behavior
#
tommorris
I'm not sure that they still would though
#
tantek
every value in XFN is based on finding many many blogs with explicit mentions of people with said relationships
#
tantek
it's just that the behavior changed over time
#
tantek
blogs shed blogrolls
#
barnabywalters
were the blogrolls maintained manually/separately to other contacts info?
#
barnabywalters
adactio’s list of “people I have let into my house” is an interesting take on the blogroll
#
tantek
aaronpk - hopefully this documentation is straightforward enough that we can all share it on our posts: http://indiewebcamp.com/comment#How_To
#
tantek
barnabywalters - yes, blogrolls were typically separately maintained
#
tantek
everyone did things slightly differently
#
tantek
some used them just as reading lists (as tommorris pointed out)
#
tantek
but even with those, you could annotate a subset of those with XFN relationships
#
tommorris
wonders whether pointing people to wordpress as a way to respond might help them on that front.
#
tommorris
although Wordpress isn't your-own-domain
#
aaronpk
wordpress can be I thought
#
aaronpk
tantek: awesome, I will link to that from my site
#
barnabywalters
well, it can be, at less expense (i believe) than ADN
#
tommorris
aaronpk: not by default
#
aaronpk
is it a paid upgrade?
#
aaronpk
well these people all paid $50 or $100 to get on app.net, so I think they can handle another $10 for their domain :)
#
tantek.com
edited /comment (+334) "/* How To */ Supporting Software"
(view diff)
#
tantek
tommorris - I pointed out that you can use Wordpress (software or .com) to post comments in this way
#
tommorris
excellent
#
barnabywalters
the rel=in-reply-to thing — who here is publishing that right now?
#
aaronpk
barnabywalters: I think my replies list does
#
tantek
did we decide rel=in-reply-to or class? I forgot
#
tantek
(too early and I haven't had coffee)
#
aaronpk
I don't remember a resolution on that point... thought it was still up in the air
#
barnabywalters
aaronpk: they have rel=previous-in-thread, which is interesting
#
aaronpk
ah that must be what I found on the microformats wiki
#
barnabywalters
last discussion I had with tantek resulted in me using u-in-reply-to
#
tantek
barnabywalters - thank you
#
tantek
inside the h-entry presumably
#
tantek
ah, in-reply-to comes from Atom
#
tantek
sort of
#
tantek
just researched it a bit further and found out that it's more complicated than that
#
tantek
however it's worth proposing
#
tantek
bbiab
#
christopheducamp.com
edited /why-fr (+5471) "[fr: translation sync'd with original]"
(view diff)
#
aaronpk
now dalton is chiming in on the thread
#
barnabywalters
aaronpk: ooh, cool — link?
#
aaronpk
hmm now I feel a little bad about "the fact that this was written as a blogpost rather than "engaging" to ask questions first via various mediums is not something I appreciate." -dalton
#
barnabywalters
true, although I suppose the whole point is that there should be no difference between the two :)
#
aaronpk
well, public vs private I guess
tantek joined the channel
#
christopheducamp.com
edited /why-fr (+412) "[/* Concentration sur le Positif */ quotes translated - draft]"
(view diff)
scor and bnvk joined the channel
#
christopheducamp.com
created /comment-fr (+2830) "[fr:translated from comment]"
(view diff)
andreypopp, tantek and caseorga_ joined the channel
#
tantek.com
created /comments (+21) "rr"
(view diff)
#
tantek
checks the archives
#
tantek
aaronpk, yeah, a blogpost *is* a medium of engagement, that's the point
#
tantek
blog posts *are* conversations
#
tantek
that's what we believe, that's how we act
#
tommorris
urgh, nginx configuration files aren't an improvement on what went before
#
tantek
bnvk: "I would love to be included on that email thread as well " <-- this is why blog posts about this make more sense
#
tantek
there are lots of of us that would "love to be included on that [conversation]"
#
aaronpk
hah! didn't even see that yet
#
aaronpk
nice, yea that drives the point home well
#
bnvk
tantek: yes, I almost included that in my response "but this is point, we should all be able to reply like we do in email"
#
christopheducamp.com
edited /Getting_Started-fr (+2309) "[fr: translation sync'd with original]"
(view diff)
#
bnvk
but, the reason I did not, as I really don't think all conversations should *always* be public all the time. opposing views (hot heads and trolls) get involved and add too much noise- hence a more private method like email is needed
#
bnvk
IMHO
#
tantek
bnvk - however, anything about open standards, and implementing them, should be an open conversation by default.
#
bnvk
at all points in time?
#
tantek
*default*
#
tantek
if you want to change from the default, you need to justify it with a VERY good reason (or set of reasons)
#
tantek
this is a crap excuse "hot heads and trolls" - as blog posts are far higher quality than that typically
#
tantek
threads on twitter or app.net OTOH have plenty of name-calling and other ad-hominem attacks
#
tantek
like all the people attacking aaronpk *personally* rather than debating his arguments
#
tantek
I would just block all of them - they're not adding anything constructive
#
bnvk
See, that's my point
#
tantek
what, that app.net has low quality responses?
#
bnvk
the main "voice" in the thread on ADN who responded to Aaron's article (and added very little productive value to the convo) was Darnell
#
tantek
I think it's right to have conversations about open web conversation standards using our cutting edge open web conversation tools
#
tantek
because the folks commenting on silo posts aren't really being (that) helpful
#
christopheducamp.com
edited /Getting_Started-fr (+8) "/* Portez le vieux des silos vers votre propre site */ typo"
(view diff)
#
tantek
it's like, if you can blog about it on your own site, then clearly you have a better idea of what's going on
#
tantek
if all you can do is write a short tweet/app.net post about it, you're likely to be an armchair commenter that doesn't get it and/or doesn't really care about being helpful, but prefers short-form snark (as evidenced by those app.net threads)
#
tantek
the problem of private/closed/email conversations is that they inevitably leave out constructive contributors who then may feel ignored/neglected, and worse, whatever private email club you make up will almost certainly miss good ideas from the broader set of folks working on open web things
#
bnvk
quality of responses in general is challenging
#
tantek
bnvk - I disagree - quality of responses on blog posts is very high
#
tantek
as in, responses written as blog posts
#
tantek
long form writing forces deeper and more considerate thinking
#
bnvk
I remember when the first Federated Social Web summit happened
#
tantek
same here
#
bnvk
I followed the threads on Google Groups. How would you describe the quality of conversations there?
#
aaronpk
(ignore me)
#
tantek
Google Groups? are you kidding me? that's just another email list.
#
tantek
public email lists are even worse than twitter/app.net threads
#
bnvk
Hah. Ok. But it was "open" in that sense that anyone who wanted to could participate
#
tantek
sure - anyone with an email address
#
bnvk
Sure. But the nature of the conversation was not "private" and closed and limited. Is that fair to say?
#
tantek
= too low of a bar, = worse than threads on twitter or app.net
#
tantek
I said open, with blog posts
#
tantek
no one is advocating for yet another noisy email list
#
bnvk
Can you give me an example of where positive "open" conversations strictly limited to blog posts led to solid products which people use?
#
tantek
bnvk - do you have your blog posts from Federated Social Web Summit 2010?
#
bnvk
I didn't actually attend it as it was "closed" and invite only and no one in the community knew me yet.
#
bnvk
cool
#
tantek.com
edited /selfdogfood (+128) "posts"
(view diff)
#
tantek
bnvk - right about the being "closed" part
#
tantek
which is why we organized IndieWebCamp as open but having to meet openly stated indiewebness requirements
#
tantek
and we got amazing new people participating
#
bnvk
Right
#
bnvk
All I'm saying is @aaronpk intentionally made entry to IndieWebCamp limited to people who can setup their own domain first with OpenID, then IndieAuth
#
tantek
that's perfect contrast set of examples of why open with an open bar to meet is better than closed/private/invite-only
#
tantek
bnvk - that was a design decision that I specifically chose for IndieWebCamp
#
tantek
and it filtered quite well - as in, everyone who showed up was a productive contributor
#
bnvk
that is by design (to some degree) exclusionary, right?
#
tantek
we had ZERO "hot heads and trolls"
#
tantek
right, we're excluding those that won't put enough time/energy into their own site
#
tantek
because if they're not willing to put that time/energy into their own site, they're unlikely to add anything but hypotheticals/theoreticals to the conversation
#
bnvk
yah, I loved that about IWC vs. the FSW invite style registration
#
tantek
and we've already had too many years of mailing lists drowning with useless circular hypotheticals/theoreticals
#
bnvk
I see that intention of filtering out attendees very similar to saying "lets email a little bit about this before we make public claims"
#
tantek
bnvk - totally different
#
bnvk
how so?
#
tantek
since *anyone* can read our work
#
tantek
our discussions
#
tantek
everything was openly readable by anyone on a browser since day 1
#
tantek
we put all our notes on open etherpads
#
tantek
and heck, the folks that couldn't be bothered to setup their own sites could even jump on IRC and etherpad if they wanted to
#
tantek
but we made it very clear that it was two tier in that way
#
tantek
either you're a serious indieweb/socialweb creator - and are proving it with your own site
#
tantek
or you're not serious, and you're a peripheral contributor
#
bnvk
Yah, I think it is all super great- what we're doing here, which is why I identify with this tribe and participate
#
tantek
that kind of "class" distinction is explicit and new about indiewebcamp
#
tantek
since it's not by who-knows-who
#
tantek
but by what-have-you-done
#
bnvk
But, I also don't think there needs to be one (and only one) true approach to solving these complicated problems
#
tantek
and it goes against the assumed egalitarian nature of all things internet
#
tantek
since we're explicitly stating there are two tiers
#
tantek
creators and everybody else
#
tantek
bnvk - that's why indiewebcamp welcomes multiple approaches
#
tantek
you have read the principles page right?
#
bnvk
Dalton, et. al. is working hard and building something I think is valuable. And MOST importantly, he is totally willing to engage and try to work towards our causes- I'm willing to meet him on his terms
#
bnvk
which (at the start) is email
#
tantek
some of us have already wasted enough time on email that we're done doing so
#
aaronpk
bnvk: indiewebcamp registration is not making the event private and is not exclusionary
#
tantek
especially since there's now a critical mass of us using the open social web to discuss and create and iterate on the open social web.
#
aaronpk
it's an intentional bar you have to hurdle
#
tantek
there's a critical mass of us that have moved past email
#
bnvk
Fair, you have every right to not want to engage with people who want to communicate via email
#
tantek
it's like, I also don't engage with people who want to communicate via fax machine
#
tantek
and increasingly, via phone call
#
tantek
we've got better tools now, why bother with email?
#
bnvk
aaronpk: yes, "an intentional bar you have to hurdle" which is such a hurdle for so many people, they see using Facebook / Twitter as a better option
#
tantek
bnvk - people that see using Facebook / Twitter as a better option obviously don't care about the indieweb
#
tantek
so why do you want to engage them in discussions? they're unlikely to be productive.
#
tantek
you might as well go talk to random people in a coffee shop (most of whom are likely on Twitter and FB)
#
bnvk
I'm not talking to people on those networks about building IndieWeb infrastructure
#
tantek
then I don't understand what point you're trying to make - please restate it.
#
aaronpk
right, and talking to people who don't want to use IndieWeb infrastructure about building IndieWeb infrastructure is not productive
seyz joined the channel
#
tantek
we're not advocating a particular IndieWeb infrastructure - we're advocating *any* IndieWeb infrastructure that they want to use/create/build
#
bnvk
I dunno you guys, you might be completely right- engaging with Dalton and them (via email) might be a total waste of time,
#
aaronpk
i'll respond to his email if he sends one... but my response will likely end with "can I post this online?"
#
bnvk
If you both see absolutely no value added by App.net thus far, I can understand where you're coming from
#
bnvk
I think what he meant was "I wish Aaron had asked me some questions (in any platform) before he wrote a big blog post slamming what we're doing here"
#
bnvk
or "not doing" rather
#
aaronpk
ok I probably could have phrased that post a little more gently
#
aaronpk
but still my main points are there
#
bnvk
sure, I agree with your points 100%
#
bnvk
they have NOT delivered those things yet
#
bnvk
aaronpk: out of curiosity did you hear about (and look into) https://directory.app.net/app/118/customwebapps-pubsubhubbub-for-adn
#
aaronpk
I havne't looked at it yet
#
christopheducamp.com
created /web_hosting-fr (+1354) "[fr: translation of web-hosting = Hébergement Web =]"
(view diff)
#
bnvk
tantek: out of curiosity, did you participate in any of the Occupy activities last year?
#
tantek
I donated and helped donate things to Occupy
#
bnvk
cool
#
tantek
e.g., when BarCamps/ScienceHackDay etc. had extra food, we dropped it off at Occupy
#
christopheducamp.com
edited /How_to_set_up_web_sign-in_on_your_own_domain-fr (+165) "[fr: translation sync'd with original + added title]"
(view diff)
#
tommorris
we should change Indie Web Camp → Occupy Web 2.0
#
Loqi
definitely
#
christopheducamp.com
edited /How_to_set_up_web_sign-in_on_your_own_domain-fr (+4) "[fr: Comment régler une Connexion *Web* sur Votre Propre Domaine]"
(view diff)
#
tantek
"see absolutely no value added by App.net thus far" - meh, it's another silo experiment, seeded with a payfor model.
#
bnvk
fair
#
tantek
app.net has nothing to do with the open social web
#
bnvk
also fair
#
tantek
but they claimed that they would
#
tantek
lots of promises up front
#
tantek
aaronpk is just calling bullshit on all that
#
tantek
I see no problem with that
#
bnvk
fair, also, they've only existed for like 7 months and have created a quite sizable developer community
#
tantek
why would you develop on app.net instead of just developing for the open web?
#
tantek
do any of these developers have/use their own personal web sites?
#
bnvk
they created this really cool micro funding (< $20,000) monthly pot that enables developers to get paid to do development- but the process of who gets funded is voted on by the community
#
tantek
ok so that's interesting then
#
aaronpk
that would be even cooler if it was related to an open/indie web
#
tantek
is there a URL describing that process?
#
tantek
aaronkp - we can emulate that same model
#
tantek
we could setup a way to make indiewebcamp donations
#
aaronpk
heh yea true
#
tantek
and then have the community of indiewebcamp participants vote on it. one personal URL, one vote ;)
#
bnvk
I think the way it was designed was so any developer COULD develop as per whatever standards / protocols they chose
#
tantek
bnvk - that's laughable
#
bnvk
how so?
#
tantek
because NO developers can develop for app.net using existing open standards / protocols
#
tantek
that's the point of aaronpk's post
#
tantek
there's no "choice"
#
tantek
there's app.net's snowflake proprietary API - just like every other proprietary silo
#
tantek
you don't have a choice. you have app.net snowflake proprietary API or nothing.
#
bnvk
I believe the few attempts towards our IndieWeb integration like https://directory.app.net/app/118/customwebapps-pubsubhubbub-for-adn/ was met with such little interest / understanding of the value it dired up
#
tommorris
it's not just that it's a snowflake API, it's a snowflake API you have to pay for... because the only people who might want to use an API are people who are building "apps"
yarns joined the channel
#
tommorris
by which they mean custom UIs for different devices that hope to turn a profit
#
tommorris
rather than people pissing around
#
bnvk
Mmmm... I don't think that is at the core of their intentions / goals
#
bnvk
"turn a profit"
#
tantek
bnvk - who said anything about intentions / goals?
#
tantek
you said "designed was so any developer COULD develop as per whatever standards / protocols they chose". so that's obviously not true. there's no "could" and there's no way to "chose".
#
bnvk
tantek: unless I misunderstood tommorris: just did
#
tommorris
so, app.net is just twitter with picket fences to keep the lower orders at bay.
#
bnvk
It just different than that, but I see how it appears that way
#
bnvk
what I meant by that was "if I (a developer) wanted to build some sort of syndication hub / messaging que / magic spaceship that had complete interoperability with IndieWeb services as well as App.net, a developer COULD create such a thing"
#
aaronpk
tommorris: that's how that comment thread makes it come off right now, yea
#
tommorris
bnvk: oh sure. you can do that.
#
tantek
bnvk - how is it any different than a pay-for twitter? write a blog post because plenty of smart people don't see it.
#
bnvk
and THEN said developer COULD get community approval and votes on said piece of technology as well as a little bit money to pay bills while developing it further
#
tommorris
I'm sticking with ifttt. if app.net want to up their game in terms of standards compliance, I'll start giving a fuck. until then, I'll just syndicate out there using ifttt. if that breaks, fixing it will be fairly low priority.
#
tantek
ok so it has a crowd-funding component
#
tantek
URL to description of the crowd-funding component?
#
tantek
seriously, if this stuff isn't obviously documented on the web, WTF?
#
bnvk
looks for best link
#
tantek.com
created /App.net (+722) "stub yet another silo"
(view diff)
#
tantek
bnvk - feel free to edit / add to: http://indiewebcamp.com/App.net
#
tantek.com
edited /App.net (+117) "POSSE to"
(view diff)
#
tantek
tommorris - mind describing how you use IFTTT to POSSE to App.net? http://indiewebcamp.com/App.net#POSSE_to_ADN
#
bnvk
where is the page on indiewebcamp.com that has the IRC logs?
#
bnvk
thanks
#
tantek
wonders how long it will take before the IndieWebCamp page on App.net hits the first page of google search results for app.net
#
tommorris
tantek: will do
#
bnvk
Yep. Realized that right after you typed it arronpk: I had only browsed the IWC interface and didn't find the link
#
aaronpk
hm /me wonders if that should be a link in the sidebar
#
aaronpk
new section "Communicate" or "Contribute" or "Join the Conversation"?
#
aaronpk
with links to IRC, comment, others?
#
bnvk
I'm kind of obsessed with this idea that most problems humans face are really just an issue of race conditions re: the distribution of the right information at the right time
#
tantek
"Join the Conversation" lol
#
bnvk
aaronpk: I like that idea
#
bnvk
I just invited Dalton to join us in here :)
#
tantek
bnvk - nice
#
aaronpk
publicly?
#
bnvk
no, in a DM
#
tantek
bnvk - btw - the response threads on app.net to aaronpk's posts prove that app.net's pay-to-participate model is no longer keeping out trolls
#
bnvk
tantek: I agree 100%
#
tantek
but hey, if you can make money off the trolls, more power to you ;)
#
tommorris
is bnvk an app.net-er?
#
tommorris
may have missed that ;)
#
bnvk
tommorris: I am formerly on IRC @brennannovak a proud IndieWeb dog fooder, but I was an early supporter of App.net http://bnvk.me/uK6
#
bnvk
oops, bad grammar, apologies ;)
#
tommorris
ah okay
#
bnvk
tommorris: I think we met at IWC IRL :)
#
tommorris
has memory like a sieve.
#
tommorris
Brighton?
#
bnvk
I think it was Portland in 2012. You're a Googler dude right?
#
tantek
bnvk, tommorris - you two have not been at the same IWC yet
#
bnvk
tantek: thanks, you with a memory a steel trap ;) who am I thinking of then?
#
tantek
bnvk - my extended replicated memory helps with that ;)
#
tantek
bnvk - there were several googler's at IWC 2011
#
tantek
googlers even.
#
bnvk
wasn't there one at 2012 as well?
#
tantek
Chris Messina, Will Norris at IWC 2011
#
tantek
and Tab Atkins too
#
tantek
and at IWC 2012, Brett Slatkin
#
tommorris
bnvk: ah, there's another tommorris who works at Google. incidentally, on semantic web stuff too.
#
tantek
he'll likely show up at 2013 too
morrocco_mole joined the channel
#
tommorris
hadleybeeman can testify to the problem of tommorris-vs-tommorris-at-Google mixup.
#
bnvk
Yah, Brett Slatkin, that's who I am thinking of. And I recently read the Tom Morris (at Googles) stuff about semantic web- I figured this was the same guy
#
tantek
tommorris - have to find it a bit funny that a semantic web person is having an identity conflict, with someone at Google no less
eschnou joined the channel
#
tantek
hence, URLs! :)
#
tommorris
indeed.
#
bnvk
which tom morris owns http://tommorris.org ?
#
bnvk
ah it's the one in here :)
#
hadleybeeman
tommorris, you're both very different people. But the challenge comes when you send emails around the same time, and my mail client only shows your names!
#
tommorris
bnvk: that would be me. ;)
#
tommorris
so, I'm starting my personal wiki to go with my blog
#
tantek
hadleybeeman, clearly you should configure your email client to show the URL of the sender ;)
#
tantek
tommorris - yeah? what wiki software? or are you writing your own?
#
tommorris
MediaWiki
#
christopheducamp.com
created /Why_web_sign-in-fr (+9344) "[fr: draft trranslation to be cont.]"
(view diff)
#
hadleybeeman
Yes, this is probably a solvable problem, tantek. :) On the other hand, that would take away the surprise for me!
#
tantek
tommorris - impressive
#
tantek
(and makes sense given your background/expertise)
#
tantek
how are you going to deal with the MySQL problem?
#
tommorris
well, I just ended up installing MySQL
#
tommorris
which is a bit annoying as I have Postgres installed on the same server for my blog
#
hadleybeeman
tommorris, that's brilliant. I tried a remotely hosted one at one point, and had trouble with it. I love the idea though.
#
tommorris
the difficult bit was getting PHP-FPM setup. haven't done that for a while
#
aaronpk
MW dosen't support postgres?
#
tommorris
it does, but it's "community supported", which really means "not supported" ;)
#
tommorris
one silo solution for real-life-namespace conflict that I saw was a plugin for Gmail that would pull details from Facebook and elsewhere and display it in a sidebar
#
hadleybeeman
tommorris, are you playing with Semantic MediaWiki at all these days?
#
tommorris
ah yes, http://rapportive.com/ was what it was called
#
hadleybeeman
(or are any of you?)
#
tommorris
hadleybeeman: not really. wikidata is the new hotness. ;)
#
hadleybeeman
wikidata is software? I thought it was just... data.
#
tantek
never got what was "semantic" about Semantic MediaWiki - looked like just a bunch of forms/templates stuff - no actual semanticness / output.
#
hadleybeeman
(+1, tantek. It was going to be my next question)
#
hadleybeeman
I've spent a lot of time lately fighting with smw on w3.org. Was just curious.
#
tantek
classic case of someone ironically misnaming something "semantic"
#
hadleybeeman
tommorris: :p
#
tommorris
SMW seems very kludgey and unintuitive.
#
tommorris
but I don't need the semanticy stuff on my personal wiki. it's a public notetaking place for myself, and it's a way to try and not use Google Docs/Dropbox/whatnot for stuff that should be on the web
#
tantek
has been trying to figure out how to wikify his existing static pages
#
tantek
as in, what are the essential differences between a static page and a wiki page
#
tantek
is it just: a) editable, b) versioned, c) browsable version history, d) browsable versions ? anything else?
#
tommorris
pretty much.
#
hadleybeeman
multi-authored, usually. Though that may not be useful for you
#
tommorris
well, the multi-author bit is negotiable for me
#
tommorris
like, some pages will invite contribution, others will be locked down
#
jedahan
whoops, wrong channel
#
hadleybeeman
If we're making a requirements wishlist, I'd love a multiplatform interface. Something like evernote in terms of input, but accessible on the Web like a wiki. With tagging.
#
tantek
hey jedahan, welcome to #indiewebcamp. Are you coming to IndieWebCamp 2013?
#
tantek
web standards based UI = multiplatform interface
#
jedahan
I really wanted to but i think it conflicts or costs to much for me to fly and stay somewhere coast to coast
#
tantek
are you one the east coast?
#
jedahan
i've been trying to save for a larger trip
#
jedahan
i am going to be in portland soon though for a conference work is sending me to
#
jedahan
so I'd love to get drinks
#
jedahan
with anyone in the indiewebcrew
#
tantek
so multi-author for a wiki is *typical* but I think framed within the current discussion, that is, a *personal* wiki, multiauthor seems like a post 1.0 type feature
#
tantek
it's not like we worry much about multi-authored blog posts on our personal sites
#
aaronpk
jedahan: let me know when you're around! I'd love to meet up!
#
tantek
aaronpk - yeah! when other indiewebcampers visit town you should organize indieweb drinkups or something! Maybe Green Dragon (now that beerandblog is defunct)
#
hadleybeeman
I'm happy with that. Also, for anyone paying attention to the new UK regulatory frameworks, it should exempt you from needing membership in the Leveson body.
#
tantek
wonders if indiewebcamp could take over beerandblog seeing as there's a lot of overlap
#
christopheducamp.com
edited /Why_web_sign-in-fr (+134) "/* Mais les e-mails sont très largement compris */"
(view diff)
#
tantek
tommorris - so ok, back to personal wiki and storage
#
jedahan
I definitely will aaronpk I'm gonna be in portland from the 14th to the 21st
#
tantek
I'm actually thinking of, how do I just use a static HTML page itself as the storage for a "wiki" page
#
tantek
and I think with versions stored elsewhere (e.g. the filesystem) it may be possible
#
tantek
all it might take is a date-time-stamp suffixing convention, and a new link rel to link to the previous historical version
#
tantek
that would satisfy b,c,d
#
tantek
and for (a), editability, it might be a neat trick to write software that made any static HTML page on your site automatically "editable" if you happened to be signed in with Web Sign-in / IndieAuth
#
tommorris.org
edited /App.net (+195) "/* POSSE to ADN */ explanation"
(view diff)
#
tantek.com
edited /App.net (+12) "/* POSSE to ADN */ link posse"
(view diff)
#
aaronpk
jedahan: cool, I will be back by then! ping me when you're around!
#
aaronpk
tantek: beer and blog is pretty much dead, but something is ready to take its place!
#
tommorris
universaleditbutton was such a horrendous hack of the type attribute. ;)
barnabywalters joined the channel
#
tantek
just trying to think of ways to make a page editable without putting any meta/link in the page itself
#
tantek
e.g. with HTTP LINK headers
#
tantek
it should be possible to use HTTP LINK rel=edit to an editing script (for browser plugin discovery)
#
tantek
and perhaps an HTTP LINK rel=script to execute an editing script (for browsers that don't know about rel=edit)
#
tantek
those are my current indieweb wiki brainstorming thoughts
#
tantek
1) static HTML page as its own storage
#
tantek
2) *-date-time suffixed versions of said page in the file system (or perhaps in the YYYY/ directories), linking to latest version and previous versions with rel values
#
tommorris
mediawiki has implemented http://www.mementoweb.org/ - W3C's attempt at magicking a "find old versions of this page" standard.
#
tantek
3) HTTP LINK headers for browser-editing-plugin discovery and javascript-based fallback
#
tantek
tommorris - last I looked at it - memento was horribly overdesigned
#
tommorris
ideally, it should be possible to just use <link rel="[some value we haven't coined yet]" to link old versions together
#
tommorris
so one can hit page X, and then follow one's nose to older versions
#
tommorris
perhaps "older" fits as a nice value for that. ;)
#
tantek
tommorris - that's exactly what I mean
#
tantek
when I said "a new link rel to link to the previous historical version"
#
tantek
rel=previous is for sequential navigation, not historical versioning
#
tantek
so we can't use that
#
tantek
it's a different meaning
#
tommorris
so, one day, I'll be able to hop into my DeLorean, swivel some dials on a dashboard like this - http://i.telegraph.co.uk/multimedia/archive/01674/future-date_1674140c.jpg - and see the old web. ;)
#
tantek
tommorris - the memento stuff has way too many rels, none of them actually useful: original, timegate, timemap
#
hadleybeeman
Yeah, the versioning topic is messy. It was originally in our charter for the Gov Linked Data working group, and we ended up not having sufficient time/resources to tackle it properly.
#
tantek
tommorris - re: UIs, I like the slider in Etherpad
#
tantek
I'm imagining something like that or what archive.org puts up in the header on archived page results
#
tantek
hadleybeeman why is versioning messy?
#
tantek
it seems like people are making it a lot harder than it should be
#
tantek
tommorris's suggestion of rel=older may be sufficient
#
tantek
though "older" feels wrong as a noun
#
tantek
I really want a noun that means "the previous revision"
#
tommorris
rel="supersedes"
#
tantek
rel=revision ?
#
tommorris
the current doc supersedes the previous doc
scor joined the channel
#
tantek
tommorris - we already have rel=canonical for that
#
hadleybeeman
tantek: Well, when you're talking about versioning a dataset, the policy part is messy. How do you deal with changes in individual values, vs the overall dataset? Updates vs next edition of a dataset? Changes in metadata (at what point is it no longer related to the original?) etc
#
tommorris
if we still had rev, we could do precedes. ;)
#
tantek
hadleybeeman, what's this "dataset" you speak of? I'm talking about one URL and a previous version of what was at that URL.
#
hadleybeeman
But for what you're talking about, an entire post, then I think the indie approach is great. Jump in and create the behaviours.
#
tommorris
hadleybeeman: so the answer to that is to have some kind of zen mind-meld and see that there is no datasets, there are just resources within the dataset. ;)
#
hadleybeeman
That was deep, tommorris. :)
#
tantek
hadleybeeman, what metadata?
#
tantek
there's just the HTML document
#
tommorris
a dataset is just a strange thing we do with data where we clomp it together into zip files and distribute it to people because we haven't put it on the web properly.
#
tantek
tommorris - that sounds like obsolete technology
#
hadleybeeman
It's definitely legacy, I agree.
#
tommorris
tantek: so, hadleybeeman works on government data, which tends to be mahoosive CSVs with a blob of metadata (who created them, which government department, when they were collected etc.)
#
tantek
is this like communicating with folks that still use fax machines or MS Word or PDF?
#
tantek
ah ok
#
tantek
well we'll solve the personal use-case first
#
tantek
and let others see what they can do about org-based use-cases
#
hadleybeeman
Yes, please do!
#
tommorris
versioning those is harder because the semantics of what a "new" version is are often difficult to agree
#
hadleybeeman
That will be of value to us.
#
tantek
so personal use-case, the HTML file is the boundary
#
tantek
there is no "meta"
#
tantek
either you changed something in the file or you didn't
#
tommorris
so, imagine the slider in Etherpad hooked up to one of these - http://store.griffintechnology.com/powermate
#
tantek
tommorris, I don't think you need rel=supercedes
#
tantek
all we need are two rels
#
tantek
one for - here is the current version over here
#
tantek
and another for - here is the previous historical version here
#
tantek
ironically this is what the visible links at the top of nearly every W3C spec link to
#
tommorris
yep, I was gonna say W3C specs do that pretty nicely.
#
tantek
so all we need are rel values that reflect those already visible semantics
#
tantek
tommorris - well they do it in an ok way in the visible markup
bnvk joined the channel
#
hadleybeeman
traditionally, the metadata for a blog post would be date of publication, tags, and author. (Though we're nixing author here, right?) But it sounds like you're treating that as content, tantek?
#
tantek
there are no discoverable semantics on those links though
#
tantek
you'd have to scrape for english words like previous / current and then look for nearby hyperlinks
#
tommorris
wonders if there is a simple way to get a list of all of one's friend's websites from the silos (FB/Twitter)
#
tommorris
preferably without touching APIs
#
tantek
hadleybeeman - traditionally date of publication, tags, and author are all visible in a blog post - thus they're data. nothing "meta" about it.
#
tantek
yes, treating it as content
#
tantek
just because things are properties does not make them "meta"
#
tantek
IMO this is a huge misconception (over-generalization) of "meta"
#
tantek
(and I blame SemWeb culture/community for this confusion)
#
tommorris
oh, no, practical semweb people hate it too.
#
hadleybeeman
Well, meta can be a relative label. What is metadata in one context is data itself in another.
#
tantek
hadleybeeman, perhaps it was never meta, except in someone's misconception
#
tommorris
there's plenty of stuff that's attributed to semweb people that I really hate. metadata/data distinction is one. and the idea that the world needs 'controlled vocabularies' is another.
#
tantek
tommorris that's because the same people that self-label as semweb also overuse the term meta
#
hadleybeeman
Do we really want to get in to the nature of metadata here? :) I'm taking your point that you want to consider what I've called metadata at the content level "content" for the purposes of what you're doing.
#
hadleybeeman
Which makes sense to me.
#
tommorris
so, Wikidata is currently debating the use of a property which tries to split the world up into 7 types. and just occasonally hexagonal pegs don't quite fit square holes. so they get firmly shoved.
#
tommorris
so the 'person' type also includes families like the Medici family.
#
tantek
sounds like an org
#
tommorris
anything that doesn't fit becomes simply a 'term'. Python is a 'term', not a programming language, because they don't seem to think programming languages are 'creative works'.
#
barnabywalters
tommorris: RE list of friends from silos without APIs, php-mf2 might do this for twitter. You can register a built in set of class conversions which will treat some of the twitter.com class names as mf2 class names
#
tommorris
to which they say "ah, but we need it so we can work out how many articles we have about people!" - and I respond "yeah, but if the data is garbage, any statistics you derive from it will be bullshit"
#
barnabywalters
So if you can find a static URL for followers/following, you can feed it into phpmf2
#
barnabywalters
I think the demo allows you to turn on twitter class name conversion
#
tantek
tommorris - nice
#
tommorris
barnabywalters: I've ended up just firing up a Ruby shell and interrogating the API. ;)
#
tantek
GIGO FTW!
#
tommorris
we've reached deep semweb nerdery on the thread about it. and I am in the strange position of advocating a type system based on anarchy rather than the studious work of librarians.
#
tantek
tommorris, we need controlled vocabularies because semweb folks can't even agree on what's meta and what isn't :P
#
tantek
folksonomy!
#
tantek
(in the UK)
#
tommorris
well, controlled vocabularies just boil down to having only seven answers when being asked to fill in the blank: "Python is a ___________".
#
tantek
as long as one of the answers is "other" - you're fine ;)
#
tommorris
and the funniest thing about it is there's some guy who is all like "yeah, nothing useful ever got built on the basis of anarchy and bottom-up categorisation"... on the data platform for Wikipedia.
#
hadleybeeman
is torn between defending the semweb world (some use cases lend themselves to controlled vocabularies!) and agreeing that too much dictatorship doesn't fit the way we use English.
#
tommorris
I almost felt like asking him if he ever has any worries that anybody can edit the wiki...
#
barnabywalters
Ha, I found yet another inconsistency between twitter implementations — on desktop twitter.com you must be signed in to see who people follow, but it’s available to the publis on mobile.twitter.com!
#
barnabywalters
So I’ll make another set of fake mf2 class names for mobile.twitter so we can get that info APIless for free (for the moment)
#
tantek
tommorris, hadleybeeman, I started this with some thoughts (re: versioning, discovery, rel value) - http://microformats.org/wiki/version
#
hadleybeeman
tantek, are you deliberately excluding mediawiki "History" pages as examples?
#
tommorris
yeah, was going to add them, but they are rather hidden (behind the view history tab)
#
christopheducamp.com
created /FAQ-fr (+3410) "[fr: FAQ translation ]"
(view diff)
#
tommorris
so, of the 2165 people I follow on Twitter, 1902 have a URL set. ;)
#
tantek
hadleybeeman a mediawiki "History" page is not the page itself
#
tantek
so that would be a different link relation (perhaps multiple)
#
hadleybeeman
Also, whatever happened to REL=PREV from here? http://www.w3.org/TR/relations.html
#
tantek
we could have a rel=history that linked to the history page for a specific page
#
tommorris
now, lots of those will be corporate accounts and brands and crap, but... that's a lot of people we can convince to start using those URLs again. ;)
#
tantek
hadleybeeman rel=prev (and next) are for sequential navigation/pagination, not historical revisions
#
hadleybeeman
Okay. They're slightly different activities, but I'm not sure the relationship isn't similar.
#
hadleybeeman
(Though I obviously don't know much about the history on that)
#
tantek
theyre not slightly different - they're totally different
#
tantek
in pagination, each of the pages has different content and is a sequence like a list
#
tantek
or a series of chapters
#
tantek
with historical revisions, each page (typically) has mostly the same content with just a few changes of differences
#
hadleybeeman
But aren't they still nodes in a graph with a single edge between pages?
#
hadleybeeman
A unique, directional edge between pages?
#
tommorris
getting a list of URLs from Twitter is fantastic. I'm discovering all sorts of old friends' blogs.
#
tantek
hadleybeeman - all hyperlinks are nodes with a single edge between pages - your statement doesn't add any information
#
hadleybeeman
it's a 1:1 unique relationship between nodes in a sequence, tantek. Which is slightly more specific than a general, unstructured graph.
#
tantek
sure, they're both sequences
#
tantek
but they're very different kinds of sequences
#
hadleybeeman
How so?
#
tantek
see above about content
#
tantek
"in pagination, each of the pages has *different content* and is a sequence like a list, or a series of chapters. with historical revisions, each page (typically) has mostly the *same content* with just a few changes of differences"
#
hadleybeeman
Okay, the content is more likely to be similar in a versioning relationship than a pagination sequence... But does it matter?
#
hadleybeeman
(For what you're doing?)
#
tantek
"does it matter" = what's the use case
#
tantek
and yes, the use cases are very different
#
tantek
sequential navigation -> reading from beginning to end
#
tantek
historical navigation -> browsing through/across versions to see differences
#
tantek
(typically in opposite order as well)
#
tantek
people usually read from page 1… N
#
barnabywalters
My quick test for “is rel next/prev suitable” is “would the content make sense if not paginated”
#
tantek
whereas people usually browse revisions from current to previous version etc.
#
barnabywalters
As pagination is a hack anyway
#
tantek
pagination is often used as an alternative to delivering the content as a whole
#
tantek
e.g. news articles
#
tantek
whereas you never see previous historical versions all appended together as if that's "the whole thing"
#
hadleybeeman
Ah.
#
barnabywalters
It makes no sense to read the same thing over and over again, with slight differences :)
#
tantek
they're structurally, and usefully (real world usage) *very* different
#
hadleybeeman
Okay, that is significantly different. It's not just a question of order, it's a difference in the fundamental unit of text you're looking at.
#
tantek
and use
#
barnabywalters
Another way of visualising it: rel prev/next on horizontal axis, revisions on vertical axis
#
hadleybeeman
Ah, I like that, barnabywalters. Good analogy!
#
barnabywalters
hadleybeeman: If in doubt, visualise a graph :)
#
hadleybeeman
I'm a fan of those :)
#
tantek
yes, pagination is in space, versions are in time
#
hadleybeeman
K, I can see now why rel=prev isn't suitable for what you're doing.
#
hadleybeeman
Thanks for the clarification!
#
hadleybeeman
Right, I think it's unplugging time for me. Thanks for the nudge, tommorris
#
hadleybeeman
I'm sorry I've been neglecting this channel of late.
#
hadleybeeman
Good to catch up with you lot!
#
tantek
hadleybeeman - tried to capture the difference here: http://microformats.org/wiki/version#what_about_rel_prev
#
tantek
since I doubt you'll be the only one to ask that question
#
tommorris
hadleybeeman: I'll nudge you again randomly when I feel we need your input. ;)
#
hadleybeeman
appreciated, tommorris. :)
#
hadleybeeman
And I'm glad it was helpful, tantek!
#
tantek
any opinions on rel=version vs. rel=revision?
#
tantek
and rel=current vs. rel=latest ?
#
tantek
(that's directed at hadleybeeman, tommorris, barnabywalters and anyone else that wants to contribute)
#
tantek
or if you have alternative terms, those too
#
tommorris
is just in the middle of bootstrapping his personal wiki. ;)
#
hadleybeeman
Will catch up with it in the AM, tantek. :)
#
tantek
thanks hadleybeeman
#
tantek
appreciate the questions/suggestions
#
barnabywalters
tantek: Remind me why current/latest isn’t the same as rel canonical? Think I missed that
#
barnabywalters
Wow, christopheducamp is on a translation roll! Thanks :)
#
barnabywalters
Okay, got it
#
tantek
loves being able to answer questions by providing a URL :)
#
brennannovak.com
edited /App.net (+125) "/* Posts about */"
(view diff)
#
barnabywalters
URLs are the best :)
#
tantek.com
edited /App.net (+28) "/* Posts about */ subheads, threads"
(view diff)
#
tommorris
so, tantek and barnabywalters, you asked if anyone was publishing friends as XFN. I've just started: http://wiki.tommorris.org/Friends_and_contacts
#
tommorris
and barnabywalters, sorry, I know you are missing from the indiweb list. along with others. ;)
#
tommorris
will add soon, but it's rapidly approaching 11pm and I have work in the morning
#
tantek
tommorris, might want to remove the errant nofollow from your link rels :P
#
tantek
and yet - how did you get rel values on mediawiki links?!?
#
tantek
I have enough trouble getting class names on mediawiki links (still wondering how)
#
tommorris
will be pulling the nofollows off shortly
#
tommorris
and the nofollows are gone.
#
tantek
ah but no class attribute support, oh well
#
tantek
wants/needs class='u-url' in several cases, e.g. inside h-cite
#
tommorris
so Toby Inkster worked on https://www.mediawiki.org/wiki/Extension:Link_Attributes - but it needs packaging up properly as an extension rather than as a hack
#
tommorris
I might fork the more recent parser function script and hack it to add class
#
christopheducamp.com
edited /silo-fr (+505) "[fr: sync'd with origina + added App.net to be translatedl]]"
(view diff)
#
tantek
since class is useful for microformats, I put it there
#
tantek
aaronpk - want to try installing https://github.com/microformats/LinkAttributes on indiewebcamp.com and see if it works?
#
tommorris
shall I test it on my rather virginal wiki first? ;)
#
aaronpk
sure :)
#
tommorris
'cos mediawiki does some interesting things with class attributes already
#
tantek
tommorris - I reviewed the extension code
#
tommorris
It works!
#
tantek
looks like it will append any class names you add in the markup to whatever mediawiki already has
#
christopheducamp.com
created /App.net-fr (+1389) "[fr: translation]"
(view diff)
#
tommorris
[http://adactio.com/ Jeremy Keith|rel=met|class=vcard fn url]
#
tommorris
produce
#
tommorris
<a class="external text vcard fn url" rel="met" href="http://adactio.com/">Jeremy Keith</a>
#
tantek
tommorris - that doesn't mean what you think it means
#
tantek
digs for an FAQ
#
tommorris
this may be the time to switch to µf2 ;)
#
tantek
tommorris - the only time you can put a root class name and a property class name on the same element is when the root class name is basically providing an entire object as the value of the property which then belongs to some other object higher up the tree
#
tantek
e.g. the location of an hCalendar event also being an hCard
#
tommorris
will fix that in the morning
#
tantek
tommorris - for uf2 only, you want:
#
tantek
[http://adactio.com/ Jeremy Keith|rel=met|class=h-card]
#
tantek
that's sufficient because the implied "u-url" and implied "p-name" rules will figure out the rest
#
christopheducamp.com
edited /App.net-fr (+100) "[fr :/* Billets en rapport */ ajout traduction en cours "Un Défi Ouvert pour App.Net" ]"
(view diff)
b0bg0d joined the channel
#
tantek
updates http://microformats.org/wiki/hcard-faq#nesting-properties with more information to hopefully clarify the answer for tommorris
#
tommorris
thanks tantek
#
tommorris
I'll just use class=h-card and ride the new wave. ;)
#
brennannovak.com
edited /App.net (+1344) "/* POSSE to ADN */"
(view diff)
#
brennannovak.com
edited /App.net (+41) "/* Open Source Components */"
(view diff)
#
tommorris
oh, in unrelated interesting news: latest research shows that 102,673,683 hours went into building Wikipedia
#
brennannovak.com
edited /App.net (+0) "/* Open Source Components */"
(view diff)
#
bnvk
Dah friggin Wiki syntax, GIVE ME SIMPLE HTML!
#
bnvk
Informal poll: how many of you devs in here use Gmail as some part of your email solution?
#
aaronpk
raises his hand
#
tommorris
bnvk: +1, although somewhat suspicious about continuing given Google's affinity for getting rid of things
#
bnvk
Cool
#
aaronpk
the nice thing is you don't need to know that when sending an email to me :)
#
brennannovak.com
edited /User:Brennannovak.com (+786) "/* Brennan Novak */"
(view diff)
morrocco_mole, tilgovi and caseorganic joined the channel
#
tantek.com
edited /wiki/ (+5627) "add disambig, page type, projects, and dump my brainstorm thoughts about building new personal wiki software based on static HTML as storage"
(view diff)
#
bnvk
aaronpk: I don't need to know that you use Gmail to send a email there- but does it really matter if EVERYONE ends using the same solution ;)
#
tantek
bnvk - there is certainly a monoculture risk there
#
tantek
and thanks for expanding the app.net article
#
bnvk
for sure :)
#
tantek.com
edited /User:Tantek.com (+289) "/* working on */ wiki functionality"
(view diff)
#
tantek
tommorris - when you get back, feel free to expand on my attempt at structuring our wiki-related discussions: http://indiewebcamp.com/wiki/#Page_type and the rest of that page
#
tantek
aaronpk - since you've done a lot of work on personal wiki software as well, would appreciate your contributions there too