#social 2015-07-22

2015-07-22 UTC
jasnell and the_frey joined the channel
#
aaronpk
cwebber2: mattl: cool I'll publish that then
#
aaronpk
i'm not friends iwth you guys on facebook but RSVP if you can! https://www.facebook.com/events/878538935556649/
#
aaronpk
on a plane! will check back when I land!
#
mattl
aaronpk: i flew in from boston today. 8am PDT might be kinda rough for me, given that i'm still up here :)
#
aaronpk
Ha I can do later
#
aaronpk
I have about 60 seconds to update it before ✈️
#
aaronpk
I changed it to 9. Feel free to comment on Facebook (or on your site and send me a Webmention ;-) if that doesn't work!
bblfish, jasnell, bblfish_ and pfefferle joined the channel
#
cwebber2
aaronpk: I should be able to meet you there! but I don't have facebook so I can't RSVP
#
cwebber2
see you then!
bblfish joined the channel
#
aaronpk
Great!
#
aaronpk
If by chance you can put up some html on your site you can rsvp that way ;-) http://IndieWebCamp.com/rsvp but no worries :-)
#
aaronpk
Guess it was good I didn't invite the other Chris Webber who has a mutual friend
pfefferle, the_frey, jasnell, the_frey_, jaywink, bblfish and melvster joined the channel
#
melvster
KevinMarks: there's some confusion about the indexical ... it's not "what you call yourself" ... you're perfectly fine to say "my name is kevin mark's homepage" ... it's about what other people are going to call you ... if other people name you in a consistent way, then the value proposition is unexpected reuse of those links, so collisions are desirable, not something to protect against ...
#
melvster
you may name yourself "the artist formally known as prince" ... but some people are just going to call you "prince"
#
melvster
that's how society works
#
ben_thatmustbeme
thinks melvster uses a different definition of "collision"
#
ben_thatmustbeme
traditionally collision means the same name for more than 1 thing
#
ben_thatmustbeme
this is aliasing
#
ben_thatmustbeme
multiple names for 1 thing
#
melvster
that definition works for me
#
melvster
but i hope i got the sense across
#
melvster
this is a beautiful feature about breaking out of silos
#
melvster
the open world assumption
pfefferle joined the channel
#
ben_thatmustbeme
rereads discussion with this new language in mind
#
ben_thatmustbeme
s/language/definition/
#
ben_thatmustbeme
i still don't see how adding #this really helps the situation though. this would cause more collisions not another alias as #this could refer to the person or the part of the page with id="this"
#
melvster
ben_thatmustbeme: anyone that knows to use id=this will know how to use a primary topic if this is not the primary topic, so i dont think it's a worry
#
melvster
but in the open world assumption even that's not a problem
#
melvster
it's like me calling you "ben"
#
melvster
when there is another "ben" in the room
#
melvster
this is how social works in the real world
#
ben_thatmustbeme
I don't think so, primaryTopic is specific to foaf is it not?
#
melvster
foaf contains that term ... but the "concept" isnt specific to foaf ... i think schema.org has one too
#
ben_thatmustbeme
html does not
#
ben_thatmustbeme
and most people use that first
#
ben_thatmustbeme
actually 100% of the web uses html :-P
#
melvster
ben_thatmustbeme: sure it does, the "body" tag and the "main" tag ... :)
#
melvster
The <main> tag specifies the main content of a document.
#
melvster
so we could use #main
#
melvster
but meh
#
ben_thatmustbeme
so if there is a main tag on a site then range14 goes away?
#
melvster
ben_thatmustbeme: no, the problem is not even specific to http, it's specific to documents, your passport is a document and it has a passport id (number) ... but that passport id is distinct from the person it describes ... this is simply a question about 'if i am taking down the details of your passport, what information should I enter into my database'
#
melvster
in general, it's a problem for *me* not a problem for *you*
bblfish joined the channel
#
ben_thatmustbeme
yes, its a consumer problem, not a producer one. as to passport dicussion, thats a bit disingenuous as we are discussing social, better to keep it in that context. we aren't looking to solve the academic problems of a librarian. also, that example is more about storage and db schema than about any sort of naming
#
melvster
ben_thatmustbeme: the first federated identity system on the web was called passport :)
#
melvster
ben_thatmustbeme: it's completely about naming ... the question is what URI should I call you as a person, to disambiguate from your homepage
#
melvster
the web is a giant database
#
ben_thatmustbeme
the question is, if my homepage says in the <main> that its me the person, do you need to use anything other than the actual URL?
#
melvster
well we're talking about the web of data here, not really html
#
ben_thatmustbeme
and honestly, there are few situations where thats a problem. most of the time, you know which based on context
#
melvster
the thing about the web is that it has one big downside
#
melvster
and that is that the switching costs of changing URIs is high
#
melvster
but in return you get a large audience
#
melvster
so when choosing a name, you want one that you wont change later on
#
melvster
that's the problem in its essence
#
melvster
the really nice thing about # in http is that the server will always ignore it
#
melvster
so if i have url#foo
#
melvster
the url is always the page
#
melvster
you get that for free
#
ben_thatmustbeme
nothing is free
#
ben_thatmustbeme
also, not always
#
ben_thatmustbeme
because you are overloading #
#
ben_thatmustbeme
# can mean an ID on the page, or a part of the text, if using fragmention, or literally anything else if you are using JS
#
melvster
the great thing about http is that you can use # to divide up the page ... now an important question is ... "would you ever want to do that?" ... its a valid question ... and it may have different answers ... but if you look at the history of documents, they've almost always had some way to divide them up into different sections ... on the web the way to do that is with a #
#
ben_thatmustbeme
yes, we all know how # work
#
melvster
yes, you are exactly right!
#
melvster
it is overloading
#
melvster
and overloading always causes confustion
#
ben_thatmustbeme
thats by definiton, overloading.
#
melvster
e.g. people overload email as a message service and a login service
#
melvster
in http the uri has a few overloaded meaning
#
ben_thatmustbeme
for fragmention thats a concern but we resolved it as "assume ID takes priority to fragmention"
#
melvster
depends if you follow the http spec or not
#
melvster
in http url#foo is different from url
#
ben_thatmustbeme
url#foo is a different url, the fragment is PART of the url, and can be processed on the server
#
melvster
now you can make a heuristic / rule that suits your community, as indieweb have ... but not every community will make the same design trade offs ...
#
melvster
# is never sent to the server in HTTP
#
ben_thatmustbeme
correct, the broser removes it. if the server has the url with #, it can do whatever it wants with it
#
melvster
it's not the browser, it's all http clients, that follow the spec, curl will too
#
melvster
but # is overloaded
#
melvster
both as the dereferencing function
#
melvster
and as the identification function (ie subset of URI)
#
melvster
that's the most confusing part
#
ben_thatmustbeme
what you are referring to is something you want the server to know about.. are you liking the page or person question
#
melvster
no, not the server
#
melvster
it's what i want linked data to know about
#
melvster
it's about primary keys and foreign keys
#
melvster
let's say i want to add you as a friend
#
ben_thatmustbeme
have to go catch a train
#
melvster
i need to name you using a URI
#
melvster
ok, nice to chat, cu
#
melvster
it's all about outbound vs inbound links
bblfish joined the channel
#
ben_thatmustbeme
Indeed (on phone) inbound links (requests for data from the webserver) don't need the disambiguation. Meta info on the url is needed for inbound links that the server needs to read (did someone like my page or me). The problem is any value as the fragment is STILL ambiguous. You need some other info like a rel=
#
ben_thatmustbeme
Unless of course you state on the page what the main topic is (in indieweb it's the first h-* is considered the main topic)
jasnell, guangyuan, ynn, AnnB, bblfish and lehawes joined the channel
#
AnnB
Hi All -- @lehawes and I are having a SocialIG meeting ..
Lynn joined the channel
#
AnnB
Hi Lynn ..
#
Lynn
Hi Ann, sorry had to reboot
#
AnnB
I had to do the same .. always happens right at wrong moment, eh?
#
AnnB
I see you coming in on WebEx.. YAY
#
Lynn
trying...
#
ben_thatmustbeme
pops his head in
#
Lynn
i'm there but can't hear...
#
AnnB
KevinMarks, we're discussing a user story about which you had comments
#
AnnB
... if you're around, we'd love to chat with you
#
AnnB
Lynn ... look at the settings on "connected to audio" at left of webex screen
#
AnnB
I always have to select it, then select for my headset for both talking and microphone
#
AnnB
I always have to select it, then select for my headset for both talking and microphone
#
AnnB
(whoops, sorry for dupe)
#
AnnB
Hey Ben .. can you get your audio going?
#
ben_thatmustbeme
just listening in, cannot speak
#
ben_thatmustbeme
in a meeting
#
AnnB
cool .. ok
#
AnnB
trackbot, start meeting
#
trackbot
is preparing a teleconference.
RRSAgent joined the channel
#
trackbot
RRSAgent, make logs public
#
RRSAgent
I have made the request, trackbot
Zakim joined the channel
#
trackbot
Zakim, this will be SOCL
#
Zakim
I do not see a conference matching that name scheduled within the next hour, trackbot
#
trackbot
Meeting: Social Web Working Group Teleconference
#
trackbot
Date: 22 July 2015
#
ben_thatmustbeme
Meeting: Social Web Interest Group Teleconference
#
AnnB
thanks ben .. I was just looking that up
#
ben_thatmustbeme
is all powerful :)
#
ben_thatmustbeme
perhaps it is important to specify that the "group" getting deleted, is really just disassociated from the group
#
AnnB
appears to us that the main objection, by @kevinmarks and agreed with by us, is line #11 in the original user story -- that Roger deletes the group
#
AnnB
we agree there needs to be some sort of 'governance" on how deletes are handled -- who gets to delete what
#
AnnB
lehawes: also, discussion on github between @elfpavlik
#
AnnB
.. and @kevinmarks is questioning whether that's pertinent to API or not
#
ben_thatmustbeme
that would be very useful
#
ben_thatmustbeme
thats why i try to listen in to the IG too
#
ben_thatmustbeme
AnnB: we skipped this week
#
ben_thatmustbeme
we are moving to every other week
#
ben_thatmustbeme
I think that can work. those times work much better for me
#
lehawes
I like idea of using a WG "off" week to schedule a session for IG to work through Minor Objections with those WG members who have registered them.
#
ben_thatmustbeme
the conversation i was on was about finding a vocabulary for profiles
#
cwebber2
aaronpk: ha, yeah is it the devops chris webber? :)
shepazu_ joined the channel
#
AnnB
lehawes: hoping Vocab Task Force can make recommendation on which vocab to use
#
lehawes
At July 8 meeting, the IG came up with two recommended actions.
#
AnnB
... foundational for profiles
#
ben_thatmustbeme
the other part of the discussion was that while the vocabulary for profiles might not be important for V1 of the social API, it may be important for v1 of FEDERATION API
#
lehawes
1. To have the Vocabulary TF vet potential base vocabularies for social profile and recommend one that can be extended and implemented. Primary candidates seem to be Vcard and Hcard.
#
ben_thatmustbeme
the "social API" is you connecting to your service (use any app to connect to your service) "Federation API" would be between sevices
#
ben_thatmustbeme
at least thats how i understand it
#
lehawes
2. Get consensus between IG and WG that the profile vocabulary will not be hooked to V1 of the Social API, but wait until V2. However, the vocabulary will be needed for V1 of the Federation API.
#
Loqi
that doesn't make sense
#
lehawes
Heres the link to the July 8 meeting minutes: http://www.w3.org/2015/07/08-social-minutes.html
#
Lynn
ann, lost your sound
#
AnnB
hmm .. what?
#
ben_thatmustbeme
AnnB: woah... WAY delayed
#
ben_thatmustbeme
you are like 30 seconds behind
#
ben_thatmustbeme
oh, i think i caught up
rhiaro_ joined the channel
#
Lynn
Ann: for profiles, work on vocabulary
#
ben_thatmustbeme
I agree with Kevin on that one, I think it may be a wording change on you cannot delete posts (aside from possibly internally)
#
ben_thatmustbeme
I'll see if i can get his attention off channel too
#
AnnB
I think we all agree on KevinMarks ' deletion question / objection
#
AnnB
.. we just haven't been able to confirm alternatives or have discussion with him
#
KevinMarks
morning annb, what was up?
#
lehawes
I will keep trying to get Kevin Marks' attention and see if my proposed wording change to the user story would remove his objection. I'll do that via GitHub
#
AnnB
YO, hi!
#
ben_thatmustbeme
KevinMarks: IG call going on right now
#
AnnB
any chance you could dial into WebEx .. we're discussing a user story objection you had
#
AnnB
we all agree with you ... just not clear on how to resolve
#
KevinMarks
can't easily dial in at the moment, but can chat here
#
AnnB
ok
#
KevinMarks
my problem is that the story doesn't make it clear what 'deleting the group' means -
#
lehawes
Hey KevinMarks! You objected to the Groups user story.
#
AnnB
we agree
#
lehawes
Specifically, the ability for any group member to delete a post or the entire group.
#
AnnB
(lehawes is typing ... )
#
KevinMarks
there is an implicit silo assumption there that it is his group and the others are posting in it
#
AnnB
right
#
ben_thatmustbeme
cannot find any B.B.King lyric "too hot to hold"
#
lehawes
We agree that is not tenable, but are wondering if you would be OK with the story saying that an administrator/community manager would be the only one with those rights.
#
KevinMarks
as opposed to posting on their own sites and the group serving as an aggregation
#
AnnB
hmm .. that's a different twist
#
lehawes
Get it Kevin, but that isn't what you originally objected to in the user story wikipage
#
KevinMarks
Amy's comment sums it up
#
KevinMarks
no, it's he same thing - he is deleting the group - is he also deleting their posts?
#
AnnB
I just read her comment
#
AnnB
lehawes: makes sense
#
AnnB
I agree to a point .. but not totally clear on how that works from a user point of view
#
AnnB
here's Amy's comment from her vote on Groups user story: "... perhaps the posts are attached to the users who made them, not the group itself, and it's just the aggregation of these posts that disappears rather than the posts themselves? In which case a group is a collection of users plus a collection of selected posts. "
#
ben_thatmustbeme
remember this says nothing about how groups work internally to a business
#
ben_thatmustbeme
it the idea, that posts could be deleted by and administrator within a single system
#
lehawes
KevinMarks: I think Amy's comment works on the wider Social Web, but doesn't represent how employees and groups function within a business.
#
AnnB
s/and admin/an admin/
#
AnnB
question for me is WHO gets to delete WHAT
#
AnnB
seems OK for me to delete my own stuff, but not others' stuff
#
AnnB
but at the same time ... in Boeing, we only let people edit and delete for one hour ..
#
AnnB
... for basic posts and comments
#
AnnB
... so as to preserve the record
#
AnnB
... if I'm the owner of a group, I can delete the group
#
ben_thatmustbeme
you should always be able to delete your own content, if there is an administrator of that user, they can delete too. a group administrator can remove the association of a post to the group perhaps, but does not own the content
#
AnnB
lehawes: I've seen it stipulated that everything from a group is archived before deletion
#
AnnB
lynn: seems like a diff between how things work in public web versus enterprse
#
AnnB
enterprise
#
ben_thatmustbeme
Anything readable on the web, can be copied, forked, etc
#
AnnB
yes, true
#
AnnB
(to both points, lynn and ben)
#
ben_thatmustbeme
I'll try to write up a long post on my thoughts at some point this week
#
AnnB
in essence we're agreeing with kevinmarks' concern about deletion
#
lehawes
I'll record here what I just said out loud. The API must allow for implementers to configure controls as to who may delete posts and groups, but it must not force implementers to apply strict controls.
#
AnnB
but at the same time, it seems like there's a separate issue or POV if a group is a group, or is an aggregation of posts by individuals...??
#
aaronpk
AnnB++
#
Loqi
AnnB has 27 karma
#
AnnB
we need to understand that better
#
AnnB
we, the IG folks, that is
#
KevinMarks
the same dichotomy applies on the web to some extent
#
aaronpk
I see value in both models
#
lehawes
Yes, Kevin! Clearly, we're all in agreement here. :>)
#
ben_thatmustbeme
always in favor of more meetings / more discussion
#
AnnB
we'd like to ask for a joint meeting with you guys, so we can get to ground on this
#
Loqi
AnnB has 28 karma
#
KevinMarks
if Roger's group is liek a blogpost with comments on his site, it's clearer that he can delete posts and indeed thw hole thing
#
aaronpk
paroneayea: mattl: I have a booth inside now!
#
AnnB
could we perhaps use one of the intermediat weeks when the WG isn't meeting ... to meet together?
#
KevinMarks
if it is a collection of their posts, like a storify, he can edit but not delet their orginlas
#
AnnB
I'm going to send an invite, hoping to make that work
#
KevinMarks
and if the orginals are rmeoved, whetehr they remain is fluid
#
AnnB
unfortunately would be 2 weeks from now
#
ben_thatmustbeme
good luck AnnB
#
AnnB
OK ... sorry but I have to run ..
#
lehawes
That's OK AnnB. Would give people lots of advance notice and we could advertise the meeting and agenda more than once.
#
AnnB
this was an interesting session .. would be soooo interesting to actually be able to talk!
#
lehawes
Thanks everyone for participating today!
#
AnnB
every time we discuss one of these stories, people come up with intriguing perspectives ...
#
AnnB
the challenge is to "get to ground" on them
#
AnnB
thanks so much for attending with last-minute notice
#
AnnB
hopefully I'll get my mom set up soon, and will be able to be more attentive
#
AnnB
gotta run ..
#
AnnB
trackbot, end meeting
#
trackbot
is ending a teleconference.
#
trackbot
Zakim, list attendees
#
Zakim
sorry, trackbot, I don't know what conference this is
#
trackbot
RRSAgent, please draft minutes
#
RRSAgent
I have made the request to generate http://www.w3.org/2015/07/22-social-minutes.html trackbot
#
trackbot
RRSAgent, bye
#
RRSAgent
I see no action items
#
AnnB
rrsagent++
#
Loqi
rrsagent has 1 karma
#
AnnB
hehe ... love it
tantek, tilgovi, bblfish and shepazu joined the channel
#
Zakim
excuses himself; his presence no longer seems to be needed
Arnaud and the_frey joined the channel