See also: IRC log
<sandro> trackbot, start meeting
<trackbot> Date: 28 July 2015
<tantek> trackbot, start meeting
<KevinMarks> https://www.w3.org/wiki/Socialwg/2015-07-28 says no more Zakim
<trackbot> Meeting: Social Web Working Group Teleconference
<trackbot> Date: 28 July 2015
<sandro> We still have the zakimbot for queue management
<sandro> fwiw
<tantek> present?
<sandro> ummm
i thought present+ was just for the logs
<sandro> nope.
<ShaneHudson> Yeah I'm muted, can hear you fine
i can scribe
<scribe> scribenick: aaronpk
<tantek> scribe: aaronpk
tantek: participation is limited to members, so please hang up if you are not a member
<tantek> https://www.w3.org/wiki/Socialwg/2015-07-14-minutes
tantek: any objections to the
minutes?
... or any +1s
<ben_thatmustbeme> +1
<ben_thatmustbeme> sure
+1
<akuckartz> +1
<ShaneHudson> +1
<wilkie> +1
<tantek> RESOLVED: 2015-07-14 minutes approved
tantek: i'm going to declare the minutes hve been approved
<tantek> https://www.w3.org/wiki/Socialwg/2015-07-28#Activity_Streams_2
tantek: first item is a pull request
<tantek> https://github.com/jasnell/w3c-socialwg-activitystreams/pull/191
tantek: most likely added by james
jasnell: yeah it was added by me.
it's a follow-on to the previous conversation
... about the purpose of a colection
... the previous converstaion, rather than adding a type
identifier or purpose identifier, the preferred approach would
be using a propertly, link-rel aproach
... that's what this does, to idenity 4 common types of
colections
... inbox, outbox, relationships
... and store, which is a link to the collection that has all
your stuff like albums or posts
... this is pretty stragithforward, adds 4 properties
... deprecates replies in favor of inbox
... rather than having a replies collection you have an
inbox
<KevinMarks> -1 on inbox over replies
jasnell: i saw some discussion on IRC about this before
<jasnell> the WD were published last week btw
tantek: yeah the IRC logs should
be recorded, let's assume the IRC logs are there unless someone
finds a probelm
... okay i see one -1 from kevinmarks
KevinMarks: my concern here is we
came up wth reply in the first place based on existing naming
practices across social networks
... there's a distinction between reply and inbox more than
just the name
... an inbox is directed at an individual, a reply is directed
at a post
... conflating the concepts is confusing implementation detail
with a meaningful distinction
tantek: do you have ea counter proposal to james' changes?
KevinMarks: i'm not sure why he's
deleting replies as well as adding inbox
... i don't think you can subsume replies with inbox without
losing that distinction
... i woudl say keep replies as well
jasnell: we have the existing in
reply to property
... if something is posted to the inbox with the in-reply0to
property you can infer it's a reply
<wilkie> I think the idea is to do filtering of inbox items into whatever you want?
jasnell: so we have the ability
to maintain the distinction without having a separate
collection
... inbox becomes more generalized
KevinMarks: so yhow would you express a collection of in-reply-tos
jasnell: you could do it as an extension, or find a way in the api to dfilter by that
KevinMarks: i'm not finding that
convincing
... when you build something very large like twitter, you end
up constructing an inbox/outbox model which is howyou do
caching
... but that is not structural
tantek: james what was the motivation for this change?
jasnell: it's going through and
pointing to the different collection types for
discoverabiilty
... the api is working with different collection types
<KevinMarks> http://activitystrea.ms/specs/json/replies/1.0/
jasnell: the pumpio API is an example, they have differnet ideas of your inbox and outbox
tantek: so this is an attempt to normalize AS2 with pump?
jasnell: yeah
<KevinMarks> "replies Used to provide information about the thread of discussion associated with a given object. Typically, the objects contained will have an implied inReplyTo property value equal to the object containing the replies property."
jasnell: normalize how we discover these separate types of collections
tantek: we don't have formal API
requirements yet, we just have approved user stories
... although someone could take the approved user storeis and
write up API requirements
... and for each requirement, cite the user story that provides
that requirement
... but i haven't heard anyone step up to do this
<ben_thatmustbeme> I see it as moving to a more generalized "inbox", something like "webmention endpoint" in the indieweb side of things
tantek: bbut if that did exist, you could cite it as an api requirement
jasnell: if we need to hold off and wait til the api matures more that's fine
tantek: we also have another
agenda item to rename the inbox user story
... maybe that will help inform this
... so unless we can come to a quick resolution i don't think
we can come to consensus on this call
<KevinMarks> I'm looking for definition of inbox that is as clear as the 1.0 one for replies
jasnell: the PR describes where i'm at
KevinMarks: i' see that we have existing AS 1.0 replies thing which is very detailed, i don't see an equivalent in defining inbox in that way
jasnell: inbox is intentionally
defined as a generalized loosely defined concept
... that can contain replies as well as other things
KevinMarks: it sounds less loose
to me
... part of this is inbox implies a receiving end. a reply can
happen anywhere, but an inbox is a queue of things that have
come from outside thaty ou have to dela with
<ben_thatmustbeme> naming issues with "inbox"
KevinMarks: in distributed stuff, replies can exist anywhere on the web, but inbox is a queue of things locally
jasnell: i don't see those as
being different things
... a reply you're going to receive, so they're conceputlaly
identical
tantek: let me see if i can find
a way to move forward
... kevin can you come up with a counter proposal or a proposal
in addition to james' changes ijn the next 2 weeks/
KevinMarks: i'll have to read it
in more detail
... i appreciate it sounds like i'm arguing about the names of
things, but i am arguing they are structurally different
<rhiaro> Can you have a Replies collection for a particular post, but not an Inbox collection for a post, Inbox just for a person?
tantek: the burden is now on you
to come up with a propsoal for how the spec should change to
indicate they are different concepts
... and we can look at that as agroup
<ben_thatmustbeme> rhiaro, thats a good point
KevinMarks: i'll look at the change in more detail
<scribe> ACTION: KevinMarks to provide counter proposal or proposal in addition to James' PR https://github.com/jasnell/w3c-socialwg-activitystreams/pull/191 [recorded in http://www.w3.org/2015/07/28-social-minutes.html#action01]
<trackbot> Error finding 'KevinMarks'. You can review and register nicknames at <http://www.w3.org/Social/track/users>.
<tantek> item: AS2: Status of publication of editor's draft?
<jasnell> they were published last week
jasnell: i would like to start a regular candence of once a month publishing a new draft
<jasnell> http://www.w3.org/TR/activitystreams-core
<jasnell> http://www.w3.org/TR/activitystreams-vocabulary
tantek: congrats for another working draft, appreciate your work on that
<jasnell> http://www.w3.org/TR/2015/WD-activitystreams-core-20150722/
<Arnaud> it's in the agenda: Published: http://www.w3.org/TR/2015/WD-activitystreams-core-20150722/ and http://www.w3.org/TR/2015/WD-activitystreams-vocabulary-20150722/
<akuckartz> jasnell++
<Loqi> jasnell has 28 karma
<jasnell> http://www.w3.org/TR/2015/WD-activitystreams-vocabulary-20150722/
tantek: okay good, i agree once a month is a good thing
<tantek> item: AS2: Proposal to replace media type "application/activity+json" by "application/ld+json" plus profile
tantek: raised by andreas
<melvster> https://www.w3.org/wiki/Socialwg/Media_type_for_AS2
tantek: i pointed out we don't
have any new information, but then andreas documented the new
information
... i'll let andreas present this
akuckartz: i'll try to keep this
short
... this one issue was created a while ago
<akuckartz> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/52
akuckartz: which was closed
because there was no further discussion or actions
... not based on the content of the discussion
... i raised this issue of the media type in the SoLiD
community and there were several people who agreed it would be
better for the linked data community to have the jsonld media
type
... what i'm interested in is it would be good to have more
arguments in favor of having the special AS media type
... the main issue that was put forwrad in this issue was the
problem of two information models
<jasnell> I'm -1 on removing application/activity+json, but +0 on adding the application/json+ld+profile.
akuckartz: but we alreadya have
multiple communities with information models and jsonld is the
way to combine the communities
... my proposal is to reopen the discusison and collect more
information
<jasnell> the impact is specifically on the API
<jasnell> essentially, it would mean allowing the API to accept: application/json+api;profile=<http://www.w3.org/ns/activitystreams> as an alternative to application/activity+json
tantek: okay, sanyone have
anything else to add?
... if not, we can leave the wiki page open to continue to
collect the different options
<melvster> just the wiki page I posted earlier: https://www.w3.org/wiki/Socialwg/Media_type_for_AS2
tantek: and advantages and
disadvantagses of each
... doesn't sound like we have a consensus yet
jasnell: i'm definitely -1 on
removing activity+json media type, but allowing json+ld+profile
we can do that also if that makes it easier for the json ld
folks
... all it takes is adding a section to the spec about the
profile parameter and what it means
tantek: you're requeting we ge tsome additional spec text provided as part of a proposal?
jasnell: yes
tantek: does that sound reasonable?
akuckartz: sure
<KevinMarks> I thought the point of using a specific type was that implied a profile, without having to include it?
tantek: why don't we leave the
wiki page open for contributions for two weeks and re-evalutate
it then
... is that a reasonable step forward?
akuckartz: perfect for me
jasnell: yes and i'll re-open the issue on github
tantek: thank you both
<tantek> https://www.w3.org/wiki/Socialwg/2015-07-28#Social_API
tantek: any update on the
brainstorming document?
... i thought i saw all 3 ofyou on the call
rhiaro: there's nothing in particular to update, there's been some requests to add solid stuff into the api outline document
<Loqi> Tantekelik made 1 edit to [[Socialwg/2015-07-28]] https://www.w3.org/wiki/index.php?diff=85250&oldid=85249
rhiaro: but nobody has offered to do that
tantek: thanks, any questions
about the brainstorming docs or specific requests?
... okay next item
<melvster> FYI: apache who were doing some work on the SoLiD documentation in the API have left the WG
tantek: i added this agenda
item
... to Rename Inbox user story to what it actually is: "Read
Social Streams" (or "Social Streams Reader")
... b based on some of the comments from the original
voting
... it seems like the term inbox is causing misunderstanding
about what this user story is about
<tantek> item: Social API: Rename Inbox user story to what it actually is: "Read Social Streams" (or "Social Streams Reader")
tantek: and assumptions beyond
what is in the user story
... let's open up the floor for discussion
... go ahead and q+ to provide feedback in any way
ben_thatmustbeme: i had brought this up wa while ago. if it's just the name of the user story and doesn't affect teh functionality, i don' tsee any problem
<ShaneHudson> +1 to changing just the name
tantek: i think in general,
because the contents and steps were voted on and approved, it's
out of scope to alter that
... if there's an alternative user story there's the other page
for that
... this proposal is just about changing the name, keeping the
steps as writtne
<rhiaro> +1
<ben_thatmustbeme> scribenick:ben_thatmustbeme
aaronpk: i just wanted to add my +1 to this, because the framing of "inbox" comes with this baggaghe of email-like which is not in the intent of the story
<aaronpk> scribenick: aaronpk
KevinMarks: folowing what i was
talking abotu earlier, the nuance is the sense of
direction
... sergey desribed twitter as email wihtout a "to"
address
... but there is a mechanism to read it
... even tho it's not directed to you
... the power of this model is that they don't behave in the
way of having an "unread" count
... you could do that, but it's underminig the more subtle
distinction which is that we hve a web of things people
say
... and some of them are in response to each other,
... but taht shouldn't create a burden on the person to respond
to it
... calling it inbox brings back that email-centric framing
that the web and social web transcends
<ShaneHudson> Agreed about the email mindset being awful
<akuckartz> melvster stated that discussion was initiated on the mailing list.
tantek: given the input so far let's put this to a straw poll
<melvster> -1 already implementing using the term 'inbox', I did not expect user stories to change at this point, any change would be more work, is the current naming a show stopper for any other implementers?
PROPSED: rename Inbox story to "Read Social Streams"
<ShaneHudson> +1
<KevinMarks> +1
<melvster> https://www.w3.org/wiki/Socialwg/Media_type_for_AS2
PROPOSED: rename Inbox story to "Read Social Streams"
<melvster> oops
+1
<ben_thatmustbeme> +1
<rhiaro> +1
<melvster> -1 already implementing using the term 'inbox', I did not expect user stories to change at this point, any change would be more work, is the current naming a show stopper for any other implementers?
<tantek> +1
<rhiaro> melvster: the story isn't changing... the contents of the story are the same
<ben_thatmustbeme> JUST THE NAME
tantek: i want to clarify, the user story itsefl is not changing
<wilkie> +1
<tsyesika> 0
tantek: so anyone implementing
the steps in the user stlory will not need to change their
work
... but this is about the WG as a whol ebeing better about
communicating what we're working on
<akuckartz> -0.5 lets see what happens in the the mailing list discussion
<KevinMarks> can you link to maling list discussion akuckartz ?
<ben_thatmustbeme> scribenick:ben_thatmustbeme
aaronpk: i just want to point out that the mailing list dicussion that is being brought up on irc is not really a discussion, it is a single email by melvster
<aaronpk> scribenick: aaronpk
tantek: melvster are you on the
phone?
... anyone else understand melvin's objection?
... the user stories aren't actually changing
... wanted to see if there was any other aspect to his
objection that we're not taking care of
... andreas you wanted to see what happens in the amiling list
discussion
... aaron pointed out that the discussion has occured on IRC
and this phone call
... do you have a specific objection to make that we should
record?
<rhiaro> ^ and the wiki
akuckartz: there isn't that much
discussion on the amiling list as a whole as far as i see
it
... my point is we should perhaps not decide on what to do
today but keep it open for two weeks and then decide
tantek: if we had new information i would tend to agree
<tsyesika> well people can -1 when approving the minutes for next meeting?
<KevinMarks> the email says inbox is poorly defined https://lists.w3.org/Archives/Public/public-socialweb/2015Jul/0034.html
tantek: but i feel like we have enough consensus to declare today
<tsyesika> if they've missed this because of holiday
tantek: absense of discussion is
not usually a readson to hold things up
... is that reasonable?
<KevinMarks> whereas the social stream story is clear
akuckartz: that's okay for me
tantek: alright let's go ahead
and declare that reslved then
... if someone does come up with new information deserving to
reopen it, we can do that
RESOLUTION: rename Inbox user story to "Read Social Streams"
tantek: we can implement that witha change ont he wiki
<melvster> wait; which of the two suggestions are we renaming it to?
tantek: let's keep the fragment
IDs so existing links don't break
... i can take care of that
... we just took the first suggestion
... and gives a "nod" to the history of RSS, the coincidence of
acronyms
... any volunteers to edit the wiki accordingly?
<melvster> well my -1 stands ... no one answered if anyone is implementing this user story
<ben_thatmustbeme> I'll take care of it
tantek: thanks ben
... just for the minutes, melvin is stil indicating he has a -1
and the reason I chose to override that as chair is my
understanding of what he wrote in IRC is his understanding is
the user story is changing and that is not true, it's just the
name
... melvin perhaps you can join the next call if you have new
information t o add so we can reconsider that
<ben_thatmustbeme> melvster: how could what is basically a rename of a URL possibly EVER effect your implementation / code ?
<melvster> how could it not?
<tantek> item: Social API: Complete: moved approved user stories to the "approved" section on the wiki
<melvster> naming is a large part of what standards is about
<tantek> https://www.w3.org/wiki/Socialwg/Social_API/User_stories#Approved_user_stories
<ben_thatmustbeme> aaronpk: its just what it sounds like. I didn't see anyone had done it yet even though approved, so i just went ahead and corrected the wiki with the 8 approved stories
tantek: great okay
<tantek> https://www.w3.org/wiki/Socialwg/2015-07-28#Charter_License_Update
sandro: i've done my part, it's
in the hands of w3c management
... i expect it will go to the AC later this week
tantek: great
... do you know how long we might expect that to take for the
AC to look at and approve/reject?
<KevinMarks> I think reader.kylewm.com counts as an implementation of Reading Social Streams
sandro: i believe it's 4 weeks
tantek: whatever we can do to minimze the time, if there's something we can request..
sandro: feel free to talk to the AB to say we don't need an AC review
tantek: this is a good test case
to suggest process improvements
... let's see how this goes and document it and se how we can
make it go more smoothly in the future
... we expect it to go to the AC this week, they'll have 4
weeks to look at it and vote on it, and at the end of that, it
will take nother couple weeks to turn the crank and update the
charter
<Arnaud> +q
tantek: if there are any objections before hand we can relate that to the WG
Arnaud: i was wondering, i didn't
see the final decision on what we put before the AC
... is it just the license?
tantek: jsut the license
<KevinMarks> woodwind implements all of that, though 5 would only be security by obscurity for now
sandro: did you revise the wiki page with the schedule?
tantek: if there's an action for me let's record that
<sandro> ACTION: tantek to update http://www.w3.org/Social/WG/Overview.html#sched [recorded in http://www.w3.org/2015/07/28-social-minutes.html#action02]
<trackbot> Created ACTION-72 - Update http://www.w3.org/social/wg/overview.html#sched [on Tantek Çelik - due 2015-08-04].
<tantek> http://www.w3.org/Social/track/issues/raised
<tantek> http://www.w3.org/Social/track/issues/pendingreview
tantek: no pending
... no actions pending review
<Loqi> Benthatmustbeme made 2 edits to [[Socialwg/Social API/User stories]] https://www.w3.org/wiki/index.php?diff=85252&oldid=85220
<tantek> http://www.w3.org/Social/track/actions/pendingreview
tantek: anyone want to declare
victory on open actions / issues?
... alrigh we've got 9 minutes left
... i'll confirm the next telcon is August 11, and Arnaud wil
be chairing
... so no telconference next week
... any other business to bring up? or we can close the call
and end early
<ShaneHudson> I have to leave now anyway. Thank you tantek
tantek: okay not hearing
anything, so i'd like to thank everyone who joined the tel
conference
... there's only so much discussion we can do on email and irc,
so your participation on the conference helps us make
progress
... i appreciate you taking time out of your day to join the
call
<akuckartz> bye!
tantek: have a producte two weeks and we'll talk to you on august 11th!
<tsyesika> bye :)
<wilkie> thanks :) bye
<wilkie> aaronpk++
<Loqi> aaronpk has 912 karma
<tantek> aaronpk++
<Loqi> aaronpk has 913 karma
<tantek> ben_thatmustbeme++
<Loqi> ben_thatmustbeme has 106 karma
<tsyesika> aaronpk++
<Loqi> aaronpk has 914 karma
trackbot: end meeting
<tantek> trackbot, end meeting
This is scribe.perl Revision: 1.140 of Date: 2014-11-06 18:16:30 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/spec test/spec text/ Found ScribeNick: aaronpk Found Scribe: aaronpk Inferring ScribeNick: aaronpk Found ScribeNick: ben_thatmustbeme Found ScribeNick: aaronpk Found ScribeNick: ben_thatmustbeme Found ScribeNick: aaronpk ScribeNicks: aaronpk, ben_thatmustbeme Present: jasnell aaronpk kevinmarks Arnaud Sandro rhiaro ShaneHudson ben_thatmustbeme tsyesika wilkie wseltzer WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 28 Jul 2015 Guessing minutes URL: http://www.w3.org/2015/07/28-social-minutes.html People with action items: kevinmarks tantek[End of scribe.perl diagnostic output]