Social Web Working Group Teleconference

13 Dec 2016

Attendees

Present: aaronpk, KevinMarks, ben_thatmustbeme, csarven, annbass, sandro, tantek, rhiaro, cwebber2
Regrets: evan
Chair: Sandro
Scribe: ben_thatmustbeme

aaronpk: the issues that are open right now are from external commenters, and every issue on there now has a PR for it and i'm waiting to hear back from the commenter about the issue
... it looks like tantek has given some thumbs up on all of them, but i'm assuming thats just from him and not from the original commenter

sandro: some of these are from Mozilla?

aaronpk: yes. They are all just clarifications, but I'd appreciate any review people want to do
... these will be the last things to go in to the draft before going to REC

sandro: that sounds like a good reason not to go to REC before the break, give commenters time to review them and respond
... it will be much harder to make any changes after this
... does anyone want to go over any of these now?
... wasn't there something about 1mb hard-written in to the spec

<KevinMarks> Can we talk them through and commit the PRs, and wait for commenters in that state?

aaronpk: yes in the security considerations, should put limitations on fetching" "for example, 1 mb or 5 seconds"

sandro: i would have imagined we would say that clients have to read at least the first MB
... i'm happy with the text on 86

aaronpk: if anyone is reviewing these now and we are happy with it, i'm happy to merge PRs in and then wait for commenter after the merge

<aaronpk> https://github.com/w3c/webmention/issues/76

<aaronpk> https://github.com/w3c/webmention/pull/85/files

aaronpk: this should be pretty easy, i basically added a background section at the top of the document
... explaining how webmention was built from and on top of pingback

sandro: pingback never specified what to do for updates and deletes

aaronpk: correct, it actuallly could not be used for updates as it said to reject a pingback that was already registered

sandro: what about trackback?

aaronpk: trackback was just actually sending the comment snippet, there wasn't anything about URLs in the spec

sandro: whenever i explain webmention, i have to use the word trackback for them to understand it

annbass: would it be possible to add some little comment about trackback as well?
... set the historical context

<KevinMarks> I suspect talking to people who haven't been on the web for a decade, trackback references will only confuse

aaronpk: i don't know, trackback is so old and overrun with spam it doesn't seem worth mentioning

annbass: i suppose but as sandro said when he explains it he has to reference trackback

<sandro> https://movabletype.org/documentation/trackback/specification.html

aaronpk: even just bringing up trackback spec confuses me, do we want to link people to it
... the spec talks about installing cURL, thats not a spec

<ben_thatmustbeme> can we just say "and other linkback methods" or something?

aaronpk: maybe i can include a sentence about webmention being in this family of linkback mechanisms

sandro: and maybe link to wikipedia for those that don't know what linkback is?

<sandro> "Webmention provides a LinkBack mechanism. It began as a simplified version of PingBack.... "

csarven: i think thats a good way as there are other methods that may be coming up

<annbass> I like that suggestion .. shows there are some other 'link back' mechanisms, with a ref, but without having to delve into any history or details

<sandro> "Webmention provides a LinkBack mechanism, somewhat akin to TrackBack. Webmention itself began as a simplified version of PingBack.... "

<csarven> I don't know if anyone actually cares about this but AFAIK, Semantic Pingback (circa 2010) predates Webmention. Whether that needs to be said somewhere is something else.

<csarven> and both derive from Pingback.. or I don't know if WM derived/simplified from SP or not.

<KevinMarks> Right, there are lots of these, so referring to the general idea is better than an attempted genealogy.

<tantek> the specific request was for Pingback since that is what is widely known and deployed (and thus folks have some understanding of its security etc characteristics in practice)

<KevinMarks> Technorati was link back as a service, Blogger has had backlinks based on Google crawl for years,

<tantek> it wasn't about a full genealogy of all experiments

<tantek> to add clarity from the commenter's perspective (Mozilla)

aaronpk: i think what i want to do is just mention that webmention is a linkback and it links to the wikipedia page

<tantek> so no need to add anything beyond that

aaronpk: i'm doing that right now

sandro: looks good

<KevinMarks> Or we end up going back to Xanadu and Memex

<tantek> no

<sandro> :-)

<tantek> the linkback abstaraction is academic

<tantek> the specific request was for pingback and/or trackback

aaronpk: since tantek opened these issues, i'll wait for tantek to go back to the original commenter

<tantek> because of practical deployment experience reasons

<csarven> Then just say Pingback!

<tantek> not for definitional ones

<csarven> No need to talk about LInkback inf act

<tantek> csarven - right, just Pingback is enough

<tantek> that's what the issue said originally

<tantek> not sure what all this bikeshedding is about

<sandro> it's done, tantek

<tantek> I confirmed that just referencing Pingback is enough, the way the pull request did when I reviewed it earlier this morning

<tantek> (pull request on the issue)

<tantek> linkback--

<Loqi> linkback has -1 karma

<aaronpk> https://github.com/w3c/webmention/issues/84

aaronpk: this is a tricky one, it sounded like he misunderstood what the section was about
... it was specifically about the POST request
... that was the section, but the comment, was about republishing private comment, essentially a totally different issue which we hadn't talked about before
... instead of changing that section, I added another section about his actual concern
... since webmention doesn't specify anything about republishing, its an note under the verification section
... it mentions that the receiver might be republishing that
... i added a note below that about not unintentionally changing the privacy on that
... just trying to make sure people are aware of it, without specifying what to do about it. thats a job for the next layer up

sandro: i'm happy with that change

<KevinMarks> That's a useful clarification.

aaronpk: closing 71 since tantek opened that and he thumbs upped it

sandro: do we know when the publishing moritorium is over?
... we'll figure it out laster

<rhiaro> January 3, 2017: Publications resume

sandro: are we meeting on the 20th and/or 27th
... i'd guess we are not meeting the 27th, we COULD meet the 20th, i'd prefer not to

aaronpk: i'd be up for dropping the 27th meeting
... that means the next meeting would be the date publishing ends

sandro: if you want to go and prepare a draft for the third, assuming all the commenters are happy
... rhiaro have you gone to sleep?
... we'll try to move things along and get it published when the moritorium ends

<KevinMarks> http://epeus.blogspot.com/2004/02/technorati-xanadu-and-other-dreams.html historic info for csarven

<tantek> sandro, did you hear back re: REC publishing? is it already too late?

<sandro> no answer, tantek

<sandro> tantek, can you get a positive response from the commenter?

<csarven> KevinMarks: Do I have to read this? What was the inquiry?

<tantek> sandro, already preflighted that with the proposals in the issues

<tantek> also, to be clear, the Mozilla vote on the PR was to Yes to publish, with *suggested* changes, not required

<sandro> Sure, but since they're editorial improvements, it's not nice to ignore them.

<tantek> especially as these are all editorial / non-normative, I believe the changes being made reflect the intentions of the commenter

<tantek> (from having spoken with the commenter in-person last week)

<sandro> excellent, okay

<tantek> (we had a Mozilla All Hands last week)

aaronpk: i've been looking up on "linkback" and it looks like the term only exists on wikipedia
... it looks like it was created there

<cwebber2> to linkback?

<tantek> that sounds likea violation of Wikipedia's no original research rule

<cwebber2> https://www.google.com/search?sclient=psy-ab&site=&source=hp&q=linkback&=&=&oq=&gs_l=&pbx=1

<cwebber2> seems to be a lot of results

<tantek> nothing citable

annbass: do whatever you feel is right

<aaronpk> cwebber2, look at the results, they're not about the 4 methods

<tantek> cwebber2 - if there were, the linkback article itself would have a citation

<KevinMarks> No, you do t have to read anything, I thought you were be interested in the history of this

sandro: i find myself using the term 'backlinking'

<cwebber2> ah ok

<annbass> the point is to help the uninitiated reader understand

<ben_thatmustbeme> sandro, but backlinking is really,r eally generic though in my mind

<cwebber2> I gotta go though

<cwebber2> later everyone!

sandro: a backlink is a name for the link itself
... that is a kind of different thing

<tantek> cwebber2 - made-up stuff on wikipedia is bad for everyone

<ben_thatmustbeme> i'm not sure i like us linking to wikipedia in the spec anyway

sandro: yeah, lets go back to the pingback version then

aaronpk: i think i like that better
... there are 3 issues, 3 PRs, tantek believes they address all the issues raised
... so i will merge the PRs and leave the issues waiting for commenter so we don't lose track of them
... i'll do that and I'll stage this

sandro: okay, it looks like we can probably go ahead for december 15th

<ben_thatmustbeme> \o/

sandro: any last minute things from anyone else?
... do people want to meet next week?

<tantek> do we need to resolve to pubilsh anything?

<tantek> (today)

aaronpk: i might have some trailing things still for other work items

<sandro> no resolution needed, really, but we did conmfirm AS2 with the changelog.

<KevinMarks> “As we all know, Ted Nelson meant hypertexts to have bidirectional links. But due to a laboratory accident in Switzerland, we ended up with this lame thing.”

<sandro> webmention REC should go out this week

<sandro> "That was NO accident"

<tantek> wow we'll actually deliver at least one charter deliverable within our original charter period!

<annbass> gotta go ..

<annbass> thanks Sandro

sandro: okay, tentatively there is a meeting for next week, might get cancelled if we end up with nothing on the agenda

<sandro> meeting as planned next week, unless nothing turns up for the agenda

sandro: and we said no meeting on the 27th

<sandro> no meeting on Dec 27th.

<tantek> (just on IRC, sporadically :P)

<KevinMarks> You were present in github

<sandro> ADJOURNED

sandro: talk to you all either in 1 week or 3, and have a happy december

<tantek> sandro++ thanks for chairing!

<Loqi> sandro has 32 karma in this channel (37 overall)

<tantek> ben_thatmustbeme++ thanks for minuting!

<Loqi> ben_thatmustbeme has 60 karma in this channel (181 overall)

sandro: don't all say bye at once everyone

rhiaro: bye

<sandro> thank for scribing ben_thatmustbeme !

<tantek> so maybe meeting next week, but definitely canceled the 27th?

<aaronpk> yes

trackbot end meeting

Summary of Action Items

Summary of Resolutions

Confirm last week's decision to publish AS2 as a new Candidate Recommendation, now that we have a proper changelog for AS2 Vocabs http://w3c.github.io/activitystreams/vocabulary/#changelog

Socialwg/2016-12-13-minutes

Social Web Working Group Teleconference

13 Dec 2016

Attendees

Contents

AS2

webmention

Summary of Action Items

Summary of Resolutions