#social 2014-10-31

2014-10-31 UTC
#
tantek
welcome timbl!
#
fjh
tantek, thanks, I’ll update the reference to the joint session that I put in our minutes. I assume this is better to use than the RRSAgent generated minutes?
#
fjh
tantek, never mind, your note was clear
#
tantek
fjh IMO yes - the quality and usability of the #social WG IRC logs has now exceeded any other, including RRSAgent generated minutes. Every IRC utterance has an easily clickable/findable/usable (and marked up!) permalink
KevinMarks joined the channel
#
fjh
tantek, thanks!
#
fjh
tantek, if others want to do something similar, can they?
#
tantek
fjh - indeed. also add yourself to https://www.w3.org/wiki/Irc-people and you'll show up in the logs hyperlinked and with an icon
danbri, fjh, harry, nicolagreco, danbri1, cmhobbs, Arnaud, bblfish, tantek, timbl, timbl_, almereyda, pfefferle and jasnell joined the channel
#
tantek
jasnell - great - I really like how this is shaping up.
#
tantek
hoping we can figure out it enough to get something into the next WD
#
jasnell
so that's the next question: what would need to be added to the next WD :-)
#
tantek
next will be to figure out the mappings from "u-in-reply-to" etc. to the different explicit AS2 types
#
tantek
I figure we can start with a "Activity Streams in HTML" section, with two subsections, "Using h-entry for Activity Streams Semantics" and "Parsing h-entry into Activity Streams Semantics"
#
jasnell
as:inReplyTo :-)
chaals joined the channel
#
tantek
with complete code examples that show HTML+h-entry -> canonical microformats JSON -> AS2 JSON(-LD)
#
tantek
to provide consumers consistent guidance for parsing AS2 semantics from HTML.
#
tantek
we can use https://www.w3.org/wiki/Activity_Streams/Microformats_Mapping to iterate on the text for these sections
#
jasnell
ok, sounds good
fjh, fjh_ and JonathanJ3 joined the channel
#
tantek
jasnell - oops edit collision, I'll try to merge
#
jasnell
I'm done for now
#
tantek
almost done mergine
#
tantek
s/mergine/merging
#
tantek
jasnell - saved and done for now also.
#
tantek
ok maybe one more tweak
jtandy, JonathanJ3, jtandy_ and nicolagreco joined the channel
#
tantek
jasnell: added some more details, including turning the h-card as:author question into more of a mapping how-to. that's a good amount of progress for now.
#
jasnell
tantek++
#
Loqi
tantek has 113 karma
nicolagreco, danbri, danbri1 and jasnell joined the channel
#
tantek
jasnell - interesting post.
#
tantek
TBH I (and others) in the indieweb community tried to make explicit verbs "fit" into our UIs etc and they just seemed awkward. So we started nouning the verbs as we came up with implementations, and in each case it worked better (simpler) to make everything a post of some (implied) type.
#
tantek
worked better in terms of simpler markup, publishing, code to generate, store etc.
#
tantek
but most of all - it was the most obvious mapping to our UIs of actually posting / view these "activities"
#
tantek
like I said in the meeting - there may still be use-cases where explicit separate verbs make sense, but so far as we've implemented them one by one, in every case it has made sense (better for use-cases) for the verbs to be nouned, and to have a permalink of such nouns for reference / verification etc., *especially* in distributed interactions.
#
tantek
I can see how pure "verbs" might be ok in a silo-only feed, where the silo is assumed to be trustworthy and there is no need to "verify" the verbs as having been done.
#
tantek
However in a distributed world, especially in a *federated* world, that need for verification drives a need for permalinks, which drives a need for posts at those permalinks, which means every action gets recorded as the *consequence* of that action. posting a like, instead of just "liking".
JonathanJ3, bblfish, bblfish_, shepazu, jtandy, jtandy_, jasnell, tantek, tantek_, danbri, Arnaud, fjh, fjh_, danbri1, shepazutu and KevinMarks joined the channel