Re: Issue-19 questions remain - a proposal

> On 21 Apr 2015, at 23:22, Harry Halpin <> wrote:
>> -- Is there a way for the wiki to send a notification when there is an update? Does that happen via the Watchlist? (Personally I find it hard to have to go look, randomly, for updates. That feels really unproductive.)
> We might be able to, although that would flood the mailing list. Thus,
> it seems wiser to simply note major changes in the telecon.
>> -- Should we agree to use the "Discussion" pages? 
> We could, or just see the note re IRC.
>> Or is IRC the place for conversation?  It's great to use Loqi to tell! someone (who's not present) something. It's great that there are daily logs. 
> Of course, the larger problem may be some people simply may not want to
> discuss with each other, due to time constraints or fundamental
> disagreements. Again, that's not a WG issue per se nor solvable by a
> resolution. We cannot, for example, make a resolution saying "Tantek,
> you have to spend whatever time it takes to agree with bblfish even
> though you two disagree about how specs should be built."
> That being said, I think the IG should volunteer to host a discussion
> over Henry's stories.
> I'd like to stick the WG to technical topics that are clear and
> delimited rather than working style differences that are open-ended.

I think Anne is asking: "how would this work even if people were willing
to discuss things". Clearly if people don't want to listen to each other
and discuss anything, but are just pushing an agenda then it is going to be
difficult to get to anyway close to a consensus, and consensus building is
the mission of the W3C. 

I understand that there are very strong divergences of methods and undersanding
of the space we are in. I have gone through all of them myself at various points
in the last 10 years. In any case at the face to face it was agreed in fact 
that the group is not going to push for one standard because the divergences
are too strong at the moment. But for the divergences to reduce then we need to 
have communication. 

So let's assume we do want to communicate, and look at the issues we can
deal with, namely buidling a process for communication. After all we are 
trying to build a social web. Now there are a number of tools that one 
needs to build to have a social web. 

One needs a way to send everyone in a group a message to alert them of some
project or idea, so that the whole group can focus its attention on a particular
topic. What tools can one use for this?

a) mailing lists have until now been very good and served the W3C and IETF 
well, as they allow a message to be sent from one to many
b) Wikis are not good unless the whole wiki has an RSS feed that people would be 
expected to add to their blog reader and poll regularly. This as you point out might
be very noisy. 
c) IRC channels have a way to ping one person, but not to ping the whole group
  ( the gitter chat for github has an @all, but that ends up working by sending every
   member an e-mail )

So if e-mail is out by Tantek's decision, and neither wikis nor irc channels are the right tool for the job,
then we have the following question:

Q1: How would one do one to many communication using the Social Web without relying on e-mail?

This is a question we MUST answer. It should be part of our user stories, since it is holding us up here. (But it is difficult to answer this if we don't have a channel to communicate about the various ideas on how to answer it, before we build it ). 

If we are to be able to do this now, using tools at our disposal, we need to use existing standards.
Lukily I think they are available, and have been for 10 years. We could do it like this:

One answer is that the Social Web WG could have a URI, lising each member of the group by their WebID, and that each WebID profile could describe that user including a foaf:weblog relation to their blog ( which has a relation to their RSS Feed where they can post their messages ). 

Eg the social Web WG would have

<> foaf:member <>, 
                                                  <>, ...

Then each of these WebID profiles would have a relation relating the user to a blog like this:

<> foaf:weblog <> .

Then by a drag and drop operation on the Social Web foaf:Group into a feed reader, the feed reader could fetch all those blogs, find the linked rss feeds, and poll those regularly ( once a day at least ), and show the group member what others have read. Perhaps we'd have to agree that blogs related to social web WG would be tagged by a special tag, so that we could filter out people's cat pictures from the discussion relevant to the topic. The W3C could index all those posts in an archive.

To do this we would not need to invent anything new, but we could use existing standards such as:
• Atom feeds
• foaf profiles

We'd still perhaps need to agree on a link relation to state that one atom entry was a response to another one. Is this all we need to do?


Social Web Architect

Received on Tuesday, 21 April 2015 22:41:33 UTC