W3C home > Mailing lists > Public > public-vocabs@w3.org > September 2014

Wednesday invite for Schema.org/W3C BOF [was Re: The Vocabulary, Schema.org governance, etc.]

From: Harry Halpin <hhalpin@w3.org>
Date: Thu, 25 Sep 2014 22:21:47 +0200
Message-ID: <5424795B.8060802@w3.org>
To: public-vocabs@w3.org
Hash: SHA1

On 09/25/2014 05:17 PM, Guha wrote:
> Peter makes an excellent point. In the past, before schema.org,
> when Google approached the W3C with Sitemaps and later with
> data-vocabulary, we were told that W3C does not do vocabularies,
> any more than it does websites.
> We also realized that we wanted to depart from some of the
> principles of linked data (URIs for everything) and languages like
> RDF (single domain / range). It is not clear we could have done
> such things as a W3C WG.
> While I have the greatest respect for the W3C and especially for
> folks like Ralph, Sandro and Phil, the W3C Working Group process is
> not suited for something like schema.org. For example, we interact
> extensively with many content providers in designing particular
> schemas. These folks often don't have the time or interest (or,
> sometimes, funds) to participate in working groups. And as the
> recent exchange re the Social Working Group demonstrated, non WG
> folks can't even talk during WG meetings. So, in that sense, we are
> a lot more open.

To be clear - everyone involved in schema.org are invited to either
join the Social Web Working Group or the Social Interest Group as
members, so then you can join the meeting at TPAC as full members of
the WG or IG. We'd love to see you at the meeting.

Also, we have arranged a special bird-of-a-feather session Wednesday
morning that can last as long as we want that has no membership
requirements in either Social IG or Social WG. So, Guha, you can
participate in that discussion, as can everyone else.

Would that work for all of you in Schema.org? Guha? Chaals? Peter? More?

So, if folks show up the discussion will happen. Just tell us what
your preferences are and if Wednesday works.

The long-winded explanation is that the WG decided they'd want
schema.org to *join* the WG if the folks are employed by a W3C member.
Given that the schema.org patent policy now in general matches the the
W3C patent policy, there should be no problem for joining the Working
Group at least as regards Actions, where there is interest from the
Social WG.  There was concern that folks from schema.org might would
try to participate in the WG without agreeing to the patent policy,
could cause problems later down the road.

Another alternative path is that since the Social Interest Group has
no Rec-track deliverables and is very open to invited experts, you all
 can join that Interest Group without patent commits and thus have
full rights and privileges for the Interest Group (and TPAC meeting).
On an aside, the Working Group is still debating the policy, but might
prefer Invited Experts that have implementation experience.

Longer-term, an Interest Group has both governance and no set
deliverables (and can even, as the Semantic Web Interest Group is, be
completely open), so this general kind of group (perhaps a Schema IG?)
might serve as a good home for schema.org if governance continues to
be an issue people bring up. The differences between the task force
and the IG would probably be minimal rather than that of a governance
structure to deal with breakdowns, of which schema.org has had quite
few so far.

As Peter Mika knows, a lot of us at W3C have been pushing for
agreed-upon vocabs for years. I gave a talk to the Rich Snippets crowd
in 2009 in Stanford - pre-schema.org - about why something like
schema.org was necessary [1].


[1] http://www.aaai.org/ocs/index.php/SSS/SSS10/paper/view/1140/0
> Guha
> On Thu, Sep 25, 2014 at 6:25 AM, Peter Mika <pmika@yahoo-inc.com>
> wrote:
>> Hi Renato,
>> The W3C in particular did not want to take on vertical vocabulary
>> projects in the past. Tim B-L emphasized in multiple talks that
>> the W3C would like to focus on developing ontology languages, and
>> let industry develop vertical solutions. (To me the examples you
>> mentioned such as SKOS and PROV are part of the language
>> infrastructure.)
>> schema.org is such a vertical solution based on the needs of
>> large web consumers.
>> Best, Peter
Version: GnuPG v1.4.11 (GNU/Linux)

Received on Thursday, 25 September 2014 20:21:53 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:29:44 UTC