W3C home > Mailing lists > Public > public-csv-wg@w3.org > April 2015

Re: Planning the tam-tam for a CSVW document review

From: Ivan Herman <ivan@w3.org>
Date: Tue, 14 Apr 2015 14:32:29 +0200
Cc: Jeni Tennison <jeni@jenitennison.com>, Phil Archer <phila@w3.org>, W3C CSV on the Web Working Group <public-csv-wg@w3.org>
Message-Id: <40894B70-7DCE-4723-9DD1-8E23EF2CDD52@w3.org>
To: Dan Brickley <danbri@google.com>
>> I guess the first two are really important for us; I do not really see privacy or accessibility issues with these documents, but better ask.
> Our charter mentions Privacy, "Ensure that the privacy concerns are
> properly included in the CSV metadata either via reference or via
> direct vocabulary elements".

Yep, so we must do this.

> Accessibility imho is important as CSV could prove to be an
> interesting approach to structured data authoring. See
> http://www.w3.org/WAI/intro/atag.php and nearby. If a suitable
> collection of csv2rdf (or json) mappings are available, actual
> authoring of data can happen in various tools that allow tables to be
> edited.

Ah. That is interesting although… I am not sure it is relevant in terms of the review of these specs. The horizontal review is really to assess whether the documents abide to the requirements of accessibility.

>> Furthermore, we would probably want to ask for an explicit review from the Data on the Web Best practices group:
>>  * public-dwbp-wg@w3.org, chairs Deirdre Lee (Deirdre.Lee@deri.org), Hadley Beeman (hadley@linkedgov.org), Yaso Cordova (yaso@nic.br), and Steve Adler (adler1@us.ibm.com); staff contact Phil Archer (phila@w3.org)
>> Otherwise a separate mail to the Data Coordination Group (member-dacg@w3.org), the SWIG (semantic-web@w3.org) and the LOD mailing list (public-lod@w3.org)
> (definitely those 2, yes)
> as well as, maybe, an announcement to chairs@w3.org
> (yes to that too)
> would be fine. Finally, we would need a Home Page News announcement
> as well as a Blog on the Data Activity blog.
>> Any groups outside the core W3C circles? Dan, do you think it is worth using the schema.org list (I do not think so, but you tell me).
> I was thinking that it is well worth posting something in
> blog.schema.org (or at least the new CG for sure). In that context I
> would be tempted to frame it alongside recent explorations of using
> Web Components (custom elements etc.) as "exploring new ways of
> publishing schema.org".

That is also interesting… do you think it would be possible to provide some schema specific metadata samples for various sub-vocabularies of schema? Then people may use a csv file to describe the data and trust some conversion to, say, JSON-LD. That may be interesting...

> We should also make sure to bounce it around various open data / civic
> hackery / govt transparency lists. Jeni and Rufus probably have their
> fingers on the right pulses there. DWBP are also a good route to those
> audiences.

Beyond DWBP a big yes for other govt lists. Leave that to Jeni & Rufus...

> Given our charter http://www.w3.org/2013/05/lcsv-charter.html says
> "The output of the mapping mechanism for RDF MUST be consistent with
> either the RDF Direct Mapping or R2RML" we really owe implementors of
> those specs at least a high level explanation of why we've taken a
> relatively different approach.

That is correct; I am not sure how of us could/should write that.

> On that front we could re-iterate the
> idea of opening a dedicated Community Group on advanced mappings, now
> that the csv2rdf basic design looks stable. We could explore the idea
> that our JSON format is an authoring-friendly shorthand (pure subset?)
> of R2RML (or RML)?.

I think it would be a stretch to bind it to R2RML. The whole idea of URI template makes it very different, there is nothing like that in R2RML. Also, R2RML (or the Direct Mapping) do not have any notion of validation of the data, which is a very important part of what we have done.

Note, however:


showing that it is easy to reproduce, for a specific csv file, the Direct Mapping output with a suitable metadata file. If that approach works, it could be an important part of our answer to this.

> The charter also mentions Forms ("Coordinate on the possible usage of
> CSV XPath expressions" - did anything happen there?) and XML. We
> didn't go the XML route and this is a fair point at which to explain
> why (and redirect followups to w3.org/community/ ?).

The short answer is: no. But that falls under the issue that we have no XML mapping in the first place. And, at this stage, we are sure there won't be….

(B.t.w., the XFOrms WG has been closed, afaik…)

> We also need to communicate carefully and appropriately to the IETF,
> given that they specify the underlying CSV format we began with. We
> shouldn't assume an IETF audience is necessarily up to date with any
> recent changes in W3C process, or has followed our group in detail.

Yeah. It is unfortunate that Yakov is not active any more; I wonder whom we could contact on this (if there is anybody still active on this)

> BTW how are we looking on the Working Drafts for today? If I recall
> right they should be with W3C webmaster by tonight?

We are ahead of schedule:-) the documents are ready on our site, and the request to the webmaster has been sent!


> From an airport -  back online later,
> Dan

Ivan Herman, W3C
Digital Publishing Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: http://orcid.org/0000-0003-0782-2704

Received on Tuesday, 14 April 2015 12:33:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 14 April 2015 12:33:03 UTC