W3C home > Mailing lists > Public > public-sweo-ig@w3.org > March 2007

Re: task force information gathering: what do we want?

From: Danny Ayers <danny.ayers@gmail.com>
Date: Mon, 5 Mar 2007 10:27:09 +0100
Message-ID: <1f2ed5cd0703050127g7bd62bfar928891392d8fff8c@mail.gmail.com>
To: "Leo Sauermann" <leo.sauermann@dfki.de>
Cc: "W3C SWEO IG" <public-sweo-ig@w3.org>

On 02/03/07, Leo Sauermann <leo.sauermann@dfki.de> wrote:

> what do you want?

Very good question!

What I personally want is *something* in place to get the ball
rolling. So the question is, what's the minimum possible that could
possibly work? I'll take your point on this:

* I want to have a working way of gathering information items

I would imagine this in practice being :

1. a collection of source URIs
2. a (set of) tool(s) for automatically gathering the data at those URIs in RDF
3. the publication of the collected data on the web (as RDF)

I agree entirely that the specific vocabularies used are secondary
(but would aim to capture as much of the source information as
reasonably possible).

Again, I'm happy with the ultimate goal being a user-friendly portal
for this data, but would suggest the primary task is information
gathering, not presentation.

If the data has been gathered, then other parties may be willing to
present it before we have a portal in place. As an extreme example,
assuming the data was made available in not-big chunks, it could be
viewed in Tabulator out of the box.

For a more end-user friendly example, I'd be happy to set up a
Longwell install on my own server (if you point the default install at
a directory containing RDF files, it provides a facetted view of the
data - it's very low effort). Lots of folks have data browser demos
online, they could provide different interfaces to that same data.

Looking further ahead, I can't remember if I've mentioned this before,
but the kind of presentation I'd really like to see is a
make-your-own-tutorial setup. You use drop-down lists to select your
current knowledge, any particular angle you'd like to take (e.g.
semweb for managers, semweb for XML people, semweb for database
people...) etc. The system would then provide a customised "course",
looking something like Franconi's DL course:


If del.icio.us tagging data was included, something like this would be
very close out of the box with a facetted data browser.



Received on Monday, 5 March 2007 09:27:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:28:52 UTC