Re: Use Cases and Requirements for Mapping Relational Databases to RDF

On Tue, Apr 20, 2010 at 9:46 AM, Michael Hausenblas <
michael.hausenblas@deri.org> wrote:

>
> > I am truly confused right now.
>
> Yeah, confusing, isn't it?
>

I know. I understood that Angela, Dan and I were going to organize the Use
Case document. Angela did the intro and motivation and Dan and I were
working on the use cases. We wanted to show usecases such as [1] or [2].
These are completely different of what you guys just published. In what we
wrote in [3], we are trying to follow that same scenario.

[1] http://www.w3.org/TR/webont-req/#usecase-portal
[2] http://www.w3.org/TR/xhtml-rdfa-scenarios/#use-case-5
[3]
http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements#Use_Cases

>
> Again, as I said, we're operating on a moving target here. Don't worry,
> bits
> are integrated as we talk. We take as an input all the relevant Wiki pages
> and create a consistent, balanced document.
>
> I've printed out all the documents now and currently comparing everything.
>
> If you want to contribute, I'd like to ask you to rewrite [1] so that we
> can
> include it. In its current form I can't understand it, it doesn't fit
> really
> in and is IMHO also not really in scope of this document.
>

I understand that the R2ML Application and Use Case Requirement section in
[4] may be a bit hairy (and academic). However, we need to have design goals
and requirements. As mentioned by some participants of the group, we can't
just present a use case of a rdb in rdf. That isn't a use case. We need to
tell a story without trying to sell RDF. We need to motivate WHY we need to
do RDB2RDF. This section opens that up (and I agree that it needs to be
cleaned up). We have a starting point (Scope of Data Processing) where the
RDB that we want to expose as RDF can be completely structured,
structured+semistructured or structured+ microparsed tagged text.

Eventually we want to integrate our RDB with other resources such as:
stuctured or semantic (existing RDF) or some mashup.

We also need to have clear the role of the ontology. Are we mapping to an
existing domain ontology? or not.

The 3 use cases that we have initially presented tell a story of WHY we need
RDB2RDF. The Enterprise RB shows that you are integrating two structured
sources (both RDB). The Bio use case shows that you are integrating a RDB
with existing RDF. The Wordpress usecase shows that I just want to have
RDB2RDF so I can get crawled by semantic web search engines.

I just wanted to get this out now and hopefully people will have the chance
to read this before the telecon so we get on the same page.

[4]
http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements#R2ML_Application_Use_Case_Requirements


>
> Happy to continue the discussion during the telecon.
>
> Cheers,
>      Michael
>
> [1]
>
> http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements#R2ML_Appli
> cation_Use_Case_Requirements
>
> --
> Dr. Michael Hausenblas
> LiDRC - Linked Data Research Centre
> DERI - Digital Enterprise Research Institute
> NUIG - National University of Ireland, Galway
> Ireland, Europe
> Tel. +353 91 495730
> http://linkeddata.deri.ie/
> http://sw-app.org/about.html
>
>
>
> > From: Juan Sequeda <juanfederico@gmail.com>
> > Date: Tue, 20 Apr 2010 09:38:37 -0500
> > To: Michael Hausenblas <michael.hausenblas@deri.org>
> > Cc: RDB2RDF WG <public-rdb2rdf-wg@w3.org>
> > Subject: Re: Use Cases and Requirements for Mapping Relational Databases
> to
> > RDF
> >
> > Michael,
> >
> > What is in [1] is not what is in [2]. I'm confused? What is in [1] is
> > essentially everything that we had originally in the use case wiki page,
> > which I then moved to [3]
> >
> > I thought that the idea of Dan and I working [2] was to start organizing
> the
> > final document, because what is in [1] is just two much.  Through all the
> > discussion, we have determined that some of the use cases in [1] are more
> of
> > case studies and not actual use cases. In [2] we are telling a small
> story
> > where non-tech people can understand.
> >
> > I am truly confused right now.
> >
> > [1] http://www.w3.org/2001/sw/rdb2rdf/use-cases/
> > [2] http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements
> > [3] http://www.w3.org/2001/sw/rdb2rdf/wiki/Draft_of_Use_Cases
> > <http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements>
> > Juan Sequeda
> > +1-575-SEQ-UEDA
> > www.juansequeda.com
> >
> >
> > On Tue, Apr 20, 2010 at 9:24 AM, Michael Hausenblas <
> > michael.hausenblas@deri.org> wrote:
> >
> >>
> >> All,
> >>
> >> As a preparation for today's telecon I'd like to point out that the
> current
> >> version of the Editor's Draft for our UC document is available at [1].
> >>
> >> Due to the events in the last couple of days, we (Eric and I) are
> operating
> >> on a moving target, so please be kind with us, we try our best to
> >> accommodate with all the brilliant contributions made, no matter how
> late
> >> they came in ;)
> >>
> >> Although the content of [1] is still changing (though the raw bits are
> >> there; the structure has stabilised), I'd like to encourage you to print
> it
> >> out and read it. Now.
> >>
> >> Please bear in mind the intended audience, any implicit assumptions one
> >> might have (esp. as you should consider yourself an expert in this area,
> >> where maybe 90% of the the readers are not) and that we are writing on a
> >> W3C
> >> document, not an academic article.
> >>
> >> That said - enjoy the first Editor's Draft and looking forward to a good
> >> discussion re its content and future development in roughly an hour and
> a
> >> half.
> >>
> >> Cheers,
> >>      Michael
> >>
> >> [1] http://www.w3.org/2001/sw/rdb2rdf/use-cases/
> >>
> >> --
> >> Dr. Michael Hausenblas
> >> LiDRC - Linked Data Research Centre
> >> DERI - Digital Enterprise Research Institute
> >> NUIG - National University of Ireland, Galway
> >> Ireland, Europe
> >> Tel. +353 91 495730
> >> http://linkeddata.deri.ie/
> >> http://sw-app.org/about.html
> >>
> >>
> >>
> >>
>
>