Use Cases Comments

Since we're getting close to reaching consensus to release the Use Cases
and Primer as Working Drafts, here's my current comments.

Use Cases: Overall, release as Working Draft with minor corrections.

Minor Corrections:

Over entire document: In general, one can take any ".)." and replace it
with ")." The period inside the parenthesis is not necessary.

Intro:

"A number of documents contain data that could be valuable if they were
automatically accessible. In particular, it would be extremely
interesting if such documents could be transformed in RDF as a pivot
language for other systems which don't use that specific document format
themselves."
=>
"Documents usually contain data that would be even more valuable if it
was automatically accessible. One method of extracting data from a
documents would be to transform the document into RDF, so RDF can then
be used as an interchange language for other systems which can not use
the original format of the document."

"to filter the events down to those" -> "to discover"

1) Scheduling : Jane is trying to coordinate a meeting.

"Browsing the calendar of her friends, Jane noticed various conferences,
talks, and other gatherings of social groups in her area" -> "While
browsing the calendar of her friends, Jane noticed various conferences,
talks, and other gatherings of social groups in her area that interest
her. "

Wait a sec...this use-case involves having GRDDL produce RDFa, and then
a RDFa web-browser. Since this is still up in the air as regards whether
or not GRDDL can produce RDFa, I say that everything from "Browsing the
calendar of her friends" down be temporarily removed until we come to
consensus on this issue.

*See also:* put projects and prototypes here. -> delete if there aren't
any relevant projects or prototypes, or point to RDF Calendar:
*http://www.w3.org/2002/12/cal/report1173.html
*

2) Health Care: Querying an XML-based clinical data using an standard
ontology

Title: "Use case #2 [:] Health Care: Querying an XML-based clinical data
using an standard ontology" -> "Use case #2 - Health Care: Querying an
XML-based clinical data using an standard ontology"

"He wants to use a content management system which includes a mechanism
to automatically replicate an XML document into [equivalent, named] RDF
graphs for persistence in synchrony with any changes to the document."
-> "He wants to use a content management system which includes a
mechanism to automatically replicate an XML document into an RDF graph
of the same information that  synchronizes with any changes to the
document." Named graphs aren't standard yet, though they should be :)

"Each is a GRDDL source document
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#GRDDLSourceDocument>
associated with transforms that extract clinical data as RDF expressed
in a universally supported vocabulary for a computer-based patient
record." -> ""Each record is a GRDDL source document
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#GRDDLSourceDocument>
associated with transforms that extract clinical data as RDF expressed
in a universally supported vocabulary for a computer-based patient record."

3) Use case #3 - Aggregating data: Stephan wants a synthetic review
before buying a guitar.

This use case seems a bit out of order. Everything you want to do up to
paragraph 3 seems to be easily do-able without GRDDL. We need to
emphasize the fact that you can extract *multiple review from multiple
sites* and *aggregate them and query them using RDF* - not hide that
fact in the last paragraph, which isn't well connected to the first three.

So, delete this paragraph:
"Reviews published using hReview microformat
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#microformat>
can be discovered using existing search services. These GRDDL source
documents
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#GRDDLSourceDocument>
can be consumed by a GRDDL Processor
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#GRDDLProcessor>
to extract the RDF which is then aggregated together in a store.
Information about the reviewers can also be aggregated from various
sources including hCard and XFN microformats and autodiscovered FOAF
profiles possibly harvested through links in Stephan's own profile. The
filtering may be achieved by running SPARQL
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#SPARQL>
queries against the aggregated data, presented to the user through
regular HTML form interfaces."

and let's put this info up front, ala:

"Stephan visits sites offering guitar reviews, in particular looking for
reviews which give a high rating for the instrument. Since this reviews
are marked up using hCard and hReview microformats, Stephan's GRDDL
Processor can extract both the reviews themselves and information about
the reviewer in the form of RDF, and all of this information is then
aggregated together. Information about which reviewers are trustworthy
can also be aggregated from various sources including hCard and XFN
microformats and autodiscovered FOAF profiles harvested from links from
Stephan's own homepage.

>From this aggregated RDF, Stephan chooses reviews from only the
reviewers he trusts, and on submitting these preferences is finally
presented with a set of full reviews which match his criteria. This
filtering is achieved by running SPARQL
<http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm#SPARQL>
queries against the aggregated data."

I'm a bit confused about this phrase: "presented to the user through
regular HTML form interfaces."

4) Use case #4 - Querying sites and digital libraries: DC4Plus Corp.
wants to automate the publication of its electronic documents.

This use-case is great except for the fact that it's all proposed as a
"possible world" scenario with Adeline "believing" and "proposing"
things - look, in the rest of our use cases we just pretend it's an
implemented reality, so let's do that here as well :)

"She believes there is a need to automate the detection, indexing and
search capabilities for these documents. "
-> "There is a need to automate the detection, indexing and search
capabilities for these documents."

"She proposes a system that relies on Semantic Web technologies"
->
"She implements a system that relies on Semantic Web technologies"

5) Use case #5 - Wikis and e-learning: The Technical University of
Marcilly decided to use wikis to foster knowledge exchanges between
lecturers and students.

I believe the convention is in English to not capitalize "Diesel" and
"Ethanol"

It's kind of long, so if it can be made shorter without sacrificing
information that would be great. However, I'm not technically
knowledgeable enough about Wikis to make any concrete suggestions.

6)
"be push-based i.e. people who want their blogs to be included can push
the appropriate entries to his blog; his blog becomes somewhat of a
magnet for similar entries of interest." ->
"be push-based, i.e. people who want their blog posts to be included can
push the appropriate entries to his blog. This allows Voltaire's blog to
a magnet for similar entries of interest."

I'm still a bit confused about this use case. Is the point that people
can *automatically* push their blog posts to his blog? Does this really
require all this machinery to do?


You know, after reading this use-case, I think a more easily
understandable use case would be to have a standard mapping of the
various RSS brands to AtomOWL via GRDDL and so blogs can then aggregate
and multiply posts from differing blog systems automatically, but that's
just a off-the-top of the head thought, and not directly relevant to
this use-case.

"Voltaire has found the use of XForms for authoring fragments of Atom
quite useful for a variety of reasons" -> *please* enumerate at least
one reason.

Is this the reason?

"In particular, the Atom Publishing Protocol's use of HTTP and
single-purpose XML vocabulary as the primary remote messaging mechanism
which allows Voltaire to easily author various XForm documents that use
XForm submission elements to dispatch operations on web resources."
->
"In particular, the Atom Publishing Protocol's use of HTTP and a
single-purpose XML vocabulary as the primary remote messaging mechanism
allows Voltaire to easily author various XForm documents that use XForm
submission elements to dispatch operations on web resources."

I admit to not fully understanding this use-case, so please someone that
does (Chime I assume) double-check this rephrasing:

"Thus Johan's client relies on a GRDDL Processor to periodically extract
the service URIs, transform the content at these URIs to Atom/OWL and
query the resulting RDF to determine if the topics match. Doing so, he
will replicate his entries at the matching URIs by POSTing them there."->
 "Thus Johan's client relies on a GRDDL Processor to periodically
extract  service URIs like the URI of Voltaire's blog, transform the
content at these URIs to Atom/OWL and query the resulting RDF to
determine if the topics match. If so, he can then automatically
replicate his entries on bird-watching at the service URIs by POSTing
them there. This allows Johan to automatically update Voltaire's blog,
among others, with his bird-watching entries." -

7) Use case #7 - XML schema specifying a transformation: the OAI would
like to be able to specify document licenses in their XML schema.

Until we know how to do this technically and if that's compatible with
the GRDDL spec as it stands now, we should not include this use-case in
the working draft. I want to include this, just like I want to include
the RDFa calendar example, but I only want to include it once we know
how to do it and if GRDDL does it.

In particular, the use-case is great until we hit these two paragraphs:
"Since the creation of XML instance documents is often distributed, as
in the OAI case, this transformation would only be specified in the XML
schema document, but would apply to the instance documents, not to the
schema document.

Since the XML instance documents are often distributed, as in the OAI
case, the XML schema itself could embed RDF descriptions identifying a
transform to apply to all its instance documents. So doing, for each
GRDDL source document, the transformation is indirectly referenced by
the XML Schema it follows."

But we need a better story telling how this can be done! I have no idea.

             cheers,
                  harry

-- 

		-harry

Harry Halpin,  University of Edinburgh 
http://www.ibiblio.org/hhalpin 6B522426\

Received on Sunday, 24 September 2006 18:19:51 UTC