W3C home > Mailing lists > Public > public-owl-wg@w3.org > November 2007

Input to UFD scope/issues (Was: UFDTF from tech plenary session)

From: Martin Dzbor <m.dzbor@open.ac.uk>
Date: Fri, 16 Nov 2007 14:05:56 +0000
Message-Id: <07EE8D9C-4B6A-4839-85F7-E09DA34596D2@open.ac.uk>
Cc: jjc@hpl.hp.com, skw@hp.com
To: ewallace@cme.nist.gov, public-owl-wg@w3.org

Hi,

just to chip in with a few thoughts about this user facing or less  
technical documentation effort, and I take advantage of Evan's point  
on "outside". One classic issue I have often faced with users (i.e.  
content managers, DB/XML schema authors, in my case) is a simple  
question (which in its blunter version is something like): Why should  
I bother with OWL when technologyX works for us? (swap "technologyX"  
for RDF(S), XML, DB schemas, etc.)

IMHO, it is true that OWL (1.0) Overview touches on the relationships  
of OWL to other technologies. Also, reference, guide, etc. emphasize  
the functional aspects of OWL, but it might be also useful (IMHO) to  
have a more detailed material on this non-technical topic = like  
"Motivation for using/migrating to OWL V.v"... I deliberately try to  
stay away from term "documentation" or "non-techie reference" -- at  
least, for the moment. :-)

The ideas below are a form of bootstrapping the wiki page on [1] with  
some more specific issuettes and also an initial contribution to what  
this TF can produce... just before proposing or amending anything in  
[1], I thought it's good to sum it here, in the list...

Personally, I think the UFD task force might be a good space for  
exploring a possibility of such a material, with these initial  
goals / issues / themes:

-- act as some sort of PR for OWL (in general and 1.1 in particular)  
introducing the updated standard also to those who do not build tools  
or reasoners, but merely want to explore how OWL content may help in  
their situations

-- present the case for OWL (1.1) in general terms = pizzas, wines,  
etc. but also collect one-two pagers containing some form of support  
short statements about benefits from specific domains (bioinf,  
genetics, geography, content mgt,...)

-- target such users as Evan and others mentioned = admins, IT staff  
(but also guys like librarians, content curators) in organizations;  
those who already manage (DB or XML) schemas, DBs, structured data

-- give these people some form of official (read WG explored and  
discussed) rationale, a set of arguments positioning OWL as a  
technology presented in a problem-centric rather than DL or RDF  
centric style 

-- possibly interesting point of such a set of arguments might be  
some form of "decision guide" = when am I better off with other  
technology, when OWL can be handy, when OWL is necessary (this was  
touched on by Vipul, I guess during the telco)

-- maybe (and this is a very raw thought) OWL Overview + Reference  
are more about telling people what different features of OWL actually  
are and what they do, what this task force might look at is to start  
something like "Why OWL and which OWL features" wizard (i.e. Problem - 
 > OWL feature / functional capability -> OWL vocabulary /  
formalization)


Maybe some parts of this problem -> OWL mappings were covered by the  
best practices WG and others, so I am not suggesting reinventing it  
here. But digesting the core aspects into an official case for  
adopting OWL or for migrating to OWL (1.1) may open up the standard  
to people who may not have considered OWL, ontologies and the part of  
universe where the two reside so far......


I guess it's time to stop it here -- don't want to try the readers'  
patience any further. Thanks for staying up to this point...


[1] ... http://www.w3.org/2007/OWL/wiki/UFDTF


Best,

Martin


[p.s. I am happy to seed the issuettes in [1] with a digest of the  
above = if it makes some sense... and if other TF people agree]





On 15 Nov 2007, at 16:35, ewallace@cme.nist.gov wrote:

> Jeremy wrote:
>> Stuart thought that the View from the Outside tech plenary session  
>> might
>> be relevant to our TF.
>>
>> I found the slides:
>> http://www.w3.org/2007/11/TPAC/view-from-outside.pdf
>> Point 1 - What is the role of a specification or standard?
>>
>>     * [snip] People read the spec and fall asleep.
>>     * MO: Web standards donít affect me because the people I work  
>> with donít use them.
>>     * MH: We were starting to talk about ontologies. MO didnít  
>> understand the words.
>> [snips]
>>     * PH: I find myself getting lost in the spec. I have to  
>> understand things more that I care to.
>>     * ST: I am able to read specs and implement them. i have a  
>> bunch of junior developers I have to help train. Just for them to  
>> reach them level we want them to achieve. Who pays for them to  
>> learn the company or the client?
>>     * MO: Business folks are only willing to pay for what they  
>> understand.
>>
> I guess the "outside" referred to here is outside of the W3C standards
> community.  However, the panelists are all web developers.  As such,
> they are outside of the user community that I represent.  I represent
> those who want to capture their domain knowledge in OWL, not those who
> build tools to capture that knowledge or build UIs to interface with
> it (others in the WG represent those communities).  Before someone
> says it... mine is not an anti-web perspective.  IMO - much of the
> usefulness of the click-able web comes from the ability of non-web
> developers to participate by publishing into it directly without the
> need for esoteric software or specialized web experts.
>
> Our standard needs to be useful to both those who build OWL tools
> and those who author OWL content.  Since these are two different  
> audiences
> and the documents aimed at them have different goals, it makes a  
> lot of
> sense to break the work product of this WG into different documents or
> document sets split along these lines.  This is what WebOnt did and
> what the deliverables in our charter reflect.
>
> -Evan
>
Received on Friday, 16 November 2007 14:06:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:13:27 GMT