W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2000

Re: SV: A certain difficulty

From: William Grosso <grosso@SMI.Stanford.EDU>
Date: Wed, 23 Feb 2000 10:00:40 -0800
Message-ID: <38B42048.A435B8D9@smi.stanford.edu>
To: Greg FitzPatrick <gf@medianet.org>
CC: xml-dev@xml.org, www-rdf-interest@w3.org

Good morning, 

Greg FitzPatrick wrote:
> 
> <William:plug for = "Protoge-2000">
>         <Greg:rebuttal>
> 

At the risk of being thought contentious, I don't think
that was a rebuttal :-) 

Instead, I think we're talking about lots of different 
issues under the same subject header. 

At least part of this thread has been about learning curve
issues. Quotes include:

> However the RDF spec. is particularly obtuse, and every 
> time i have to write something on RDF my heart sinks,

and

> It took me much much longer than a week to "get" RDF....
> Now the bad news.  The specs take you only so far in learning 
> the powers and pitfalls.  The examples floating around out 
> there are not terribly elucidating.  And the publicly available 
> tools are barely functional or out-and-out broken. 

Protege directly addresses this issue. You can use Protege
to create models, you can play with the example knowledge-bases
that come with the download, you can even (should you desire
to read pure XML syntax, look at the generated RDF files).

Learning to model inevitably involves building models and 
Protege-2000 vastly simplifies that task. That was my point-- 
Protege-2000 is a set of modeling tools, built for frame-based 
languages very similar to RDF (and then adapted to RDF). It grew 
out of the need for better modeling and knowledge acquisition 
tools and it explicitly addresses the problems mentioned above.

What's more, Protege-2000, since it is open-source, written in Java, 
built in a very modular fashion (e.g., Protege components can be 
embedded in applications, or additional objects can easily be 
added to Protege), comes with both examples and documentation,
etcetera etcetera, provides an excellent starting point for 
people to begin thinking about applications using RDF. 

What's more (and the reason for the plug in the first place)--
we're serious about the open-source stuff. I'd be thrilled for
people to download Protege-2000, try to use it, and then 
say "You know. It's a nice try but it doesn't work for what I 
need to do because [good reason]." 

I'd be even more thrilled if the sentence continued " ... but 
then I thought about it some more and I realized that the 
following piece of code fixes the problem."

Jumping out a little bit:  there may also be deeper issues 
being brought up in this thread. For example, I don't quite 
get the following paragraph:

> On the other hand there is something unsettling about 
> your solution to the "difficulties" being discussed.  I 
> know I was sloppy quoting someone offhand like that and I am 
> working in the background in order to get an "official"
> comment from that person, but we were talking about understanding;
> 
> understanding in order to implement
>                 and not
> implementing in order to circumvent understanding

I'm not really sure what's meant here. Not only am I unclear
on what "implementing RDF" means, it's not at all obvious to 
me that allowing people to ignore the syntax and focus on the
real task (building abstract models) is "circumventing 
understanding." 


William Grosso
-- 
William Grosso 			grosso@smi.stanford.edu
Phone 650-498-4255 		http://www.smi.stanford.edu/people/grosso/
Received on Wednesday, 23 February 2000 13:01:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:42 GMT