W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > October to December 2004

Re: Permanent model names in TKS/Kowari

From: Tom Adams <tom@tucanatech.com>
Date: Tue, 23 Nov 2004 09:51:15 -0500
Message-Id: <1FEAFCA2-3D5F-11D9-8861-000A95C9112A@tucanatech.com>
To: DAWG list <public-rdf-dawg@w3.org>

>> Here is a summary of the discussions that we've been having internally
>> on permanent model names
>
> There's that "model" word again... would you please rephrase
> in Feb 2004 terminology? Do you mean RDF Graph?
> http://www.w3.org/TR/2004/REC-rdf-concepts-20040210/#dfn-rdf-graph

No, I mean model, that's what we call them ;) Yes, I mean a set of 
triples. Models in TKS/Kowari are mutable containers composed of RDF 
graphs (Simon: Is that a fair assessment?).

> or do you mean something mutable? (Graphs are immutable syntactic
> things, like numerals and strings).

Yes, models in TKS/Kowari are mutable.

>> <snip/>
>> o Models should to be named in order to be able to reference them in
>> queries;
>> o Names should be independent of the method of access (i.e. protocol);
>
> Not if you mean "model" in the sense of "mutable store" ala
> "file" or "document".

True, when the only handle you have is a URL (to a file say), then 
that's fine. However, with resolvers, we're able to express a file or 
data store as an RDF graph (i.e. a TKS/Kowari model) and make it 
available to the query engine. So, to confuse the point, you could 
represent a file containing RDF/XML as a file and query it as such, or 
an MP3 file could fronted by a resolver and queried as a graph (i.e. 
TKS/Kowari model).

> And if you mean something immutable, it seems a lot simpler to use
> a hash.
>
>> o Names should be independent of the hostname of the machine they're
>> stored on;
>
> Yes, but DNS names meet that criterium. (hint: which machine
> is www.w3.org? Which continent, for that matter? A: many.)

True.

>> <snip/>
>> For example:
>>
>> rmi://knuth.bne.tucanatech.com/server1#foo
>> beep://knuth.bne.tucanatech.com/server1#foo
>>
>> This confuses two distinct properties of a model: the model name and
>> how to access the model.
>
> I'm not at all sure that's the case.

Let's take the first example. When the model is referenced in a query:

select $x $y $z
from <rmi://knuth.bne.tucanatech.com/server1#foo>
where $x $y $z ;

The model "name" is effectively 
"rmi://knuth.bne.tucanatech.com/server1#foo".

What this is currently saying is:

o That there exists a model called "foo";
o The model lives in the TKS/Kowari server "server1";
o The TKS/Kowari server is located on the host 
"knuth.bne.tucanatech.com"; and
o You use RMI to get to it.

I think that all these except the model name do not need to be in the 
name. If someone wants to put this information into a name, fine, but 
currently we're forcing them to, and not allowing any choice.

>> TKS/Kowari stores model information in a special place called the
>> system model. This model stores information such as the type of a
>> model, e.g. a full text model or a native "Tucana" model. As far as
>> Tucana is concerned, we want to solve the problems outlined above, but
>> also allow for backwards compatibility with existing queries. As we
>> want to be able to store model information in RDF, and make statements
>> about models, we need to use a resource to reference a model. We're
>> considering using URNs to name models.
>>
>> Our proposed new model naming scheme would use the system model to
>> contain information about the models including their name and how to
>> access them. So for example we could be storing the following triples:
>>
>> <urn:slashdot-org:news-rss> <access> <http://slashdot.org/news.rss>
>> <urn:slashdot-org:news-rss> <updateEvery> <5 minutes>
>> <urn:knuth-com:server1:foo> <access> <rmi://localhost/server1>
>> <urn:tucanatech-com:server1:hr> <access> <rmi://localhost/server1>
>
> Why not
>   <http://example/someplace#slashdotNews> <access> ...
> ?

No reason, that's just the scheme someone came up with a while ago. I 
don't know if it's still valid, but no one has told me otherwise, so I 
assume it is.

>> The first model <urn:slashdot-org:news.rss> is an remote model that 
>> you
>> access using the given URL (http://slashdot.org/news.rss) and you
>> update it every 5 minutes.
>
> Aha... so you _do_ mean something mutable.

Sure do!

>> <snip/>
> Please take a look at
>
>   The Myth of Names and Addresses
>   http://www.w3.org/DesignIssues/NameMyth
>
> and see if you still think urn: helps at all.

Hmmm not directly, but I see the point. I shall bring it up with the 
guys next I speak with them.

Thanks for the comments!!!

Cheers,
Tom
-- 
Tom Adams                  | Tucana Technologies, Inc.
Support Engineer           |   Office: +1 703 871 5312
tom@tucanatech.com         |     Cell: +1 571 594 0847
http://www.tucanatech.com  |      Fax: +1 877 290 6687
------------------------------------------------------
Received on Tuesday, 23 November 2004 14:51:22 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:21 GMT