Re: Facet algebra XFML+CAMEL

I agree with both Travis and Peter.

According to my opinion, for now it is better to keep the core XFML 
as it is. I agree that at some point we could define a format that 
will incorporate several well-tested extensions, but I don't think 
that we should call that XFML any more.

For instance, XFML+CAMEL allows publishing and exchanging 
algebraically defined compound faceted taxonomies. It goes far beyond 
the exchange of simple faceted taxonomies and faceted metadata.

However, what I think that is really missing from the current XFML 
and it should be probably added on a subsequent version of it, is the 
support of more expressive kinds of connections. I find the current 
specification quite poor and ineffective in practice.
I would like to see connections that allow expressing relationships 
between two maps 1 and 2 of the following form:

Topic_1 <= Topic_2
Topic_1 >= Topic_2
Topic_1 =  Topic_2
Topic_1 <= Query over topics of map2
Topic_1 >= Query over topics of map2
Topic_1 =  Query over topics of map2

Only in this way we can achieve effective information integration.


Regards,

Yannis



--- In [email protected], Travis Wilson 
<trav@f...> wrote:
> At 04:16 PM 9/23/2003, you wrote:
> >This could also work for other formats, for example, I know Travis 
has
> >done some groundbreaking work with types of facets. Maybe defining 
an
> >XFML format that extends XFML for those would be useful as well? 
(Then
> >again, maybe XFML just wouldn't be flexible enough??)
> 
> Not a bad idea, although I'd like to turn the question around and 
ask about 
> plans for the next version of XFML. Not that splinter projects 
aren't 
> useful (that's how we got the good versions of Unix, after all), 
but it 
> would be great to see the next XFML spec incorporate ideas that 
have been 
> helpful -- it would help set direction for faceted classification 
in general.
> 
> The facetmap XML format is already an extension of XFML, in the 
sense that 
> is not supported by basic XFML software but replicates the 
functionality of 
> XFML. It just happens to be a less friendly extension since 
facetmap and 
> XFML were originally independent simultaneous projects.
> 
> The idea of a set of XFML extensions that each extend the spec in 
different 
> ways is intriguing, and I wonder what would happen if a developer 
were to 
> want features from a couple different extensions. (Which casts the 
whole 
> set of XFML specs into a facet space! Who are we to extend faceted 
> classification specs with such an outdated single-parent hierarchy 
paradigm? :)
> 
> -trav


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/0bmwlB/TM
---------------------------------------------------------------------~->

----------
Thanks for playing. To unsubscribe from this group, send an email to:
[email protected]

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/