W3C home > Mailing lists > Public > public-esw-thes@w3.org > February 2009

Re: RE : broader and broaderMatch... and BT???

From: Stephen Bounds <km@bounds.net.au>
Date: Sun, 15 Feb 2009 23:24:15 +1100
Message-ID: <4998096F.3060808@bounds.net.au>
To: Antoine Isaac <Antoine.Isaac@KB.nl>
CC: public-esw-thes@w3.org
Hi Antoine,

I think the wiki is perfectly fine as a solution, and it's great that it 
is already under consideration.

Given that most people requesting extensions to SKOS seem to have fairly 
specific goals in mind (eg BS 8723 compliance), it might be worth 
presenting the wiki in two broad sections:

* registered extensions to SKOS core
* "cookbook" recipes illustrating recommended ways to meet various use 
cases with SKOS core + zero or more SKOS extensions

The only question remaining is how much change control is necessary over 
extensions themselves.  We don't want to stifle innovation, particularly 
in the early years of SKOS.  However, I can see the need for popular 
extensions to be transitioned to a W3C "extension namespace" down the 
track to ensure stability and interoperability.


-- Stephen.

Antoine Isaac wrote:
> Hi everyone,
> I fully agree with Alistair, and am very curious to hear about what you 
> (especially Johan, Christophe and Stephen, in this thread) would like to 
> have as a solution!
> I'd like to answer on Johan's first point in [1]: yes, these best 
> practices/extension may be closely related to different types of KOSs to 
> represent. You can find a a good example of this during our last comment 
> period, if you look at issues 181 to 186 in [2].
> Cheers,
> Antoine
> [1] http://lists.w3.org/Archives/Public/public-esw-thes/2009Feb/0030.html
> [2] http://www.w3.org/2006/07/SWD/track/issues/closed
> -------- Message d'origine--------
> De: public-esw-thes-request@w3.org de la part de Alistair Miles
> Date: ven. 13/02/2009 15:41
> : Stephen Bounds
> Cc: public-esw-thes@w3.org
> Objet : Re: broader and broaderMatch... and BT???
> Hi Stephen,
> Yes, this is a great idea. I brought this up withi the WG a while ago,
> the WG has been focused on reaching candidate rec so hasn't discussed
> it yet, but Tom Baker has been very diligent in making sure it stays
> on the agenda.
> Do you think a wiki page on the W3C site would be enough to act as a
> "registry" of SKOS extensions? Or do we need more than that?
> Cheers,
> Alistair
> On Fri, Feb 13, 2009 at 07:52:26AM +1100, Stephen Bounds wrote:
>  >
>  > Hi Alistair,
>  >
>  > I agree that it's probably too late for changes to be considered to the 
>  > core SKOS language.  However, making the extensibility of SKOS 
>  > explicitly part of the standard could be a really important benefit.
>  >
>  > These days, there is a common trend in open-source packages (eg Drupal, 
>  > Firefox) for there to be a place for extensions to be registered by the 
>  > *owner* of the package (not just hoping that people can publish on their 
>  > own web pages and get a sufficient rank on Google).
>  >
>  > Doing this reduces the proliferation of competing extensions to some 
>  > extent and focuses community effort.
>  >
>  > SKOS sits somewhere between a technology (eg XML) and a software 
>  > package.  I believe it could benefit from a similarly endorsed community 
>  > extension model.
>  >
>  > This could be achieved simply by inserting something into the 
>  > Recommendation like: "The SKOS standard was built was extensibility in 
>  > mind.  A list of unofficial community extensions to SKOS can be found at 
>  > http://w3c.org/xxx."  Then, just give people a way for people to submit 
>  > their extensions to the SKOS group for listing on a W3C web page.
>  >
>  > Even better would be the ability to submit under an "extension" subtype 
>  > of the W3C, eg http://www.w3.org/2008/05/x-skos/iso#bt.  I can 
>  > appreciate that this might be politically difficult to arrange, however.
>  >
>  > Cheers,
>  >
>  > -- Stephen.
Received on Sunday, 15 February 2009 12:25:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:39:03 GMT