[topicmapmail] eXchangeable Faceted Metadata Language

[PeterV]

>  | I was wondering whether the relative simplicity of XFML 
>  would make it 
>  | more suitable (than XTM directly) for modeling the IA of a 
>  website, 
>  | supporting the HTML etc. and itself based on an underlying 
>  XTM, i.e. 
>  | as a layer in a website's software architecture (whether 
>  distributed 
>  | or not).
>  
>  That's rather like saying "the relative simplicity of a 
>  bicycle makes it more suitable (than a car) for getting 
>  around". It's not directly false, but it's not true in all 
>  contexts, either. XFML is simpler, and if something simplem 
>  is all you need, then it may be better or just as good to go 
>  with XFML. 
>  
>  However, XFML won't always be enough. There is a *lot* you 
>  can do with XTM that XFML doesn't even let you dream of.

I think I can clarify this. (For the record: I wrote XFML)

Lars is (as usual) right: XFML is very limited. The reason is that it is
meant to *exchange* faceted metadata.

A lot of people want a framework to build a site on, and look at XFML
and decide it's not flexible enough. They're right - it's not. I
wouldn't recommend building a site based on XFML - why limit yourself
like that? I do think there is a need for something that is more
*focused* on IA needs than XTM, to build website achitectures with, but
XFML is not it. XFML is useful if you need a format to exchange a
faceted classification system. That's really it.

Hope that helped? Feel free to ask for more details.
Peter