Hi Matthias,
On Jan 28, 2006, at 4:56 AM, Matthias Basler wrote:
Just some quick comments:
1. I am missings "handbook/manual" in your type list. And when you're
at it, you
could add "painting" as well... and "RSS feed"?
In the end you'll have to acknowledge that new types of documents may
arise in
the future and that certain people always have specific types of
document that
you, I and all the others cannot foresee.
Yes.
One thing to keep in mind is that there is a hierarchy to the model. So
say I have a newsletter, which is not there. I can still use
"periodical" and it will be accurate. It's worth nothing there is no
suitable superclass for painting though, so I should probably fix that.
So I would definitely not want an RSS feed class; that's too specific.
It mixes format (RSS) and type/class. Also, is it the feed one is
citing, or the thing that feed represents? E.g. I have feeds for
newspapers, and often cite the articles that I get from those. I never
record the feed information, though.
In general I need to think more about internet document types.
BTW, there's another cool detail worth noting from an RDF standpoint: I
have defined some of the key classes in relation to the new FRBR
ontology, so it would allow this stuff to be plugged into a more
general view.
2. Writing a list of types in one language is one thing. But please
bear in mind
that you'll have to find a suitable (uniqe!) translation for each type
for any
other common language.
Yes, if you look at the RDF schema, you'll see the text descriptions
(rdfs:label and rdfs:comment) are all tagged with an xml:lang
attribute. E.g. I welcome translations :-)
The documentation is generated by XSLT directly from the schema.
3. My idea about this is an internet repository where people can
upload new
proposed document types AND their translations for certain languages.
Someone
will have to look over them, ensure they are correctly translated and
approve
them for public usage.
.. A sort of standardization process, so to speak.
A good idea, and sort of what I was thinking of here. E.g. we could use
some agreed model to facilitate interoperability, enable better future
tools, etc. And we hopefully will be able to use this directly in
OOoBib.
I registered the purl to keep a permanent uri, but allow where I put
the stuff to be flexible. So as a start I will put it at the xbiblio
Sourceforge site. Am not sure how to deal with longer term management;
for now it'll probably have to be people emailing me suggestions and
comments.
Bruce
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
