> > Something like "Ann. Math., II. Ser. 141, No.3, 443-551 (1995)" obviously*identifies* (rather than "defines") the article in question, so it should go
> > into DC.Identifier. (The article above is not [yet] online available.)
> >
> > DC.Source should give "information about a second resource from which the
> > present resource is derived. [...] Source is not applicable if the present resource
> > is in its original form."
> > (The above article is in its original form, so you cannot put the stuff above into
> > DC.Source.)
> >
>
> I guess it depends on how you look at it. I thought that the original source is
> the paper version of the journal article, defined by the <SOURCE> metatdata
> above. The 'second' resource is the electronic version. I agree that this
> could also be the DC.Identifier metadata for the second resource. For that we
> have used the SICI number, and also a file path identifying the electronic
> version.
You're right, the paper version is the original.DC.Source seems (at least to me, after reading the description of DC.Source)
not to be the right place for this kind of information!
As I've said, there's no electronic version of that article now.
I think the above reference to the article should go somehow into DC.Identifier.
SICIs might be a good choice for formal references to put into DC.Identifier,
but I tend to agree with the comments from Cliff Morgan that also a human
readable (and perhaps not-quite-formal) information should be given.
Not just ISSNs that appear in SICI codes.
People tend to rate the quality of a journal article by the name of the journal.
For example, mathematics people normally know that articles published in Ann. Math.
are more important than those in Bull. Number Theory.
Now, what about ISSN 1071-0957 and ISSN 0003-486X?
And how can you search for an article by that appeared in Ann. Math.
if you don't know their ISSN and the journal title appears nowhere in the DC
data?
Regards,
- Michael
--
Michael Jost
Fachinformationszentrum Karlsruhe
Abteilung Mathematik u. Informatik
Franklinstr. 11
D-10587 Berlin
[log in to unmask]
Tel (49)(30) 3923000, Fax (49)(30) 3927009
|