On a related note - though the idea to introduce some kind of "latest update date a member of the list" property is something to consider, as I stated earlier - each scheme has an atom feed which orders the members of the scheme in descending chronological order based on last modified date.
Unfortunately, I didn't really have dates when creating the Relators [1] or Countries [2] lists (the MARC CodeLists, in general), so I basically made the date Jan 1 1970. This is hardly perfect, but serves as a baseline and suggests their lengthy, pre-ID existence. It also means that accessing the atom feed for those schemes is not terribly illuminating insofar as there have been no changes since they were published at ID. However, the ISO language codes do have last modified dates, so we were able to use those dates and work with them. See, for example, the ISO639-1 feed [3] or the ISO639-2 feed [4].
Rgds,
Kevin
[1] http://id.loc.gov/vocabulary/relators/feed/1
[2] http://id.loc.gov/vocabulary/countries/feed/1
[3] http://id.loc.gov/vocabulary/iso639-1/feed/1
[4] http://id.loc.gov/vocabulary/iso639-2/feed/1
> -----Original Message-----
> From: Authorities and Vocabularies Service Discussion List
> [mailto:[log in to unmask]] On Behalf Of Karen Coyle
> Sent: Wednesday, January 04, 2012 9:20 AM
> To: [log in to unmask]
> Subject: Re: [ID.LOC.GOV] Update dates
>
> Quoting Rebecca Guenther <[log in to unmask]>:
>
> > It probably could be argued either way. I would argue that the date
> of
> > the vocabulary (which is a last modified date) is the date when any
> > change was made to the description of the vocabulary as a whole.
> > For instance MARC relators has a URI for the vocabulary as a whole
> > with metadata about that vocabulary
> > (http://id.loc.gov/vocabulary/relators.html). Each entry in the
> > vocabulary has its own metadata and its own date. So I would suggest
> > that it be the date of any change to the description of the overall
> > vocabulary rather than to any of its parts. It may be more
> > straightforward in terms of maintenance.
>
> Rebecca, I would hope that rarely do vocabularies as a whole get
> modified -- in fact, it's hard to think of why that would happen once
> the basic "beta" versions have been gotten out of the way. Such a
> change would probably result in a major version change (e.g. from 1.0
> to 2.0).
>
> If a person wants to know if this vocabulary is up-to-date, or to know
> if it is volatile or stable, there needs to be some visible clue. Even
> programmatically you would need to pass through an entire vocabulary
> (and some have tens of thousands of entries) to find a "latest update
> date".
>
> The important thing is to have any dates clearly defined. Not just
> "date" but "latest update date a member of the list."
>
> kc
>
> >
> > Rebecca
> >
> >
> > On Jan 3, 2012, at 3:05 PM, Diane Hillmann wrote:
> >
> >> Karen:
> >>
> >> I think you're on to something important here. It seems to me that
> >> there are valid use cases for both a date of the vocabulary as a
> >> whole and for the individual parts of it, so expressing that in the
> >> form of a 'best practice' is extremely useful. And then there's the
> >> question of what the 'date of the vocabulary' should consist
> >> of--should it be the latest date of any of its component parts, or
> >> something else?
> >>
> >> Thanks for bringing this up!
> >>
> >> Diane
> >>
> >> On Tue, Jan 3, 2012 at 10:57 AM, Karen Coyle <[log in to unmask]>
> wrote:
> >>
> >> BTW, I asked because I'm including some of these in a publication
> and
> >> am trying to give readers an idea of the "freshness" of various
> >> vocabularies so I was looking for update dates on them. I think it's
> >> very helpful to give a general "latest update" on the list as a
> whole
> >> because a visitor to the site couldn't determine that by looking at
> >> individual entries. Kind of like the update date on a home page that
> >> reflects the entire web site. I think we're on our way to developing
> >> a "best practice". :-)
> >>
> >
> >
>
>
>
> --
> Karen Coyle
> [log in to unmask] http://kcoyle.net
> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
|