[topicmapmail] PSIs - alternatives

[Murray Altheim]

Quoting Alexander Johannesen <alexander.johannesen at gmail.com>:

> Hiya,
>
> On 6/21/06, Murray Altheim <murray06 at altheim.com> wrote:
>> I agree with Steve -- PSIs are fine.
>
> And I agree too that PSI's are fine. It's that resolver thing I've got
> a problem with.

Not me. I think issues of resolution are what?  ...context-dependent.

>> "Breaking" over time is not necessarily a bad thing. But I'd question
>> the meaning of "breaking", i.e., the URI is still viable as an identifier
>> in 1000 years, even if the domain doesn't resolve.
>
> Wasn't that Steve's point, though, that if it doesn't resolve to
> something, it ain't a PSI anymore? What in fact does 'published' mean?

I think "published" is also context-dependent. If I publish something
for you and you receive it, it's "published." If Steve doesn't, and I
didn't explicitly mean for him to receive it (even if it happens to be
available to him), then "published" works for you and I but not for
him. That's publication in general, and applies here as well. If I'm
say, publishing something for the library or the medical communities,
I would choose a suitable publishing medium. If I'm not a member of
the library or medical communities I probably can't publish something
to JoDI or the AMA journal (resp.), which would under any circumstance
be appropriate restrictions. Authority is gained through community
vetting of content. Outside of that, anyone can of course "publish"
whatever they like. Staple it to a utility pole for all that's worth.
As for the "mundane" world, so for the networked one.

>> For example, one web service could
>> query another via OAI-PMH and receive any available PSIs or PSI sets
>> matching a query. There's a ton of work being done on federated
>> searching within digital libraries.
>
> Sounds like we need a set of PSI's for PSI's. :)

Only within the context of use that might require such a thing. For
those who are interested in the sort of "meta-PSIs" we've been
alluding to, that might be appropriate part of the ISO Topic Map
work. There was the OASIS group for this purpose at one point.

>> [...] there's an great deal of both research
>> and development being done within the library community, and almost
>> all of the issues being tackled here (such as subject identity,
>> identification schemes, federation of identity, classification,
>> classification systems, etc.) and in related areas by the W3C
>> have been under consideration for many years within the library
>> community. There's little point in reinventing the wheel here.
>
> Maybe that's why I feel it's been discussed ad naseum ; I work for a
> library, and all they talk about here is PI, federation,
> classification, repositories and DRM. Sorry to bring my assumptions
> along to this group. :)

Same here, but I don't get nauseous about it; it seems like part
and parcel of the kinds of things that libraries have traditionally
and appropriately taken on in their roles of information management,
and I am happy whenever I hear of the library community stepping up
to the plate of that responsibility. I don't think of ISO, OASIS,
the W3C or the AI/KR/OntEng academic community to be equipped to
deal with this kind of thing -- the expertise is obviously in the
library community. Figuring out how to get librarians and computer
scientists to talk to each other is probably a large part of our
tasks.

Murray

...........................................................................
Murray Altheim <murray06 at altheim.com>                              ===  = =
http://www.altheim.com/murray/                                     = =  ===
SGML Grease Monkey, Banjo Player, Wantanabe Zen Monk               = =  = =

       In the evening
       The rice leaves in the garden
       Rustle in the autumn wind
       That blows through my reed hut.  -- Minamoto no Tsunenobu