If it was FAST, then there'd be a URI for "Greece", "History" and
"Peloponnesian War, 431-404 B.C. ", not a URI for the pre-coordinated
authority.
Additionally, I think that FAST normalizes/rationalizes -- and builds
relationships between -- geographic and temporal terms in some useful
ways that are not done in the authorities.
But I'm definitely not an expert on FAST. We have only fairly limited
text explaining FAST available to us; my understanding is that OCLC has
continued to do interesting work/research on it beyond the several year
old high-level documentation we do have, but it has not been
published/shared. Definitely the intellectual property issues of FAST
are potentially a barrier; if it's useful, I guess it could be 'cloned'
in a version meant to be released creative commons or public domain or
what have you. (This might be tricky depending on the nature of the
intellectual property involved; if we have to do a 'clean room'
re-implementation, forget it).
As far as pro's and con's -- well, you'll find people who will
vehemently argue for the _usefulness_ of the pre-coordinated nature of
LCSH. It definitely does provide some theoretical features. But it's
exactly that nature -- in particular pre-coordinated _as a string
literal_ that Ed was suggesting might give us some barriers to working
with it in flexible ways. Additional analysis of some of it's barriers,
and in some cases suggestions for structural solutions, can be found in
this article by Kelley McGrath. http://journal.code4lib.org/articles/23
Beacom, Matthew wrote:
> Not the same, but pretty close. The long LCSH strings of terms like (in MARC21) 651 0 |a Greece |x History |y Peloponnesian War, 431-404 B.C. are not constructed. Only the individual terms are present. That is (more or less) FAST, no?
>
> Matthew
>
>
> -----Original Message-----
> From: Next generation catalogs for libraries [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Jonathan Rochkind
> Sent: Tuesday, November 03, 2009 4:13 PM
> To: NGC4LIB_at_LISTSERV.ND.EDU
> Subject: Re: [NGC4LIB] Tim Berners-Lee on the Semantic Web
>
> Not quite, I don't think. I believe that FAST breaks down
> (un-coordinates) and normalizes LC authorities, it doesn't just use
> existing LC authorities. id.loc.gov just lists authorities. I don't
> think they are the same?
>
> Beacom, Matthew wrote:
>
>> The LC authorities and vocabularies project at http://id.loc.gov/authorities/ is pretty much a SKOS-based FAST version of LCSH.
>>
>> Matthew
>>
>>
>>
>>
>> -----Original Message-----
>> From: Next generation catalogs for libraries [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Jonathan Rochkind
>> Sent: Tuesday, November 03, 2009 4:01 PM
>> To: NGC4LIB_at_LISTSERV.ND.EDU
>> Subject: Re: [NGC4LIB] Tim Berners-Lee on the Semantic Web
>>
>> Turn it into FAST instead?
>>
>> I wish those working on FAST would write publically about what they're
>> doing, but I've been unable to persuade them to do so.
>>
>> Ed Jones wrote:
>>
>>
>>> LCSH is a pre-coordinated system designed for the card catalog, generating discrete coherent quasi-hierarchical strings intended to be read from left to right. Even the MARC coding is primarily designed to arrange subject cards in proper order for filing in that card catalog. LCSH is our legacy data and necessarily has at least sentimental value to those of us who grew up amidst its intricacies, but I'm increasingly unsure what we can do with it now other than use it to produce an online card catalog. The more we try to squeeze it into the Procrustean Bed of an online database--for which it was never designed--the more we seem to distort it.
>>>
>>> Ed Jones
>>>
>>> -----Original Message-----
>>> From: Next generation catalogs for libraries [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Dan Matei
>>> Sent: Tuesday, November 03, 2009 12:11 PM
>>> To: NGC4LIB_at_LISTSERV.ND.EDU
>>> Subject: Re: [NGC4LIB] Tim Berners-Lee on the Semantic Web
>>>
>>> -----Original Message-----
>>> From: Karen Coyle <lists_at_kcoyle.net>
>>> Date: Tue, 3 Nov 2009 10:24:19 -0800
>>>
>>>
>>>
>>>
>>>> Order is complicated in LCSH (as is everything else ;-)). In a given
>>>> heading, the order that the subfields are written in the MARC record
>>>> is the order that must be retained. In some cases, changing the order
>>>> actually changes the meaning of the heading. HOWEVER, the order does
>>>> not give you the facet type. So in some faceted systems, you have a
>>>> fixed number of facets and they are always in the same order:
>>>>
>>>> topic / place / time / genre (or something like that)
>>>>
>>>>
>>>>
>>> Some facets are not properties of the topic, but of the indexed resources.
>>>
>>> In the card catalogue time, it made sense (to me) to coordinate a topic with a genre (say). But now ? I would prefer
>>> (at browsing) to select a topic and then the system to offer me a faceted browsing on genres.
>>>
>>> For instance, not:
>>>
>>>
>>>
>>>
>>>> $aInternational Congress of Writers for the Defense of Culture$n(1st
>>>> :$d1935 :$cParis, France)$vFiction.
>>>>
>>>>
>>>>
>>> but:
>>>
>>> International Congress of Writers for the Defense of Culture$n(1st :$d1935 :$cParis, France)
>>>
>>> Fiction: n1
>>> History: n2
>>> etc.: n3
>>>
>>> Dan
>>>
>>>
>>>
>>>
>>
>>
>
>
Received on Tue Nov 03 2009 - 16:27:26 EST