[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Metadata] Academic involvement in Metadata
What you're saying is: academic staff currently use metadata in a given
form (standard bibliographic data) which is adequate for their needs. So is
it useful to either (a) enhance this with additional metadata from other
sources, or (b) reorganise it using a different structure - like Dublin
Core - when there appears to be no demand. The latter seems to be what you
are intending. Have I got that right?
If so I'd say the following. On one level, no, absolutely not. In the
INDECS project we have established the very useful principle of "functional
granularity" - in other words, only identify something when you need to. If
you are only using your data for academic discovery purposes and it does
the job, ignore all other solutions, however elegant or interesting.
Life (and money) is too short.
But...and its a very big but. The question for anyone creating metadata
these days is whether in creating their metadata they are confident of the
limits of its use. If your "bibliographic" metadata is, or is going to be,
on the Web and used by people (and data mining software) who rely on other
metadata or schemas, like Dublin Core, then you'd better include it. Not
knowing your circumstance I've no idea whether that matters to you at the
moment or not.
Certainly for anyone producing "definitive" metadata about new resources it
matters, and this is at the heart of the DOI metadata proposals and of the
INDECS project. Now that metadata has "make once, use many" functions, a
broader framework beyond "academic", "commercial", "entertainment" etc
categories becomes essential as we cannot afford either to go on recreating
metadata in lots of different ways, or maintaining lots of unrelated
different versions.
Its also at the heart of the current difficulties Dublin Core is working
through. DC started off "only" being for discovery, and now finds that it
is also being used for other purposes for which it was not adequately
designed, so needs to be re-plumbed. Personally I wouldn't make any
significant investment in the DC-Simple categories for any serious resource
metadata which I expected to have around for any length of time as it
contains too many elements which are ambiguous or overlapping and which I
believe will require re-classifying when Dublin Core version 2, based on an
entity relationship model, emerges in due course. For more ephemeral stuff
like news stories on Netscape I think DC-Simple is fine.
For intermediaries it is a difficult call, and until the new standards are
established I'm sure you need to be cautious about implementing any
emerging standard if it's not of immediate and obvious benefit.
This point underlines again how critical it is for the DOI system to
implement well-formed metadata at the point of original DOI allocation and
pass that down through the chain.
Godfrey
At 11:37 AM 2/12/99 -0000, you wrote:
>The joys of terminology! Sorry to be confusing. The point I'm coming from
>is that we are considering incorporating dublin core elements into a
>catalogue, which academics can search to select materials for reading lists.
>So we are not using metadata per se, rather we are using the elements to
>enhance standard bibliographic data. All well and good, but it is not clear
>that there is any demand from academic staff for this enhanced information.
>It's this last point I'm interested in - theoretically speaking, is there
>justification for doing this if there is no demand. (Practically speaking, I
>think there are good reasons for doing it, but I can't find any data from
>academic staff to support it!)
>
>Helen
>
>> ----------
>> From: Godfrey Rust[SMTP:godfreyrust@dds.netkonect.co.uk]
>> Sent: 12 February 1999 19:18
>> To: metadata@doi.org
>> Subject: Re: [Metadata] Academic involvement in Metadata
>>
>> I'm a little puzzled by this, but I expect its a difference of
>> terminology.
>>
>> In the DOI approach (and all the other main initiatives, as far as I can
>> tell), all identifiers, titles or descriptive elements of any kind are
>> "metadata", so it's impossible to search for and select resources without
>> it. People use metadata constantly, whether they realise it or not. The
>> questions are about the organisation, structure and location of metadata,
>> not whether or not its a good thing.
>>
>> It reminds me of the comment I heard a few weeks ago that "metadata is all
>> very well, but it'll never replace cataloguing". Metadata's just a new
>> word
>> for a very old set of problems that have now become of central importance
>> because of the move to digital.
>>
>> Does that makes sense, Helen, or am I missing a point here?
>>
>> Godfrey
>>
>>
>
>------------------------------------------------------
>Metadata maillist - Metadata@doi.org
>http://www.doi.org/mailman/listinfo/metadata
>
>
Godfrey Rust
..........................................
Data Definitions
14 Gloucester Road
London W5 4JB
T (44) 181 567 1047
F (44) 181 579 0938
Mobile 07775 908398
------------------------------------------------------
Metadata maillist - Metadata@doi.org
http://www.doi.org/mailman/listinfo/metadata