MULLEN Allen wrote:
> I also appreciated the post, Tim. Great data. One likely implication
> is that most stand-alone catalogs may not be able to accumulate
> sufficient tags for optimal use. Which just adds to the sense that
> stand-alone catalogs are less effective and efficient than collective
> systems and services.
>
>
A 'standalone catalog' can certainly share it's tags with other
'standalone catalogs'. After all, none of our catalogs are truly
'standalone' when it comes to metadata, we certainly have lots of
experience with sharing metadata between cataloging through cooperative
cataloging. Not that the mechanisms of cooperating cataloging are what
should be used here (the current mechanisms of cooperative cataloging
probably aren't even what should be used for cooperative cataloging!),
but to remind us that in fact 'no system is an island'. A 'standalone
catalog' is perfectly capable of sharing/aggregating it's user-supplied
folksonmy-style tags with other catalogs.
Indeed, Tim has hinted that he is looking at a method for LibraryThing
to share it's user-supplied tag data with other systems that are not
LibraryThing.
Jonathan
> How might tagging work if library staff also did tagging as part of
> their work ("expert" tagging)?
>
> What I'm thinking is that subject specialists/reference staff/selectors
> could potentially be adding tags in the library world as part of their
> tasks. The volume of new acquisitions might be so large that they might
> only be able to address a portion of the new works. On the other hand,
> their tags would likely be very relevant. Catalogers generally do a
> good job of providing the topical container for work they address while
> selectors and public services staff have more of an ongoing relationship
> with users of library and library-accessible works.
>
> If this were done in a collective catalogue, there could be sufficient
> accumulation of "expert" tags applied to records for more heavily used
> resources.
>
> Another approach I would consider taking to add to the accumulation of
> relevant tags (were I a developer in either a library or non-library
> context) is to "suggest" tags that users can select based on content or
> previous tags that had been cross-matched to content of the work at hand
> and/or similar works. Say the top 5 or 10 tags that might be applied to
> the work could be suggested allowing the users to utilize these as well
> as their own tags.
>
> Allen Mullen
> Eugene Public Library
>
>
>
>> -----Original Message-----
>> From: Next generation catalogs for libraries
>> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Eric Lease Morgan
>> Sent: Wednesday, February 21, 2007 7:04 AM
>> To: NGC4LIB_at_listserv.nd.edu
>> Subject: Re: [NGC4LIB] Book tagging: Amazon and LibraryThing
>>
>> On Feb 20, 2007, at 3:41 PM, Tim Spalding wrote:
>>
>>
>>> http://www.librarything.com/thingology/2007/02/when-tags-works-and-
>>> when-they-dont.php
>>>
>>> ...Tags only work in big numbers....
>>>
>> Nice post.
>>
>> Yes, in order for tagging to work, there needs to be a high
>> number tags applied to any given item, and you need to have a
>> large number of diverse people doing the tagging. You are only
>> going to get the second thing if you allow people to tag their
>> own item, not yours.
>> These same principles are elaborated upon in James
>> Surowiecki's The Wisdom Of Crowds (www.librarything.com/work/17697).
>>
>> --
>> Eric Lease Morgan
>> University Libraries of Notre Dame
>>
>>
>
>
--
Jonathan Rochkind
Sr. Programmer/Analyst
The Sheridan Libraries
Johns Hopkins University
410.516.8886
rochkind (at) jhu.edu
Received on Wed Feb 21 2007 - 12:11:14 EST