Friday, October 12, 2007

Cataloging as Industry

Something pointed me to this paper by Alan Danskin of the British Library:
Tomorrow never knows: the end of cataloging?

It has some well-spoken statements about the great increase in materials, the need to collaborate better with others in the publishing supply chain, etc. But what really stood out for me was this:
The future of cataloguing depends on transforming the process from a craft into an industry.

He qualifies this by saying
This requires unambiguous identification at different levels of granularity to facilitate repurposing of metadata created at the different stages of the process of creating and publishing resources. It also means we may have to be less precious about some of our cherished practices.

I can't disagree with what he says here, but I must say that I have a different take on the idea of industrialization of cataloging, and that is that we should consider taking cataloging out of the library and giving it to others who will actually industrialize it. Just as we don't hand craft our own library shelves, and we don't hand craft our own library systems, perhaps we shouldn't be hand-crafting our own catalog records.

What I refer to here would probably come under the rubric of "outsourcing," some of which already takes place, especially for works in less common or more difficult languages. But what if, just what if, someone could develop a cataloging service that was cheaper than what libraries can do themselves, and had comparable quality? Is there any reason why we shouldn't go for it?

4 comments:

Anonymous said...

Link to the article needs to be fixed. Should be http://www.ifla.org/IV/ifla72/papers/102-Danskin-en.pdf

Karen Coyle said...

Thanks! Done.

Anonymous said...

While I agree that cataloging could benefit greatly from industrialization, do we really want cataloging to go the way of library systems? Libraries are increasingly held hostage by the fact that they don't control their own systems, and the vendors who do control our systems are not willing or able to keep up with our needs. And since system/software development is a lot more lucrative than metadata creation and management, I don't hold out much hope for getting metadata of acceptable quality out of vendors unless libraries are willing/able to finance the industry. Doubtful at best.

Elaine Sanchez said...

I wrote this in 1998 in a list called Cristal-Ed. What I wrote then is still applicable now. Cataloging is already an industry, of industrious folks who craft bibliographic records of timeless quality in relation to the existing works in the universe, and in our online catalog. There is more to do now, with the same or fewer people, and cooperative cataloging has decreased in quality. So automation of some tasks is necessary, but the bottom line is to continue high quality cataloging, and for standards to be enforced. OCLC, level 3 records, vendor input, LC losing quality and catalogers: we're on a downhill slope. We have to keep our ideals high, and do the best we can without shooting ourselves in the foot by loosening standards or calling for the industrialization, the One-piece-fits-all of cataloging.

Here are my 1998 thoughts:
What I write is as an academic librarian. I realize that other types of libraries have different needs, and different staff resources. We are fortunate to have staff and to have the recognized and understood requirements to provide the best access and description to, and security for, all our materials.

1. We do authority control in-house:

[Note: Although now, we are looking at outsourcing authority control for DLC cataloging, the statement is still true]

Catalogers/classified staff do their own authorities, our database management unit does global changes and subject authorities. We improve the cross-reference structure on authority records by adding and deleting references, often requested by our reference librarians. We provide standardized access points by retrospectively fixing headings used inconsistently, incorrectly, or which have changed in some regard. We have an excellent online authority database, which provides a strong, ever-changing with user needs, cross-referencing structure that allows collocation of access points. This is what catalogers do. It is a local function - vendors cannot interact with patrons and reference staff to improve access based on customer feedback.

2. We do our own labeling of materials in-house:

Classified staff in the various units create contents labels, acid-free strips for special collections materials, and our database management unit creates spine labels. The materials are our responsibility. How we identify them, both for patron use and for inventory control, affects their shelving and their security. We possibly could outsource spine labels for materials which require no special processing (books), [if titles in series were reviewed for possible classed-together titles, and if we could figure out some way to tell a vendor how to consistently handle long music call numbers and other complexities involving series and subseries;] but the unique labels we would still have to create locally. Vendors cannot supply the local labeling needs for all materials.

3. We do our own item record creation of materials in-house:

Classified staff create the inventory of library materials, and provide the item database which allows materials to have a unique barcode number with which to circulate items. We add item notes to these records to document special materials contained in the item, which otherwise might be undocumented, and thus could be easily lost. This is a local function. A vendor could not easily perform this task.

4. We catalog and classify materials:

We not only closely review original, we also review copy, and, to a lesser extent, LC copy. We are in charge of the quality of our database, and what we put in affects our customers access to materials. We provide standard full-level description for materials we acquire, and routinely enhance, [enrich and upgrade] copy found on OCLC by upgrading to AACR2, adding access points that are useful in our collection, and relating records to materials previously cataloged. This is a local function, and vendors cannot do this.
We classify based on what is already existing in our collection, and anticipate change in classification, which always happens as knowledge expands and requires shifting of the organization of this knowledge. We provide the link between the two, so that patrons can more easily find all materials, whether classed together or separately. Our bibliographic database is excellent, and is a cohesive, relational database. That's what catalogers do, and it is done locally. A vendor can provide cataloging, but cannot relate it to materials already existing in the database. That is a local function.

Are these local cataloging functions worth the time and expense of having a local staff of fulltime catalogers/classified staff to perform them?

Yes, if the quality of the bibliographic, authority, and item databases is important to the library because it affects the effectiveness of a patron's search for his or her informational needs, the unique needs of the institution, and the security of materials.