«Part II Submissions by: Judy Anderson Heather S. Miller Bryan Baldus Allen Mullen Deanna Briggs John J. Riemer Diana Brooking Karina Ricker Lloyd ...»
3) Various national and international organizations are responsible for developing, maintaining, and supporting structures and standards for bibliographic control. Do these organizational arrangements support current needs? How could they be improved? I’m sure this comment must have been made before, but I think they could be improved in terms of efficiency. For instance, I knew that last summer ISSN International had a long delay with updating their database (ISSN Portal) with changes and updates received from various international ISSN centres. I think NISO has some great goals, in regards to attempts to decrease the amount of time it takes to create relevant standard. In the last year, CCC has become a member of BISG. When I worked in a library, I probably wouldn’t have thought of BISG as an organization involved with bibliographic control, yet they are aiding in this effort. (I hope that I may be a participant on one of their committees in the near future.) The library community needs to work more broadly with related organizations to improve bibliographic control. This may also lower costs if these other organizations (in publishing, for instance) have more financial means than library‐only/primarily library‐only standards orgs.
4) A recurrent theme of the previous meetings was more fully integrating bibliographic data (such as MARC records, terminologies, authority files, et al.), which currently exist as “data silos,” into the fabric of the World Wide Web. In particular, terminologies and authorities were seen as important resources that could be used in a variety of ways. From a design perspective, how do we move from “data silos” to “data services,” that increase the potential value of bibliographic data by treating them as interconnected resource collections, addressable via URIs and accessible over Web protocols? Organizationally, how might this goal be accomplished, supported, and maintained? Economically, what factors need to be considered? Just as we have SaaS model, think DaaS, i.e. Data as a Service. We desperately need to build our own (library) influence on the web and integrate/partner with others on the web (non‐library) who already have distinct clout, not necessarily due to their bibliographic control practices, but nevertheless, they use some bibliographic control to a certain extent to achieve their means. E.g. I admit it, I look up a book in Amazon before I go to any OPAC. Why? It is easy and I like the features that Amazon provides me. But since my funds are not unlimited, I then go to the OPAC and place and reserve on that book in my local public library system. I wish I could place a reserve on a local copy of a book directly from Amazon. Of course, there are competition issues of sorts to be hashed out, but I think there is a place for partnerships wherein the cooperating parties have very different goals. It simply may take more time and thought to figure out the nuances and the “how” of the partnership to ensure that all parties receive desired benefits. Another example of a partnership would be directly with publishers, like HarperCollins, who creates widgets for books. These widgets are placed on author blogs (or anywhere I suppose) and through that widget link on a blog you can sample the full‐text from HarperCollins digital repository. Why not integrate so the widget could include a link for me to place a reserve against that book in my local library system? Or how about the other way around (link to your available “local holdings” from HarperCollins). Right, HarperCollins wants sales and profits. But there can be mutual benefit. I place a reserve from the widget and read the book from my library. I like the book and I tell my family/friends/colleagues about it via word of mouth and a link with the widget. The widget delivers options to my family/friends/colleagues, some of whom make a purchase (redirect to Amazon, I don’t think HarperCollins sells directly) or borrow from the library. I think there will always be people who want to buy the book because they a.) don’t like waiting to get it from another library b.) want to mangle it, mark it up, etc. c.) have on hand for local reference and people who want to borrow from the library because a.) they don’t want to spend money on the book when they could get it for “free,” b.) want to take advantage of the library as a public service and c.) the library contains other materials that the user wishes to access as well. This partnership brings all kinds of users initially to the same place and lets the user choose how he/she wants to access the material, if desired. This is powerful. Case in point: I am a patron who a.) uses the public library quite a lot, I go there several times a week for reading materials and b.) also buys a ton of books (from Amazon) because there are some books that I want to own, mark‐up, give as gifts, and simply re‐read without having time‐limit to return. Still, there are many times where I first read a book from a library, then purchase my own copy because I want it for any number of reasons. If the content is good, the publisher won’t lose a sale. We can’t wait “forever” to partner with other influential parties (as I believe Credence Clearwater Revival sang, “Someday never comes”) to let the world know that the records and metadata in our library catalogs is valuable in 2007 and beyond.
5) Library of Congress cataloging and its support for terminologies and authorities are central to the apparatus of bibliographic control in the U.S. and beyond. What should the role of the Library of Congress be in this developing environment? LC should not be afraid to partner with others who are working on development in strategically important areas (e.g. OCLC, with Worldcat Identities) to increase its value and share costs. But I suspect OCLC may not want to partner? To agree with Martha Yee, we do need to continue this important work. Without authority files, even the easiest shared‐characteristic search (a basic requirement, in my mind, of a decent catalog), in many catalogs, would be so far from complete. I haven’t given this much thought, but are there other organizations (besides LC) to which some authority functions could be continued and maintained? i.e. Cost distribution through multiple organizations? I thought I heard (somewhere?) that LC was considering simplification of subject headings (shorter pre‐ coordination, perhaps?). This might not be such a bad idea, given the first meeting about users who in most cases don’t require the kitchen sink (the complete MARC record that we analyze to death), but need more like “good enough” (the ONIX record, from a vendor, mapped to MARC and inserted with little or no augmentation). LC should continue to support subjects, but perhaps in some kind of tiered structure because we don’t need to create long pre‐coordinated subjects for every item a user would like to access.
From: D. Brooking firstname.lastname@example.org Date: Jun 28, 2007 1:49 PM Subject:
Comments for Working Group To: email@example.com
Dear Dr. Griffiths,
Thank you for accepting comments on your group's work. I have several different comments to make, they are numbered and listed below. Overall I am very uneasy about what I see as the dishonesty in claiming that we no longer need something (like bibliographic control) as opposed to admitting we still need it, but maybe can't always afford it or need to do it in a different way. This always seems to happen when a new technology arises. There can always be some categories of materials where a library decides, this isn't important enough for full bibliographic control, given economic realities. But we need to be very clear about what we are giving up in those cases.
1. The Future of Bibliographic Control--for the Library of Congress or for all of us?
I am concerned that conditions unique to the Library of Congress could be driving this process. When your report and recommendations come out, they will possibly be very influential throughout the library community. And I see your charge is much broader than just making recommendations to LC.
The Library of Congress is in many ways behind the rest of us when it comes to efficient workflows and productivity. I won't go into the reasons, in fact, I don't know all the reasons, it's just the way it is. For example, "whole book cataloging" is new concept for them that still hasn't made its way completely through their organization. I have read reports before from other institutions that characterize all cataloging based on their own situations. I read these reports and think, "My God, we haven't done it that way in years!
We could never afford that."
The LC series decision is a good example of the fallout of one library's problems affecting all the rest of us in a very negative way. Another example is Wright State University's decision to disband their cataloging department and outsource everything.
This decision was based on severe internal management problems, but was actively promoted as a progressive and daring move into the "future" of cataloging that all libraries should follow.
You need to keep this very much in mind when you draft your final documents. No library likes to have their dirty laundry made public, but glossing over the real problems can have disastrous effects for all of us.
2. Quality control (based on Clifford Lynch's questions at May 9 meeting) I think it is very hard to balance the costs and benefits of quality control in cataloging. It is not the same as making cars or doing surgery. No one dies if our quality is not up to snuff. What happens is people will not be able to find resources. And we have seen over and over again that our users don't miss what they can't find, unless they have some reason for knowing ahead of time that it should be there. And most of our users don't know what should be there and what is missing. This makes it very difficult to come up with objective measures of the benefits of quality control for our end users. The costs of doing quality control are more obvious.
There are of course costs associated with bad data. Garbage in, garbage out. It is less costly to do it right the first time than to try to correct large amounts of legacy data after the fact. The problem is in determining how good is good enough.
And we just don't catalog for the people today, we catalog for people a hundred years from now. There are stewardship issues here that need to be recognized. That's why Clifford's comment that "systems can limit the data that users see, and therefore the data that are usable" disturbs me so much. He should know, *systems change.* Creating data based closely on today's system limitations is a bad idea. We now see OCLC trying to implement a less than perfect FRBR algorithm. Wouldn't it be easier if there were better data, more uniform titles? Maybe no one could have foreseen the need to catalog for FRBR, but in fact better quality control on the existing rules we had at the time would now be providing better data for this new need.
3. Working with other communities on standards
I think there are practical limits to how many different communities we can work with.
It is easier if these communities share similar goals to the library community. So for instance, museums and archives seem to be natural partners for us in the creation of standards. There will be times when there are good reasons for library standards to be different, if we have important goals or needs that apply just to us. We shouldn't get too caught up in trying to please other people.
On the other hand, there is no point in being different just to be different. The original draft of RDA had very different descriptive rules for unpublished manuscripts than the archives community has. And there was no point in that. (Luckily, RDA has changed those rules now thanks to archival catalogers in libraries.) It's a balancing act.
4. Text strings, labels, identifiers
Clifford asks about creating such things as "author identifiers for name disambiguation" in our vocabularies. In fact, we already have OCLC authority record numbers and LC control numbers for each name that could be made use of right now. Aren't those numeric identifiers? I am not sure what Clifford thinks is missing.
I fear that we may make numbers a fetish in some way, the magic bullet that solves all problems. People seem to be turning this into an either/or thing. At some point someone will need to associate one or more text strings with these so-called identifiers.
(And I admit all the terminology surrounding this is really confusing. Identifier vs. label vs. attribute, etc.) End users want to see something they can read.
Author: 49584903 Title: 548395834054839 I can't imagine that's what is intended, but often the people who speak in favor of such identifiers then also claim that we don't need to worry about primary access points anymore, we don't need to have an established textual heading, we don't need rules for citations, etc.