Tuesday, December 8, 2009

Museums Explain Their Databases

It's a good thing that museums are increasingly putting their curatorial databases online for public browsing and searching. A minor aspect of this trend is the language institutions use to explain the state of their data. This post collects some examples of the genre.

The Brooklyn Museum has the following on its opening search page:
The material presented here represents only a fraction of that rich collection. The Museum is committed to making its collections accessible to the widest possible audience, and this site is an important part of that process. It is, however, a work in progress. We intend to continue to expand the number of works of art included on the site and to update information currently posted. We are making every effort to ensure that the information provided about our collection is accurate and up-to-date, but the nature of scholarship is that there are sometimes changes in information and new discoveries. If you believe you have information we should have about any of the works you find here, we would be happy to hear from you.

The Museum of Fine Arts, Boston shows this to start:
Note that some of the electronic records indicate that they have not been reviewed recently by curatorial staff and might need revision; also, please note that a small percentage of the MFA’s collection is not presently searchable online.
The indicated MFA records have this:
This electronic record was created from historic documentation that does not necessarily reflect the MFA's complete or current knowledge about the object. Review and updating of such records is ongoing.
That text is essentially identical to what appears on the Yale Art Gallery site
Note: This electronic record was created from historic documentation that does not necessarily reflect the Yale University Art Gallery's complete or current knowledge about the object. Review and updating of such records is ongoing.
In turn, that's not so far off from the Harvard Art Museum text:
This record was created from historic documentation and may not have been reviewed by a curator; it may be inaccurate or incomplete. Our records are frequently revised and enhanced. Please contact the curatorial department listed above for more information.
[[Note that I don't give a link into the Harvard website. The (almost unbelievable) explanation is that if you click on http://www.artmuseums.harvard.edu/collection/detail.dot?objectid=289448 you get a blank page. I clicked on the link from a search results page and saw information about a sherd of Roman pottery. Is it really possible that Harvard is checking a session id or the referrer and only displaying the info as part of an existing visit to the site? If so, that's highly (highly!) lame.]]

The Metropolitan Museum in New York has the following:
Due to the extremely large number of objects in the Museum's permanent collection, not all artworks are currently available in the Collection Database. Furthermore, information contained in the database records is, in some cases, incomplete, and all information is subject to change according to ongoing research and new acquisitions.
The British Museum just puts "Noticed a mistake? Have some extra information about this object? Please contact us." on its individual object pages. That's in the same spirit as the more extended explanations from other institutions.

Overall, these texts represent a mode of sharing data that is welcome. Better to make slightly incorrect or outdated data available than to hold on tight to it. That's especially the case when there are images of the objects as well.


sgillies said...

Confirmed your experience with the Harvard site. Intentionally or not, it prevents perma-linking, and that's a bug.

Jfer said...

The Harvard Art Museum recognizes the permalink bug and is currently working to remedy the issue. You will be able to bookmark soon!

Also, please note that our URL has changed: http://www.harvardartmuseum.org.

Sebastian Heath said...

Jennifer, thanks for the response. Sorry about the extra "highly". Take it as a sign of my enthusiasm for the topic of stable URIs.

Unknown said...

In the meantime, you may use the following to link to the specific record:

Tyler said...

The National Museum of the American Indian has published scans of the index cards that used to comprise their collections records. Some of the cards are handwritten and go back 100+ years. Ex.:
Card scans are accompanied by this text: "Catalog card scans are provided to illustrate the information that originally accompanied the objects. Please recognize that unacceptable or offensive terminology represents historic data and not NMAI's current usage."

John Muccigrosso said...


As of tonight the Harvard museums site can handle something like:


Not sure where your number is from, but it still doesn't work.

Oddly, I also randomly searched another number (1980.1) and got a hit. Well, I actually got three hits for 1980.1, 1980.10 and 1980.100, which makes sense for the search engine.

However when I put that id into a URL manually, the site remembered that I had found the corresponding object before and presented it as "object 3 of 3", the other two being 1980.1 and 1980.10. Deleting all my harvard.edu cookies took care of that and then it was only "1 of 1". Finding it via another search also reset the counter.

This seems fine to me, if a bit unexpected.

Unknown said...

love to see this discussion! It’s great to see you all working through the issues and also, it’s great to see recommendations for testing. In the end, it’s what your actual users do and prefer that should be your biggest driver in making these decisions.

get level