Archival Description

When inadequate data is transferred from a handwritten description into an online environment, there is potential for promulgation of error.

For the 2014 Cambridge Open Archives Tour (http://www.cambridgearchives.org/OpenArchives) we pulled out some of our fabulous collection items.  Among them was a recipe box.

Thanks to the need for exhibition labels, a colleague did a little background research and uncovered the true story behind it.  From the old description (below), you’d probably conclude that this was a box of recipes very much like you find in your own kitchen–things picked up over the years, family favorites, and aspirational dishes that you’ve never even tried.

Here’s what the old record looked like. Click the link in the call number to see the current version.

Author :	Greenough, Marietta McPherson.
Title :	Recipe file, ca. 1920.
Location :	Harvard University Archives HUG 1436.40
History notes :	Marietta McPherson was the first wife of Chester N. Greenough, a professor of English at Harvard University.
Summary :	Card file of recipes collected by Marietta McPherson Greenough, ca. 1920.
Subjects :	Cooking, American.
	Cooking — Massachusetts — Cambridge.
	College teachers’ spouses — Massachusetts — Cambridge.
Keyword Subject :	Harvard University — Faculty spouses.
	Harvard University — Women.
Form/Genre :	Recipes.
HOLLIS Number :	009359473

Giving all due credit to the “quick and dirty” catalog record for making the item discoverable, I nonetheless roll my eyes at the only context the record provides. “Marietta McPherson was the first wife of Chester N. Greenough, a professor of English at Harvard University.”

Full disclosure: I’m pretty sure I created this record.

In point of fact, Marietta McPherson Greenough was an author, cookbook collector, and an advocate of home economics as a field of study. Under the pseudonym Mary Green, she published Better Meals for Less Money in 1917. The recipes in the card file seem intended to cope with war-time shortages. Some were vegetarian or required only a small amount of meat, and the desserts call for only small amounts of butter or eggs.

This biographical note that limits itself to who her husband was–when the recipe collection is so clearly the product of her own research–arguably constitutes error. (If one of my students had done this, there’d be some fairly copious comments on their assignment!)

When inadequate data is transferred from a handwritten description into an online environment, there is potential for promulgation of error and bias, as well as the welcome possibility of being called out on error and bias.  Such old descriptions are often the source for “quick and dirty” catalog records. The inaccuracy and bias in this case derive from an outdated view of the archives, an outdated view of the world, and an outdated expectation about what researchers want.  Once upon a time “wife of professor X” was the only presumed role of significance, the professor the presumed topic of research, and the professor’s papers their presumed resource of choice.

One of our visitors at the Cambridge Open Archives Tour remarked on how many of our staff were women.  I’d like to think any archivist today–male or female–would appreciate context and recognize research value.

In a world of bad finding aids, there are some that make you cringe. Did my archives do that?

This finding aid stands out because of the obvious hard work that went into it, but it is perhaps the least user-friendly thing I’ve ever seen.

Here’s the handy key from the bottom of page 3:

Now that you know how to interpret it, have a gander at page 21 of the finding aid:

Let’s translate the “1910″ line. Here’s an enlargement:

It reads “1910    2, 1, (1)     1xc, 1relay”

This tells us that there are 6 photographic prints of the “track” team (which in this finding aid means any team that runs).

In the position under the “Varsity” column header, “2,1,(1)” means four photographs are of the Varsity track team.  Of these four, two are presumably from the same negative and definitely from 1910, one is from a different negative and definitely 1910, and one is from a third negative and probably from 1910.

There are no photographs indicated under the column for the Junior Varsity team, but there are two “other” photographs.  One is the cross-country team, and the other is the relay team.

Of course, there’s a lot that this  finding aid is not saying.

There’s no information about the photographs’ sizes, or anything to help you retrieve the right folder.  (These are portfolio-sized folders stored flat in drawers, which is scary, since a lot of the photos are small, but we’ll let preservationists shudder about that.)

Near the top on the right-hand side are the years 1930 and 1931.  Does it take steeping in an Ivy League tradition to understand that “H-Y” means this is a photograph of a Harvard-Yale track meet?

We’ve decoded this finding aid and converted it to EAD, but we didn’t convert some of the information– such as whether a date attribution was doubtful or whether more than one print might have come from the same negative.

And we’re still being slightly mysterious.  There are placeholders in the EAD version waiting for links to several dozen digitized images.

So, is this more user-friendly? http://nrs.harvard.edu/urn-3:HUL.ARCH:hua25013