Mapping User Interface to Web Services – Observations from A2

Here are some observations after completing A2:

1.  Sorting and formatting results from GET and POST calls to the collection space web service is very necessary and important to understanding what you’re working with.  It is hard to read the results, and it takes time to find what you’re looking for in the document.

2.  There are a lot of fields in the user interface.  It takes some decision making (subjective) and ideally shared conventions to decide what things should go in which fields.  This could be hard to standardize between institutions.  And, it would seem that the number of collectively agreed upon fields (i.e. standards) would be rather small, although that could still be very useful (ie. logistics, loans, and insurance).

3.  In the rare case that you’re someone who has to work both on the web services and user interface side, or work with people doing this work, it would be useful to have a mapping between the organization of the UI and web services (XML, JSON) results.  In the ideal case there could be a anchor link between the respective locations in the two documents.

Media handling in CollectionSpace

Looking at the “Recently Created Records” page on the CollectionSpace demo site, there seem to be two main categories of records: cataloging and procedural (which includes loan in/out, acquisition, location and movement, exit, intake, group and media handling).

Since I’m interested in how systems like CollectionSpace deal with different media types, as I poked around the existing records in the demo, I tried to explore the “media handling” tab for some of the objects and explore the “related media handling records” (I didn’t seem to have permission to edit the media handling record). I like the linkage of data here — I’m curious, is every object in CollectionSpace considered media? So I can link any object to any other object using the “media handling” fields?

For my new record, I chose to add a digital photograph object through the “media handling” heading. I was excited to see how easy it was to either upload media or provide a URI to media stored elsewhere on the web. As I added metadata to my object, I was simultaneously impressed and overwhelmed by the number of fields from which I could choose. For example, since I was uploading a digital photograph with no physical surrogate, I didn’t need to use all of the “Dimensions” fields. But having the option is important — and I wonder if there is a way to collapse some of these extra fields when viewing a record, for simplicity’s sake. I’d also like to better understand the “Person” authority as CollectionSpace intends it. As far as I can tell at this point, “Person” is used to identify the original owner of certain objects — but the “Person” entity could be used in a number of other contexts, too. I wouldn’t be surprised if CollectionSpace allows for this and I just missed it in the records I browsed.

My First CollectionSpace Record

I started out trying to make a record for a photograph, and, my, aren’t there a lot of ways to describe a photograph. For a collection that already has some sort of organizing system, manually entering that much detailed info for each item in a collection is prohibitive. I imagine that there’s some way to batch upload preexisting catalog information from a CSV file, although that wasn’t readily apparent to me.

Next, I experimented with Loan In. Clicking around in the right side bar led me to a location record called “Creaky old shelf at the bottom of the stairs.” The Cataloging section of the record tells me that the creaky old shelf holds a painting and a camera. This two-way relationship (“x is on the shelf”/”The shelf contains x”) is really handy. It’s easy enough to find the location of a single item, but it’s rarer to be able to access all the items in a location. It struck me that the descriptions of the two items in the location vary in specificity. The painting is described just as “painting,” perhaps because no title is listed. The descritption for the camera, however, is “Bell & Howell 35mm Eyemo Motion Picture Camera.”  This description is found under title, which seems like an oblique fit to me, as cameras generally have a make and model rather than a title.

CollectionSpace Observations

I examined a couple of existing items, such as the Lemon. The other tabs (Loan In, Loan Out, Acquisition, Location and Movement, Object Exi, Intake, Group, and Media Handling) are empty, which was somewhat frustrating because you had to be on that page in order to discover this. (Still don’t know what the Lemon was – the Brief Description said “Accessories”, and Responsible Department was Decorative Arts.)

Thought it was interesting how the green tab with the plus sign was perched on top of several categories as a way to add an additional field for that particular category. Its function was instantly recognizable and much more aesthetic than, say, an “add” button or link. Found it interesting that there were so many fields for collecting information about artifact history – I didn’t see that many fields used.

Created a new cataloging records from Template, selected Photograph from Template. Entered information for a standardized image of a player piano. There were some fields that I ended up not using. I went to the next tab, and edited some of the Cataloging information. Some of the fields were hard to distinguish (such as, what is the difference between a Technique and a Technique Type? What does the Role of a Production place mean?) In attempting to add a Production Organization, I found that it automatically searches existing organizations to match (the pseudonym organization I used wasn’t found, and I was given the option to set Default Organization Authority). Retrieving this cataloging information was difficult – it didn’t show up in the Cataloguing on the right. I tried searching for it in Related Cataloging Records, and it appeared there, but grayed out. I’m still not quite sure how to access the Cataloging Record that I tried to add here.

Observations on CollectionSpace

I played around with CollectionSpace a couple of ways. The interface is clean and straight-forward. Not knowing what to put into the input fields made it a little confusing but that is just a matter of knowing the jargon and having data to work off of. In creating an “Acquisition” entry I was not sure whether or not to add terms that weren’t found in the existing vocabulary that would appear in the pull-down so I just left those input fields blank. Also it moves rather slowly between pages.

I also created a “Photographic” catalog entry (“Rabbit Man”) which has a lot more input fields than the “Acquisition” entry. I did not fill out all the input fields because I lacked the information to do so but the form itself was thorough; there was input considered for the contents of the object itself, its physical characteristics and its provenance, as well as extended boxes for interpretation (though I wonder if that is too subjective.)

Even though there is a lot of information that can be put into each new entry, overall the look of the interface is clean and simple enough to not be overwhelming.

Blog access info

The students who are in the I School should be added as contributors in the next day or so. For those of you in other schools/departments, please email me and I will get you an account (we need your email address to set it up). Alternatively, you can self-register (that may take a day or two, as we moderate all registrations to combat sp*m).

Thanks – Patrick

Welcome everyone!

Welcome to the course blog for IS290-rmm, Resource and metadata management in Museums, Archives and Research Collections. We’ll be using this blog to comment on readings, topics and issues raised in class, and on the tools we’ll be using.

We’ll also use a course mailing list, i290-rmm-sp12@lists.berkeley.edu, to distribute general info and keep you up to date. Please subscribe to this email list if you have not already.

The course syllabus and related info has been posted, although some of the readings for later lectures will be updated in coming weeks.

This should be a fun class, combining an interesting domain with a chance to put your I School chops to work. I’m looking forward to it!