November 30, 1999
Ms. Derrick should be congratulated on this remarkable project. She has managed to design and flesh out a database that encompasses many of the materials encountered by the conservator. Its simplicity helps insure that it is easy to use, responsive, and informative. In this regard, it is quite successful, even at the present stage of its development, compared to other efforts with which I am familiar. As it is conceived, it will eventually fill a specific need in the field of conservation, which has become so complex, specialized, and filled with highly technical terms and materials that it is often daunting for the individual practitioner to know where to start in a search for relevant information. I applaud the decision to build the database on a widely used and famously user-friendly product, FileMaker, which, while simple and straightforward, is effective and responsive. (I have constructed databases with both FileMaker and MSAccess, and continue to find that the former is a very efficient program both for the designer and the user, especially after it expanded from flat file to fully relational capability.) In general, Ms. Derrick has launched a terrific idea which, with continued work and input, will undoubtedly become a favorite “Swiss army knife” tool for conservators everywhere.
The concept of the database is good — to create an easily accessible source of information that will help the conservator find basic explanations and definitions about a wide variety of materials which may be known under several names, some traditional, others technical. Having spent some time “challenging” the database to yield information on a number of subjects, I remain impressed by the sheer quantity of knowledge that Ms. Derrick has managed to include. Some who have commented on the database suggest that it is unfortunate that it doesn’t include more on procedures and techniques of conservation. I disagree somewhat with this assessment, but in trying to articulate exactly why I disagree I think I came to a better understanding of what such a database should be.
To some extent, the term “database” may even breed some difficulty. Fundamentally, this product is not so much a database as it is a “finding aid” or “dictionary” that is constructed on the framework of a database. It makes good use of the ability of a database to search quickly through a field containing certain kinds of information and thereby enhances its usefulness, unlike, say, a traditional index in a book. In my daily work, the kind of information I am most likely to need to obtain — and the kind that is often most difficult to unearth — has to do with the definition of some material, its composition, and how to find out more about it. While techniques and procedures are, of course, of enormous importance to the conservator, I am skeptical that such knowledge is best presented in a format of this kind. Rather, a catalogue of the kind already produced by several specialty groups of the AIC is more congenial as a means of conveying procedures.
It also occurred to me that this database — I’ll dub it the “Conservation Materials Dictionary,” or CMD for short — is likely to be used most by someone who is trying to locate a basic definition of some material that is not ordinarily encountered in his or her own specialty, and who hopes to find signposts toward more information on that material. As I was using the CMD I had to keep reminding myself that it cannot be a substitute for the twenty shelves of reference works in my lab. After all, one is likely to know far more about one’s own area of specialization than would be possible (or desirable) to include in any database format, or at the least is likely to know where to go to find the information. For this reason, the part of the database I found most fascinating was generally in areas outside my own. When I opted to browse through classifications such as fiber/textile/leather, architecture, or environment, it was hard to stop reading entry after entry because I kept finding myself saying, “How about that! I didn’t know that!” I didn’t really expect to have fun reviewing this! On the other hand, within or tangent to my own sphere I also found that I enjoyed occasional notes about the date when something was first introduced, whether it was masking tape or cellophane, and that this was a pleasant “bonus” to be had in strolling through largely familiar topics. The guiding principle in further development of the CMD should be that it will continue to be an “enhanced dictionary”: a source of good definitions augmented with references to sources for further information.
Regarding the current design and functions of the CMD, I have a number of observations and suggestions for future consideration.
- The program installed easily and was invoked on my PC (Pentium 166 MHz processor, Windows 95, 32 MB RAM, with a 15″ monitor) without difficulty. Once the program was operating, however, it prevented access to the desktop, preventing one from alternating between other tasks such as word processing. This is at the very least an inconvenience in that it makes it hard to refer to the CMD easily while engaged in other computer tasks.
- There are fairly serious problems with the display of the various views in the program. In all layouts it was constantly necessary to scroll up and down and left and right in order to read the entire page. I don’t know what kind of system was used in the initial design of the database, but I suppose that the display was set for a higher level of resolution than my PC. If so, it would be helpful to the user to say something about how the resolution of the monitor should be set.
- An introductory page or layout with instructions about how to use the CMD would be helpful. A button (and associated FileMaker script) could be made to take the user to the “Search” layout.
- There are problems with printing. When I first printed out a record, much of the right side of the information was missing. This suggested to me that I needed to bring up my printer settings and change the format to landscape, but it is possible that another user might need some “coaching” to select the correct setting. Having selected landscape format, however, the result was still unsatisfactory (see attached printouts). Some of the text at the right was still truncated, and a second page was printed out (it was empty of text, except for the header and the image of the buttons across the bottom of the page) even though the database indicated that I had only found one record.
- There are a few quirks that suggest that some procedural instructions would be useful. I tried a search for a word that turned out not to exist in the Name field, and this returned a message, “No records match this request.” If one then hits “Cancel”, the Search page is somehow disabled so that clicking on any button produces a message, “The field is not modifiable.” The solution in this case is not to use the “Cancel” button in the first place, but to select the “Modify Find” button instead. It also turns out not to be possible to check more than one box under “Classification.” In this regard, I think it would be a useful feature to incorporate the capability of Boolean searches, i.e., to be able to use “or,” “and,” or other similar operators. There is a de facto “or” function in the “Description” field, as it appears that the database will search on any words typed into the entry slot. I could imagine that it could be useful to do a compound search that might include the word “adhesive” in the “Name” field, and a check mark in, say, the “Paintings” field, thereby limiting the search to only those records that would relate to adhesives as they pertain to paintings.
- As a default, the database will search for any words that contain the word entered in the search field. For example, entering “alum” will return results such as alumina, aluminum foil, and aluminum. This can be especially helpful if one only has a vague idea that the word one wants to identify might contain certain letters. On the other hand, it would sometimes be helpful to limit the search only to a particular word. In this example, if one only wanted to get a definition of the word “alum,” obtaining a larger group of words each of which contains “alum” is not helpful. It is common practice in databases to be able to use quotation marks to indicate that the search should be conducted only for an exact match of the specified word, but this is not possible in the current version of the database. In another example, typing in merely the letters “tess” leads to “tessera.” But if one types in “tesserae” (the plural form of “tessera”) the database finds no records. Perhaps in a set of instructions the user could be coached to try various portions of words if an initial search is unsuccessful.
- An editor needs to attend to certain problems of consistency. A search for “amate” has nothing entered under the heading called “Other names,” but in the text directly above it says that one should also see “amatl.” If so, it seems to me that it should be noted in “Other names.” On the other hand, in the entry for “amati,” in the “other names” heading it mentions “amate” and “Aztec paper.” Selecting the phrase “adhesive, pressure sensitive” (found among the list for the “Adhesives” classification) produces “see ‘pressure-sensitive adhesive’.” A search on the latter returns the user to “adhesive, pressure-sensitive.” To break out of this circular situation, selecting the phrase without use of the hyphen, i.e., “pressure sensitive adhesive” produces what is probably destined to be a more fruitful result, although at the present time the “full record” is blank and the “working record” contains quite a bit of interesting information.
- In the “Full record” layout, a portion of one of the adjacent records is also visible. This is rather confusing and can probably be corrected without difficulty (it’s as if the “list view” format has been designated while doing the design in FileMaker).
- In the “Full Record” and “Working Record” views, the ruler across the top is unnecessary and takes up valuable screen space. There are buttons across the top that are grayed out and hence non-functional. These should also be eliminated. I would also suggest that the vertical panel along the left side be permanently concealed from view. This is a useful “trick” in FileMaker that will free up still more screen space. If certain functions (quick access to the other layouts, for example) are deemed desirable, it uses up less screen space to design small buttons that activate appropriate scripts.
- For general searches carried out under one of the main categories, it might be helpful to have a note somewhere that would remind one of which of the categories in which one is browsing.
- I suggest that the word “browse” that is seen while in the “Full Record” and “Working Record” views is not the best choice. Perhaps the button could be labeled “Return to List View,” “View as List,” “View Results as List,” or something comparable might better describe the action.
- I find that it is an annoyance that a complete sorting of the entire database occurs whenever one exits the program. Instead, perhaps it would be better to have this occur only when one selected the “Browse” button on the home page, indicating that one would like to browse the entire database.
As an ardent fan both of databases and arcane literature on conservation materials, it has been a pleasure and privilege to try out the “beta” version. I am impressed that Ms. Derrick has managed to do so much while working essentially on her own. As she has said, more needs to be done, but this raises the question of how best to proceed with building on this excellent beginning. I tend to agree with those who have suggested that it might be good to consider enlarging the team of contributors in order to enhance the depth and usefulness of the CMD. With additional funding, a group of contributors could be convened regularly and, with the assistance of a coordinating editor, could push this project to the next level of excellence. It is tempting to think ábout whether funding for this effort could be forthcoming from the American Institute for Conservation (which offers some support for publications, and it could certainly be argued that the CMD is a “publication”), or possibly from the Getty Trust, especially if it seemed appropriate to produce a more international version in a few different languages.
Regarding how best to disseminate the CMD, it is especially tempting to consider “publication” via the Internet. Already there are countless databases to be found on the Internet, and the latest version of FileMaker is designed to be “web-ready.” While I remain, of course, very fond of the book format, I am persuaded that the flexibility, ease of updating, and world-wide reach of the web, which is already becoming a “first stop” in the search for information, make it admirable suited for this purpose. Furthermore, it has the capability of hyperlinks to a virtually unlimited number of other related sources that could be added by the editorial board: links to other publications, sites relating to conservation, technical data from manufacturers, etc. The Conservation Materials Dictionary in its current form represents an exciting beginning with tremendous potential to become an indispensable tool for conservators throughout the world.