In my last post I expressed regret that it looks like the CAD will not be moving to a full electronic version. I say this because Martha Roth basically said at SBL that once the U/W volume is published, they will be packing it up. Her intention is that the CAD will provide the general framework for the field, but new dictionary projects will concentrate on more specialized areas like astronomical texts, etc. Further, at a certain point they stopped collecting text references so that there will be no revision of the earlier volumes.
Let me say first that the CAD is a wonderful reference tool, and I have nothing but respect for the work done over at the Oriental Institute. Also, the initiative to get this information into the digital domain for free is generous, bold, and forward thinking.
As a series of searchable PDFs, the CAD is basically a digitized book. You can read it on your computer and search within a single volume. However, over the last day, I have been thinking about what a true electronic version of the CAD would look like. Two models come to mind first: the Comprehensive Aramaic Lexicon (CAL) and the Pennsylvania Sumerian Dictionary (ePSD).
I am familiar with the innards of the CAL since it is hosted here at HUC. It is more of an electronic text repository/concordance than a dictionary at this point. The first goal was to enter a comprehensive corpus of Aramaic texts covering all dialects. These texts are tagged with information related to the lexicon, but not morpho-syntax. That is, the root and binyan of a verb are important, but not its person/number/gender/etc. Each root has a separate lexical file, which lists English glosses for each dialect. The lexical information was originally input from existing dictionaries to give a baseline, but has been updated when necessary based on the broader data from the text corpus.
As a user you can enter a root and either view its static lexical file, or you can create a dynamic concordance limited by dialect(s). The benefit of this is that it allows you to make your own lexical decisions based on the data from the concordance. The negative is that the lexical file does not include any text citations to indicate why the lexicographer made their decisions. Since the backbone of the lexicon is a text repository, you can also use the site for dictionary help in reading through an individual text.
I am not as familiar with the ePSD, but I do use it regularly for Akkadian class. The ePSD is not a concordance and does not contain searchable texts. Each Sumerian entry has a lexical file that includes a short English gloss and a list of Akkadian synonyms. Because Sumerian uses logograms, the signs are an important part of the lexicon. Each word contains a list of attested spellings along with a table of their historic distribution. Following this is a section of short text citations giving attestations of each gloss. If you are interested in concordance work, each entry has a link to the ETCSL where you can find texts.
So, it would be nice to see something similar to the CAL/ePSD for the CAD. That is, the CAD could be used to create static lexical pages for each entry (note that the CAD is not organized by root) complete with the text citations that the lexicographer based their decisions on. All Akkadian references in the text citations should be cross-referenced to their own entry, and vice versa. Further, it would be nice if each entry could also be linked to an electronic text repository in order to do the concordance work for yourself. There are already several projects underway such as CDLI to create electronic texts. If they are tagged with lexical information similar to the CAL, then the electronic CAD could be the glue that pulls them all together.
Those are my quick thoughts, and yes it would be a lot of work. Have I missed something major or been shortsighted? How would you like to use the CAD electronically?