This post is the second in the series describing and interpreting a set of questions that I have used when exploring databases containing Shakespearean texts. The last post presented and explained the first four questions. This time I shall cover questions 5-8 exploring documentation, purpose, fitness for purpose, and the quality of the text. Let’s start then with question 5.
- Is the documentation of the database clear and succinct?
Documentation of responsibilities and sustainability (even if theoretical) purposes and origins may well be a characteristic feature of a database. These should be explored in detail and even the problematic points may be clarified. It is also relevant here whether this documentation can be found easily or is buried at a location that can hardly be found. Naturally it is open to debate what is meant by “detailed” documentation, especially because I can hear the counterargument that a good project does not need documentation as the database and its usability speak for themselves, it is, thus, not needed to document this. There is much truth in this claim, since who needs the documentation for something that works properly and with considerable success, who reads the documentation of Twitter and Facebook. I reckon the case is different with applications and databases that if only partially but still would like to attract scholarly audiences as well. For the scholarly community to be able to take the results of a project and research or query seriously must be able to look behind what is immediately visible. What is going on behind the scenes is as important for a serious user as the results of a query provided by the database. So a clear and succinct documentation is indispensible for an intended audience that would like to be taken seriously.
- Is there a clear statement about the purpose of the database?
This question speaks for itself. It is reassuring to know what the database was built for. It does good both to the creator, because to have a clear purpose help one stay focused. Also this does good to the user, because then (s)he knows what to expect, be confident to use the database for what is was created for.
- Quality of the texts.
The quality of the text is one of the cornerstones of a database. Even if this is only one of the four aspects of a database, this is the first aspect that a Shakespeare scholar will enquire about, and if it does not live up to scholarly standards the database will not be used.
- Is the origin of the digital text documented?
This has two advantages. If the textual and editorial choice have been explained, it is very likely that the creator of the database has given thought to the choices made. In this case it is very unlikely that it can happen that someone created a powerful, fancy and interesting tool, and then feeds into it an unedited text found somewhere in the public domain without say checking that the King Lear under consideration was written by Nahum Tate. If a text is documented appropriately it is very unlikely that such a mistake is made. The second advantage of documentation is that the Shakespeare scholar does not have to spend or waste time with discovering slowly that the text is unreliable and useless for a scholarly purpose.
- Is there somebody responsible for the digital text?
The documentation should not only reveal the origin of the digital surrogate but also should name the creator of the text. Even in cases if the text was not created by the creator of the database but (s)he uses someone else’s text. This is important because even if a text is left without editing, when preparing the machine readable text there must be decisions made, and it is indispensible that somebody takes responsibility for these decisions. This is part of scholarly honesty.
- Are the editorial decisions explained and documented?
Of course, the expectation is not to explain every single editorial decision, because that would mean the creation of a documentation similar to a critical edition. The expectation, however, lies in the exploration of general editorial decisions with a few examples for the sake of clarification. Decisions are the ones that pertain both to the text and to the encoding of the text.
- Is there a harmony between the purpose of the database and the search engine, quality of the texts, level of encoding?
It can happen that the purpose of the database and the search engine, quality of the text and the depth of encoding have not been harmonized. It can happen that a database intends to serve scholarly purposes for which a powerful search engine has been installed, which should secure the scholarly outcomes of the queries. The search engine, however, cannot secure scholarly purposes in itself, if not accompanied with an appropriate text. The excellence of the engine cannot compensate for the weakness of the text. Unfortunately in this case there is no real compensation, the weakest part determines the power of the database. It can also happen that the search engine has not been tuned for the depth of encoding. It can happen that the level of encoding does not harmonize with the power of the search engine, or it can also happen that the text is encoded in more depth than what the search engine has been tuned for.
This time I focused on the aspects of documentation that should ensure the quality or at least the transparency of the database. These qualities may well attract or distract a Shakespeare scholar to or from the database. Next time I shall continue the list and explanation of the questions that help analysing a database focusing on Shakespearean texts.