Thursday, July 31, 2014

The Morpheus Database: Thinking with Computers

"Learn how to program a computer, learn a computer language, because it teaches you how to think"

The Lost Interview
Steve Jobs


The Morpheus Database (Mark 2) represents a second stage in the development of an interactive archive of transformation in literature. At this point, I begin Beta Testing the database by allowing online access to the information in order to assess the viability of the program. 


Progress Report 

Since the announcement of the Morpheus Database for Critical Theory last month, the data-entry has been ongoing. Currently over 120 individual monograph chapters and articles have been entered. This provides [1] general information (titles, author, date), [2] an outline of the entry's argument (thesis, antithesis, quotes), [3] and coding to designate its relevance to other entries.

The basic structure of the archive has been based on the Morpheus Database for Literature, initiated during Mark 1 development and currently ongoing. The general data-collection has been relatively straightforward. 

The outline of the arguments is an innovation specifically geared towards the actancy of a critical theory text as opposed to a narrative that runs on characters, events, etc. This section has been very effective at capturing a snap-shot of a large quantity of information allowing for a map to be made of discourses on change within and between texts. It has however proven to require a lot of pithy summary and transcription of quotes can be time-consuming but critical to later work.

Particular difficulty has arisen in the final phase of coding: inter-connections. 


Thinking Like a Computer

The way a database is coded will reflect and determine the types of questions that it is asked. As a result, a developer must simultaneously thinking like a theorist (what do I want to know from these texts?) and also like a computer (what limits and sensors do I want to program?). 

Once these tagging mechanisms are formed, the database will go to work behind the scenes creating maps while I am working up front inputting the data. This is why it is important to consider the coding early and frequently. Working with a computer is like working with another brain. It will do things you asked it to do but you may forget down the line what your specific instructions were and what the consequences of those commands might be down the road.

In many ways, queer and disability studies prepare a digital humanist for the complicated and multiple ways of thinking required to develop a useful database. I've elsewhere quoted Disability Theorist Robert McRuer saying, "Every way of knowing is also a way of not knowing something else." This holds true for code. Despite what Science Fiction will have you believe, there is and will never be such a thing as a "Master Code" that will give you access to all computers everywhere (Sorry Neo). Likewise, no method of tagging or coding will give you all the possible information from database. Broad architecture can be built in to allow for a variety of options later, but what senses you give the mainframe determine what it sees, hears, or feels. 

It is true that certain methods of coding will work well across many hardware and software projects, but computers, like the people who use them, are all made differently. People will use source code in strange ways. They may introduce anomalies,  "glitches," into the programming. Then, as with everything else, over time things transform.


Theory v Literature

While the literary database tracing the transformations of bodies has functioned on general codes (e.g. types of change, gender, relevance to a variety of fields of study) many of this tags are either not transferable or no longer relevant.

For instance, it is less useful to try to interconnect entries based on its relation to general fields of study (such as thing theory) because the topic of a monograph will tend to lump all the chapters of a text together rather than noting distinctions. 

Furthermore, because the critical theory texts have been selected specifically for their relevance to a project on gender, sexuality and disability, nearly all the text fall into all of these three or else some combination. 

A tagging system that worked well with a general, somewhat arbitrary collection of texts and transformations is no longer useful in a more specialized study. New methods of coding must be developed.


Questioning the Database

Moving forward, the questions will arise primarily from the basic lines of thought that lead to the theory database's creation. The gender and disability focus bring together a medieval historiography on transgender within a wider theoretical framework. This will serve as a general guide for the reformatted tagging.

There is no reason to reinvent the wheel, especially when digital composition is concerned. Although coding-it-yourself is a great motto and (as stated) there are no master-codes, the digital world is highly adaptable. Borrowing and tweeking old structures are a great way to get what you want done while leaving your time and energy on the specifics rather than redeveloping the system itself.

Is this case that means adapting less useful tagging methods to make them better serve the purpose of creating connections between the entries. Previously, I installed text-boxes where a summary of the article's relevance to other research interests can be written. This data-entry form was perfect for thesis writing where each entry had a unique aspect. These connections however became largely formulaic.

While most entries had some variation in their interconnections, most drew from a general vocabulary of key terms (e.g. transvestite, prosthetics, transsexual, operation, transgender, madness, etc). A redesign could look to the terms that emerged and change the text-boxes into tabs (with yes/no options). This would allow later inquires to be ran where all texts that connected with the "eunuch" node to be selected, or all critical theory that may be applicable to a reading of "the Book of Margery Kempe."

By choosing to adapt rather than create a whole new method of tagging (at this juncture), I will save time and energy. Revisions can be made to each page in relatively short order because I would not be re-reading and re-entering data but rather looking at the information in the text-box and selecting the appropriate yes/no tabs that correspond. With these tabs in place, future coding will easily slide among the other entries and later this summer larger trends can be mapped.

That's all for now, let me know what you think!

Explore the Theory Database!

Sign up to be an authorized Beta-Tester! Add to the conversation about transformation and critical theory!


1 comment:

  1. What is jewelry to you? Do you have knowledge about the different types of jewelry and how you can determine high-quality jewelry? The ideas presented here will aid in make you a more informed jewelry purchaser. | | | | |