Algorithms are Thoughts, Chainsaws are Tools

Thought I’d share this brilliant clip by Stephen Ramsay for the recent Critical Code Studies conference.  The result has been another case of what I call the Rationalization Effect — technology disrupts practice and forces us to explicitly think about, and perhaps define, tacit ideas. In this case, the idea is the notion of “live performance,” which Amanda French immediately remarked upon, and which gets taken up by Sorensen himself on the CCS blog.

Algorithms are Thoughts, Chainsaws are Tools

To Do 2009-03-18

  1. Write up lesson plan for MDST 3703 (Introduction to the Digital Liberal Arts) studio session on SQL.
  2. Prepare presentation about SHANTI for ACCS conference tomorrow.
  3. Watch demo of Kaltura-Sakai integration tool.
  4. Prepare GraphViz visualization for client (’cause your a nice guy).

Back from Dentist, onto Lesson Planning

I am crafting today’s lesson for the studio track of my class, Introduction to the Digital Liberal Arts.   Students have already been introduced to basic PHP, regular expressions, XML, and image formats.   Now they move on to a key layer in the application stack, the model layer, which provides the most powerful means of organizing and transforming content.  They will be learning basic SQL; next week, they will learn how to communicate with the database from within PHP.

To my way of thinking (I am a data determinist), this is a pivotal part of the course, as it introduces the student to data modelling and knowledge representation, areas that Unsworth (and others) long ago recognized as a central dimension of humanities computing.  Not only does this layer focus on modelling, a rewarding practice in its own right, it is also the part of the application that will enable the student to perform the most magic at the interface end.   For a flexible interface design space requires a flexible data model.

I have long favored SQL as the data structure tool of choice for this kind of work, over “thick XML” and other contenders (such as the current spate of NoSQL choices), because of the stability of the model, the language, and existing applications, and the vast support for the technology in languages.  As far as I know, one only finds genuine database management systems in SQL;  XML databases and now RDF repositories seem never to be complete solutions.   And their query languages just seem harder.   SQL is not good for everything, nor for every data structuring dimension of applications where it is suitable, but it is hard to image a web application that can’t use SQL somewhere.