3 ms·

This reminds me of a project I started as a Religious Studies undergrad to try to use Python for scriptural analysis by analyzing word frequency. I abandoned it

This reminds me of a project I started as a Religious Studies undergrad to try to use Python for scriptural analysis by analyzing word frequency. I abandoned it, however, after I tried running my program on several different translations (i.e. KJV, NRSV, etc.) and would get wildly different results. I realized that I couldn't rely on translations for accuracy and, being an undergrad, had no knowledge of Greek or Hebrew to be able to rewrite the program for the respective original languages. It bummed me that I abandoned it, largely because I think there's a void for CS-aided Humanities endeavors. I still consider myself a (mediocre) humanities academic, rather than a (horrible) CS/IT worker. I'm happy to see this article perhaps giving some hope to other majors that you don't need to go all-in on Comp-Sci to be able to use it.

ebiester 11y ago
It goes deeper: the greek texts are divergent as well. The work you proposed, believe it or not, has been done by hand (though I'm sure it has been replaced by computers.) Concordances have been around since 1262, and the first (according to wikipedia) was an effort of 500 monks.
- singingfish 11y ago
  What's the formula for converting monks to cpu cycles?
  - TheOtherHobbes 11y ago
    It depends on the denomination of the CPU.
onetwotree 11y ago
I did something similar while working for the IT department at the University of Chicago -- we helped a humanities prof build what was essentially a plagiarism detector, with the twist that it was designed to be applied to "great texts" in order to examine plagiarism, which wasn't considered a bad thing back in the day. Spoilers: Rousseau copied like 25% of most of his texts from earlier authors. At the same job, I also helped some linguistics grad students use python to do something involving linguistics that I didn't completely understand. They were pleased as punch to be able to automate a tedious manual process, and expand their throughput to entire texts instead of a few sentences. So there is a lot of potential for using code to help the humanities, but it takes active involvement from hackers who are willing to work with humanities folks. Props to NSIT at the U of C for attempting this.

ebiester 11y ago

It goes deeper: the greek texts are divergent as well. The work you proposed, believe it or not, has been done by hand (though I'm sure it has been replaced by computers.) Concordances have been around since 1262, and the first (according to wikipedia) was an effort of 500 monks.

singingfish 11y ago
What's the formula for converting monks to cpu cycles?
- TheOtherHobbes 11y ago
  It depends on the denomination of the CPU.

onetwotree 11y ago

I did something similar while working for the IT department at the University of Chicago -- we helped a humanities prof build what was essentially a plagiarism detector, with the twist that it was designed to be applied to "great texts" in order to examine plagiarism, which wasn't considered a bad thing back in the day. Spoilers: Rousseau copied like 25% of most of his texts from earlier authors. At the same job, I also helped some linguistics grad students use python to do something involving linguistics that I didn't completely understand. They were pleased as punch to be able to automate a tedious manual process, and expand their throughput to entire texts instead of a few sentences. So there is a lot of potential for using code to help the humanities, but it takes active involvement from hackers who are willing to work with humanities folks. Props to NSIT at the U of C for attempting this.