Sunday, August 31, 2003

A reader posted a comment to the effect that my post on Jonathan Edward's Resolutions for Life was invalid RSS because of an endash character in "peac_able" (inadvertently copied, and now removed). I'm not a 3-star RSS maven, but i wonder if this is true, and if so, why? If i look at the XML feed itself, clearly there's a character outside the 7-bit ASCII range. But the RSS feed produced by Radio Userland doesn't include an encoding attribute (maybe it should?), and if i drop the whole thing into XML Spy, my XML tool of choice and usually quite fastidious, it has no complaints. Do some RSS readers adhere to stricter standards than what Userland produces?

If somebody can shed some additional light on this, please drop me a line or post a comment to explain.

9:17:07 PM #  Click here to send an email to the editor of this weblog.  comment []  trackback []

An article by MSNBC describes how

"the behaviors that characterize teens’ use of technology—socializing, multitasking, going mobile, making a personal statement—are increasingly influencing what is being developed in industries ranging from telecoms to electronics to software"

So Microsoft has hired anthropologists (!) to try to convert these "disruptive technology" trends into new products. But here's that part that seems significant to me: "... kids are having perhaps the most important and far-reaching impact in the area of collaborative computing..." I wonder how these trends will be manifested in the religious lives of future teens? Virtual Bible studies and prayer groups?

8:48:31 PM #  Click here to send an email to the editor of this weblog.  comment []  trackback []

A picture named louw-nida.jpgI've recently discovered and acquired a wonderful tool for understanding the semantic content of Scripture, Louw and Nida's Greek-English Lexicon Based on Semantic Domains. What they've done differently from other lexicons is to clearly distinguish and group different meanings for the same term. 

In addition, they've indexed them according to an semantic classification scheme (what we'd call an ontology today). There are 93 semantic domains, each with subdomains, and terms are classified into them. For example, domain 28 is "Know", and includes these subdomains:

  • Know
  • Known (the content of knowledge)
  • Well Known, Clearly Shown, Revealed
  • Able To Be Known
  • Not Able To Be Known, Secret

The first subdomain (Know) has 16 different entries that express this concept, and it includes words from different parts of speech (verbs, nouns, adjectives) as well as different usages of the same term (for example, use of the Greek middle form has semantic consequences for many verbs). So it allows you to find the words whose meanings are related to one of interest.

Attempts to classify and organize terms and knowledge have a long history: Princeton's WordNet project has achieved some real success and practical value, as opposed to Japan's Electronic Dictionary Research project in the 90s which seems to have been a colossal and expensive failure in the grand tradition of AI Done Large. As i've pointed out before, however, the New Testament has a tremendous advantage as a technical field for such studies: while it's broad enough to provide a real challenge, it's a fixed corpus whose entire vocabulary size is something over 5000 terms.

Nida is one of the pioneers in the application of linguistics and lexicography to Bible translation, as attested by this 50-year history of publications. I would love to expand the NT Hyper-Concordance to include this kind of information as a navigational feature.

By the way, if you purchase this, make sure you get both volumes. The second volume contains all the indexes for the first, and one isn't useful without the other.

10:14:50 AM #  Click here to send an email to the editor of this weblog.  comment []  trackback []