[B-Greek] GRONT score website
mike at vnoel.com
Tue Dec 28 15:10:55 EST 2010
On Mon, Dec 27, 2010 at 8:25 AM, George F Somsel <gfsomsel at yahoo.com> wrote:
> Precisely. If a few are submitting all of the ratings, the ratings become
> little more than the opinion of those few which may be rather skewed to their
> knowledge of Greek. For the score to achieve any true validity as a comparison
> to NT Greek some standard would need to be established comparing vocabulary
> frequency and also syntactical usage. Otherwise all you have is a statement
> that **I** think it is comparable to NT Greek (or not). What that person may be
> saying is that he is / is not able to read it with some ease which may in turn
> reflect the fact that he may have read Homer, Plato, Aristotle, Herodotus,
> Thucydides (or not have read them).
The thing we are looking at here is a perceptual issue. That is, how
close to the NT/good/easy does the Greek of a particular work SEEM to
be? The benefit is for the reader who is looking for something that
SEEMS close to the NT or easy to read or good Greek.
With access to the texts it would be fairly easy to do some
statistical analysis and comparisons. In additional to the ones
mentioned above, I would also want to look at bigram frequencies. It
would be very interesting (to me) to see if there was a correlation
between the results of the statistical analysis and the perceptual
judgments. Back when I was doing computer speech recognition research
this question came up a lot. We frequently compared statistical
models of recorded speech with perceptual data. Often the results did
not correlate well at all -- which meant that our statistical models
were not sufficient for what we were trying to do but that's a
In the end, I hope that the GRONT scoring is valuable simply because
it does provide a view of peoples' opinions about various Greek works.
If it helps other Greek readers to find something they can read and
enjoy and learn from, then I consider it a success.
More information about the B-Greek