Hermione,
Thanks for the info on this. Steve Farmer provides a good critique and the posts provide more info.

As for my opinion on "Computational Linguistics", and finding "order in symbols"...

Quote

The joke is that there is "order" and "meaning" of some sort in
virtually all symbol strings. Finding that has absolutely nothing to
do with whether the symbols are linguistic or not.

If you had a wealth of texts, you could look for common words and patterns to create a dictionary. This seems to be what they've done. A nice first step.

Of course, with an Egyptian Analogy...
if I give you a word in Gardiner Symbols: F31 N29 X1 O1
Determining what this means, the context, the language rules, why this order, etc... that's not so easy... and it doesn't seem like they have gotten this far.

Nevertheless, it is a good idea to have someone document all of the patterns between all of the texts... it is a nice first step.
*****
Looking at the local languages may or may not be of value, but it seems like the only reasonable place to look, if you are trying to match up symbols to a language.

Quote

the authors compare the Indus signs with only four
languages: English, Sumerian, Sanskrit and Old Tamil. Any claim for supposed similarity
between Indus inscriptions and any Dravidian language would need to be based on comparisons
with far more languages than this.

*****
I don't really have a problem with the methodology being employed. Turning these "symbol patterns" into useable results though may be an impossible task though, do to scarcity of texts. Perhaps the "symbol patterns" will inspire a future research in 10 or 20 years. Who knows.