Hi Lennart, and friends,
The unzipped file for reversed sorted list is 35MB. This home pc is grinding under memory strain.
However, managed to extract 2 lists with endings in 'a' (6.4MB) and 'aa' (3.8MB) through small bites of copy and paste.
Looking through the smaller a and aa lists, your reverse sorting may provide the methodolgy for meeting the requirements of Ven. Ashin Pandita and Sis. Bhadra. (Please see highlighted note below)
But its demands for resources would preclude it from usage on ordinary pc's. For instance, words with this suffix, -va.n.n.ana, amount to 7,138. However it might be an alternative to the dictionary housed at Chicageo U. Demand for resources and time for such a project is huge.
Perhaps we may be able to start with a scaled down and much smaller list mirrored against, say, Buddhadatta's dictionary that might be useful here and now for the Ven. Ashin Pandita's Pali students. And as we learn, ideas coud be self generating. A reverse sorted list would be broken down into folders under words with endings in a, aa, i, ii, etc. Then each folder would have subfolder for nouns, verbs, etc. Then students could have at least 2 tools to tackle this 'tejaa' assignment - with 1. reverse sorted list under nouns with 'a' or 'aa' endings and 2. from ordinary dictionary under alphabet 't'. Information from these 2 sources could then be used to map against declension tables.
Please kindly let me know.
From Ven. Ashin Pandita:
However, What Bhadra and myself have envisioned is not so ambitious. It
is a text parsing tool to help students to learn Pali in the Burmese
way. In Burmese tradition, a Pali student should have the ability to
analyze any noun they encounter into the respective noun stem and
case-ending. Accordingly, I have to force students to learn declension
tables by heart, which is usually a miserable job for the teacher and
The tool we have thought of is a workaround for this difficulty.Its
concept can be explained using an example:
Suppose input to the program is "tejaa". It is to be checked against the
declension tables kept in the database (say, those given in the New Pali
First, it would be checked against the declension of purisa (mas. noun
ending in a). And the algorithm would be roughly like this:
"tejaa" is similar to "purisaa" (since both end in aa preceded by a
According to the declension table, "purisaa" has the stem "purisa" and
four possible cases --- nom. plu., voc. sing. /plu., and abl. sing.
Therefore "tejaa" has a possible stem "teja" of mas. gender and four
possible cases --- nom. plu., voc. sing. /plu., and abl. sing. (This
line should be the output)
Then "tejaa" is checked against the declension of citta (neu. ending in
a). The pattern of reasoning should be the same, and the output would be:
. . . "tejaa" has a possible stem "teja" of neu. gender four possible
cases --- nom. plu., voc. sing. /plu., and abl. sing.
In this way, the word "tejaa" would be checked against declension tables
one by one, and there would be many lines of output. Afterwards, the
student must use a dictionary and the context of the word to filter out
the irrelevant and get the noun stem and case that the author really meant.
At the next stage, we can add a dictionary or word-list to help the
student pick the correct stem quickly. In the example above, "teja" in
mas. gender does not exist in actual usage; the software can point out
this fact to the user after referring to the word-list.
The principle would be the same for conjugated verbs, only more
complicated. The software is to help the student analyze a conjugated
verb into its hypothetical components, and pick the correct ones.
Lennart Lopin <lenni_lop@...
> wrote: Dear friends,
Creating a reverse sorted list was easier than i imagined. You are invited to download the reverse sorted Pali word list (sorted according to the Pali alphabet) from this URL:
Any suggestions on what to extract/analyze furthermore are welcome...
What are the most popular cars? Find out at Yahoo! Autos
[Non-text portions of this message have been removed]