compound analysis in e-texts

[Jakub Cejka]

Lars, I forgot to say one more think, that although I find it very 
important for any sort of further work with a computerized text to be 
analysed, I always wondered how could you rely on the typers, as to 
which alternative should be selected if a compound or a sandhi can be 
dissolved in two ways (such as whether it is alfa privative or a prefix 
-a that merged with a preceding a/-a; or when a feminine noun ending in 
-a before voiced consonant is N.sg. or N/Ak.pl with dropped visarga etc.)

And also, how will you deal with text which is meant to be read in TWO 
different ways (changing meaning and sometimes gramm. structure), will 
you then type it twice, analysed as per one reading and the other 
respectively ?

This is not  by any means to say that in such cases analysis should be 
avoided, rather to ask for the solution you propose - there should be 
some way, by which you'll indicate to your machine (when further 
processing the text) that there are two possibilities how to analyse a 
certain passage and a) it is not surely decided whether a right one was 
selected  b) both should be taken in account 

______________________________________________________________________________
Mr. Jakub Cejka
Dept. of Sanskrit, University of Pune
Ganeshkhind, Pune, India  411 007

e-mail:  jakub at unipune.ernet.in   (till July 97 the latest)