Wish List

Craig Anderson caa at noaacrd.Colorado.EDU
Wed Oct 6 13:32:10 EDT 1993


----- Begin Included Message -----


Raymond Venneker <venn at geo.vu.nl> writes:

>O.k., suppose the user community makes a couple of good modifications,
>or comes up with new features. But what are the chances it will become
>part of the next edition of GRASS? I assume it depends on judgement and
>willingness of CERL. I've made modifications to a few programs. However,
>it is getting a pain in the neck to make sure they are continued in every
>new version/update.

	 Admittedly it is up to the CERL folks to incorporate user donations/
modifications into subsequent releases of GRASS after it has been in use
by the community to ensure sufficient testing and debugging. 

	One problem that I see right away is if a user puts something in the
src.contrib (aka. src.beta) directory, how will CERL folks know when the magic 
time for user testing has past and it is now a SAFE program.  Once this step 
occured, it should theoretically be included in src.alpha for another period 
of testing before inclusion into src (aka. main).  This might be unlikely to 
occur unless: 

	a) The programmer keeps a log of user comments and fixes on a program 
while it is in beta testing. 
	b)CERL eventually gets a copy of this log from the programmer to remind
them of a potential need to now include this program (assuming it gets to the 
proper person) into src.alpha where they take over responsibility for fixing
potential bugs.

	As for having CERL include user additions to existing src or src.alpha
programs, I personally believe that only the squeaky wheel syndrome (user
voicing concerns) would ensure its addition to existing code.  This might
be a problem for CERL folks since with such an addition no beta testing by
the entire community has been done and now they have to support it fully.

	I realize that this process is lengthy and sometimes creates
a paradox of user/manager areas of responsibility, but this has to be 
adequately recognized and evaluated by entire developmental base if we hope
to enhance overall system capabilities.  Your pain in the neck could in one
sense, be a necessary sacrafice for the benefit of all users.  Hopefully others
are as willing to endure their share of the pain also, but no guarantees.

Craig



More information about the grass-dev mailing list