GENBANK TRANSACTION PROTOCOL" AUTHORIN VAX/VMS

David Kristofferson kristoff at genbank.bio.net
Mon Jul 15 12:29:44 EST 1991


>
> Is there at least some C code people could have to attempt a port?  (BTW:
> This is NOT an offer, we use Macs, thats all we need....)
>

There is nothing proprietary about the source code, but it was
developed using the XVT interface libraries, and I believe that one
needs a source license from them to do the work.  Julie Ryals is the
project manager and could comment in more detail.

A suggestion in the next posting also mentions letting "a thousand
flowers bloom."  This is one area in which I personally have grave
doubts about the wisdom of letting the program be developed by 1000
people at 1000 different sites (make that 2000 sites since people are
always changing labs).  That is a recipe for **disaster** in my book.
It makes far more sense for one group to be responsible for the
software that allows the data to be automatically read into the
database (i.e., read "contractually obligated" so that someone can
"get on their case" if something doesn't work).  Making sure that the
software is updated if anything changes in the database format,
maintaining a list of all users and making sure that they receive the
latest updates, providing technical support for the program, etc., are
important tasks because obsolete versions might cause problems (or at
least be less efficient) for the database efforts.  We have already
experienced some difficulties of this nature as it is, and I believe
that the thousand flowers approach would undoubtedly produce a weed
patch instead of a rose garden.  Given all the flak about data
problems, etc., that the databases have received over the years, this
approach would lead to more confusion and would *not* be a service to
the user community.

				Sincerely,

				Dave Kristofferson
				GenBank Manager

				kristoff at genbank.bio.net



More information about the Bioforum mailing list