>> I don't think newsgroups work too well to distribute software.
>This is done quite frequently on USENET. The more I think about it,
>however, the less sure I am that we would want a parallel mailing list
>to something like this since there are e-mail servers already and we
>don't want to blast BITNET more than we already are 8-). For example,
>we do not have a mailing list for bionet.molbio.genbank.updates. If
>all we're talking about is adding a USENET group without setting up
>mailing lists at the four BIOSCI nodes, we can establish this
>trivially and I'm not sure that we need to go through the usual BIOSCI
>voting procedure.
>>I have also seen in some forwarded messages from Rob concern about
>viruses being propagated around. Since UofH, IUBIO, and EMBL are
>already handling software and I presume checking for these, the group
>could be moderated and contributions would have to be funneled through
>the moderators (these groups would be the obvious choices).
>>Other comments?
>>Dave Kristofferson
The blasting BITNET issue merits serious consideration. This is a
very inefficient mode of mail and file transfer that requires them to hop from
node to node to node down chains. If one of the nodes in a chain goes down,
the mail and files pile up on the antecedent node, then all come pouring
through when it comes back on line. The more "unnecessary" traffic kept off
BITNET, the better off everyone on it will be. It makes more sense to post
primarily notices of availability to newsgroups (and occasional "here's an
example of code which does such and so" stuff), and let the substantive
software be fetched from servers or archives only by those who really want it.
For those whose network supports FTP, why post the software itself to
a newsgroup when archives can be accessed so easily and quickly?
The virus propagation issue also merits serious considertion. Is the
software deposited with servers and archives in fact checked for viruses
before it's distributed? I didn't think so and therefore delete any
executables in packages I get from them, even when they've been deposited by
people we've come to know (electronically) and trust (in one case, the source
code wasn't included and we "really wanted" that software, so I ran it on an
abandoned MicroVAX for a while before installing it on WFEB2).
Fote
=========================================================================
Foteos Macrides Worcester Foundation for Experimental Biology
MACRIDES at WFEB2.BITNET 222 Maple Avenue, Shrewsbury, MA 01545
=========================================================================
