In article <32E9BA59.702 at gab.unt.edu>,
C. J. Fields <cfields at gab.unt.edu> wrote:
>> Listen, I am *really* not trying to piss anyone off. I think we agree on
>> most things - actually, I'm not sure exactly how we *dis*agree...
>>The disagreement lies in the misconception that any lab can get these
>programs. The disagreement lies also in the fact that, in order to
>publish in a highly regarded journal (Gene, J. Bact., etc.), your
>results must be 'publishable', and most freeware isn't able to do this.
I agree that software such as GCG really needs to be bought on a
separtmental or institution basis; it's too expensive for anything
else, and not just because of the cost of the software. The real cost
in maintaining a GCG service is that you really require a full time
person to maintain databases, answer users' queries and perform
routine system administration (this is the task I perform here in
Cambridge). Of course, support staff don't come cheap, but they are
essential for efficient use of your time. They can only really be
afforded on a much larger scale than a research group.
>BTW, the original message has nothing to do with the present argument
>(i.e. you are right). However, that doesn't diminish the argument
>itself. Commercial software, esp. for molecular biology is way too
>overpriced (come on, $3000 for a program that can't be networked and
>requires a passkey?).
I agree, that really is pricing out of the reach of most groups.
>The only package I've seen that seems to be worth
>the high price is GCG, and even that is coming under fire (no source
>codes are being released in the new version). At least with GCG you can
>network it for less than $10000 (I think about $4600, to be exact), so
>maybe the entire school, and not just one lab, can benefit. I'm truly
>surprised that they don't dominate the market with their package.
Hmm. Networking GCG is *not* cheap. You need the supoprt staff (as I
said above), and the hardware you require is expensive.
Unfortunately, GCG are refusing to support Intel based UNIX systems,
otherwise we'd be laughing. Parts of GCG are *very* RAM hungry, so
even if you buy a low-end workstation to run it on, you still need
lots of memory. Our current machine has 128 MB RAM, and this is fine
for the main searching programs (even BLAST, which is notoriously RAM
hungry), but the srsbuild program caused this machine to crawl; this
one program allocated over 300 MB of virtual memory while rebuilding
SRS indexes so that lookup would work, and took days to complete. I
would say that given the size of databases now, your UNIX box tht you
run GCG on really should have at least 256 MB RAM, particularly if
there are a lot of users (mole.bio.cam.ac.uk has well over 1000 users)
We are currently in the process of buying a replacement machine; a
four-processor Silicon Graphics Origin 200 with 512 MB of RAM, which
will help address the problems we've been having with the larger
databases. This machine has cost us about $70,000 not $4,600...
>C. J. "I'm not a GCG saleman" Fields
:-)
Tim.
