Don't be too high-tech!!!

Doug Brutlag brutlag at CMGM.STANFORD.EDU
Sat May 8 15:06:30 EST 1993


     With respect to the interface issue to computer resources, Ernie Retzel and
Bill Pearson want to move their user community to the leading edge and to
provide them maximum functionality as fast as their user's budgets allow.
Reinhard Doelz and Alex Reisner are providers of services who don't want to
disenfranchise any large segment of their user community.  Both philosophies are
valid but each in its own environment.

     At Stanford we follow Bill Pearson's approach of copying important
resources (Brookhaven, Entrez, etc.)  to hard disks and make them available to a
large user community over a network.  We can do this because of the
infrastructure here (high speed networks, lots of Macs and PCs on the network
and the cost distributed over a very large user base of 800 local molecular
biologists from 80 research groups).

     On the other hand, the latter approach, providing low tech access to these
resources, may require giving up novel functionality, but if you can reach a
much larger user community over a much narrower bandwidth for less money, it is
often the preferred approach.  Alex Reisner's ANGIS system provides X-windows
access to many of its resources locally, but it can serve an entire country via
terminal emulation.

     When the Bionet Resource was developed many years ago, it wanted to serve
biologists whose only network access was a terminal connected to a modem.  How
many biochemists were on the Arpanet in 1983?  Later, by 1985, a large number
had at least e-mail via some network and Bionet introduced FASTP-MAIL and later
FASTA-MAIL again aiming at the lowest common denominator and minimizing the
USER'S investment required to get access to powerful resources.

     The NCBI Blast and Retrieve servers continue this tradition even though
interactive access to these resources could provide higher functionality.
Another excellent example is the ICGEB (International Centre for Genetic
Engineering and Biotechnology in Trieste Italy) that provides access to
molecular databases for over 40 developing countries (UNIDO member countries)
using dial-ups and X.25 links in addition to Internet.

     Personally, I feel that the Internet is one of the most important
equalizing factors in providing resources (via terminal connections and client
servers like FTP, X-windows etc.) to the largest number of people now.  But just
because someone has access to the Internet, does not mean that locally they have
a UNIX workstation with a 21 inch color display.  They may be reading mail on a
VT 100 terminal connected through a long distance telephone to some computer
with X.25 access to an Internet gateway. When you put a new service on the
Internet you should judge its success not by how many people you reach, but by
how many people your resource disenfranchises.

Doug Brutlag
brutlag at cmgm.stanford.edu






More information about the Bio-soft mailing list