Keith Robison (robison at nucleus.harvard.edu) wrote:
: a handsome chappy!! (jenkins at aidsun.nibsc.ac.uk) wrote:
[...]
: : The reason i'm asking is that we as a network are going 'internet' soon if that makes sense.
[...]
: : To access these services, what software package(s) are required.
[..]
: For simple usage, E-mail servers require no special software --
: you just format a mail message and send it. Some software does
: come with hooks to E-mail servers -- the one I am most familiar
[...]
When will Biologists ever learn? Electronic Mail is the most unfriendly
and least controllable software environment which is available. Since
the 'Bitnet' type of mail with semi-interactive jobs disappreared (Don't
understand me wrong- Bitnet had its weaknesses, but was excellent in other
aspects where Internet mail has a long way to go!) the Internet developed
much more exciting and challenging protocols than electronic mail. If
Biologists get hammered in that they should use Mail 'cause there is no
need for special software this is only partially correct. I heard quite
often that you need 'simply' to format a message. You don't swim across the
Atlantic as you don't need anything than a bathing suit which is easy
to jump in?
Seriously, the Internet in general (i.e., the several-million hosts
on the IP suite of protocols), and the application software out there is
much more powerful than you might gather from this group's traffic. First,
the X-windows system is already a network application on the LAN (I know,
most Biologists don't have it). Then, tools from the synchronous information
retrieval like GOPHER and WWW emerged into reasonable biological tools which
you should definitively run (LYNX, for example, is perfectly happy with VT100
as terminal, and can be used to browse many archives with WWW).
Admittedly, the asynchronous nature of searching does not easily fit into
the current protocol applications as those are mostly synchronous. However,
they don't run so badly that they weren't worth mentioning! The NCBI offers
their BLAST network application already for some time, and on the EMBnet
we run HASSLE since 1992 on international scale. The biggest problem is
to tell granting agencies that there is a need to DEVELOP those tools, and
to tell system managers to INSTALL what is already out there!
Sorry for the heavy arguments against electronic mail, but repeatedly
advertising e-mail based services is the wrong direction. Unless we (the
biological community) really start using the network as NETwork rather
than a Postman Pat schema no one will take us serious enough to develop
the promising thoughts further on.
And, certainly, the same applies for FTP archives - Copying files around
might be the cheapest way for system managers but it is the most troublesome
excercise to access biological data, and certainly not convenient or
economical. If you tell me that it is economical that all use eMail just
because the access to software is FREE (?) I would suggest that we launch
a small competition to figure out how expensive it will become if all of
use rely on a few centers and have no longer the software on-site which
allows more sophisticated use of data. Remote data and retrieval? Certainly.
Networks of servers rather than a few? Most probably. More and better pro-
tocols to develop (This is SCIENCE!) in order to achieve this? Unavoidable.
Electronic Labs with out walls? No way to go without. EMail? Archaic.
Regards
Reinhard
--
+---------------------------+-------------------------------------------+
| Dr. Reinhard Doelz | Tel. x41 61 2672247 Fax x41 61 2672078 |
| Biocomputing | electronic Mail doelz at urz.unibas.ch |
|Biozentrum der Universitaet+-------------------------------------------+
