[1319] in SIPB-AFS-requests
Re: server migration
daemon@ATHENA.MIT.EDU (Derek Atkins)
Fri Apr 15 15:08:29 1994
To: sipb-afsreq@MIT.EDU, charon-maintainers@MIT.EDU, rtfm-maintainers@MIT.EDU,
webmaster@MIT.EDU, usenet@MIT.EDU
In-Reply-To: Your message of Fri, 15 Apr 94 14:45:50 -0400.
<9404151845.AA05852@hope.MIT.EDU>
Date: Fri, 15 Apr 94 15:07:02 EDT
From: Derek Atkins <warlord@MIT.EDU>
[Propsals deleted]
The biggest problem with moving charon services over to the ss10, as
opposed to a ds5000.25, is that many of them require binaries that
only work on the dectstation (like the BITNET gateway), and have
byte-order problems (like discuss).
I'm not saying that it would be impossible to get around these
obstacles, but it is something that we should think about, since it
would take a non-trivial amount of time and effort to get everything
to work on a new platform. (Remember all the time and effort it took
to move charon from a vax to a decstation? And that didn't even have
the byte-order problems! :-)
Just to reiterate what has already been said by others, I too think
SIPB *NEEDS* to have a machine that people can use to test
experimental programs.
Also, if www is taking up that many of our resources, maybe it is
better to have IS run the service. One of the good things about SIPB
is that we can expend the time and energy to try new things. We have
the ability to augment Athena/IS, and try to support things that they
wont, or bring in new ideas. For example, LaTeX was first supported
by the SIPB, before Athena started to support it in about 1990 or 91.
Also, if my history is correct, it was SIPB that put up the first
zephyr server, and then Athena took that over when it became too much
for SIPB to handle.
Basically, I am saying that there is a precedent for having IS take
over services that SIPB started, and if WWW becomes something that
starts reducing our ability to bring in new services, then maybe
turning the service over to IS is the right solution.
Thanks for listening.
-derek
