[402] in Enterprise Print Delivery Team
Re: 7/31 Enterprise Print Delivery Project Status Report
daemon@ATHENA.MIT.EDU (Bill Cattey)
Wed Aug 2 07:25:11 2000
Resent-From: David F Lambert <LAMBERT@MITVMA.MIT.Edu>
Resent-To: Enterprise Printing Delivery Project Team <printdel@MIT.EDU>
Message-Id: <ctVtgNIG4U4l16eK40@mit.edu>
Date: Tue, 1 Aug 2000 23:52:57 -0400 (EDT)
From: Bill Cattey <wdc@MIT.EDU>
To: David F Lambert <LAMBERT@MITVMA.MIT.EDU>
In-Reply-To: <10008011750.AA28614@MIT.EDU>
For Rocklyn & Lynne. I already acknowledged Bill's note. -Dave
----------------------------Original message----------------------------
I noticed mention of DCE in the past two enterprise printing reports.
Caution: DCE is very large, and a dying infrastructure model.
It's analagous to the Microsoft domain controller model in many ways:
It's hard to take just the nose of the camel in your tent. In one way
it's different: it has very few new customers.
If you've got an established system that has DCE and you can kludge it
to use an existing kerberos v5 infrastructure, that's cool. But you may
end up dedicating a server to the DCE time service and the DCE name
service,
and possibly other services, just to keep the elephant fed.
I used to often say, "DCE: Free elephant shipped with every packet."
The Athena printing redo gave us an opportunity to thing things through
and simplify them. The worst code in the old system was based on a
simpler precursor to DCE that fell out of favor. Thinking about using
DCE for a print service brings to mind the memories of the failed
Palladium Athena project where tons of complexity and ECMA compatiblity
came in, but reliable printing of pages never resulted.
BEWARE!
-wdc
