> Nobody will write such a compatible server. I know it sounds like a
> gratuitous and dictatorial statement, but it's based on experience. I
> have sent LISTSERV sources, over the last 3 years, to a dozen experienced
> VMS programmers (some of which also had a good knowledge of VM) who
> wanted to port it to VMS. None of them has produced anything that would
> handle even just the distribution list business, which is a relatively
> small fraction of LISTSERV.
As a member of the "nobody" group 8-) I'll say a few words on the subject:
1) Yes, it is a lot of work. Since we have file serving under VMSSERV (a
micro-subset of LISTSERV, but still very useful), and mail list serving with
PMDF, this has become less important to us. I'm not saying we'll never do
it, but it has become less important.
2) Having a number of VMS systems, it is easier to code a new (hopefully
generic) VMS-based system than it is to port a different implementation.
This would let us test and re-test until we got it right, and we could also
conduct net-wide tests without jeopardizing the existing LISTSERV backbone.
3) We would want this mechanism to function across other nets besides Bit-
net. While LISTSERV does this for requests now, emulating that behavior on a
VAX is rather convoluted due to the way mail is handled between multiple
transports.
4) We have not yet come up with a solution which would allow remote admin-
istration of the LISTSERV-on-VMS by a coordinator such as Eric. While REXX
updates can be shipped and be immediately usable, there are some questions
about how they could be (automatically) converted and merged into the mono-
lithic C program our LISTSERV-on-VMS would be.
5) There are a number of architectural differences. For example, on a VMS
system, filenames can be up to 39 characters (for the name, not counting the
type). SImilarly, spooling is low overhead on VM, whereas it can be painful
a VAX.
I welcome the list's thoughts on this, as well as comments on the number
of sites which would be interested in such a program. As the percentage of
VM systems on Bitnet drops, the remaining ones will be picking up more and
more LISTSERV traffic. I fear the time may come where that will be too much
of a burden for them to bear, and we will see a few large list servers, such
as we have with Internet gateways today - a small number of hosts carrying
the entire network load.
Terry Kennedy
Operations Manager, Academic Computing
St. Peter's College
[log in to unmask]
|
