LSTSRV-L Archives

LISTSERV Site Administrators' Forum

LSTSRV-L

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
"F. Scott Ophof" <[log in to unmask]>
Wed, 14 Jul 1993 21:21:06 -0400
text/plain (103 lines)
On Wed, 14 Jul 93 16:58:38 EDT Wes Morgan said about Duane's posting:
>Now that he's vented his spleen, I'm going to vent mine.
 
>Our buddy wrote:
>>LISTSERVER is a unixfied version of what the author thinks LISTSERV
>>does.   I understand it is writtin in C.   A student worker here said
>>the code is poorly written.
>Flaming code that you've (apparently) never examined and (apparently)
>couldn't understand is as useless an exercise as I've seen.  I'd sug-
>gest that you put that nice young "student worker" right to work on
>writing your LISTSERV clone for you.  It can't be that hard, right?
 
Wouldn't it be better to ask Duane to ask this student worker for
more details?  Maybe a thing or two could be LEARNED which might be
worth implementing in this or that MLM.
 
 
>The *classic* characteristic of PD software in the Unix world is
>a simple one -- GROUP EFFORT.  I know of very few major packages
>that have not involved dozens, even hundreds, of people.  Every-
>one takes the time and effort to help out, testing software on
>their particular installations and providing patches/workarounds
>for everyone.  Do you think that packages like gnuplot, mush, elm,
>perl and IDA sendmail just appeared in an FTP archive one day?
 
This characteristic is also quite classic in the BITnet world, at
least in most stuff I've seen from '88 on.
What drives *me* up the wall is that when people make suggestions
for stuff which they think might be an improvement, and/or ask
whether it would be an improvement in general, and even when they
CLEARLY state they cannot implement it for whatever reason, the
result ON THE INTERNET is quite often that these people get flamed
from here to Arcturus (and back), are told off, or are summarily
replied to with "hack the code".  Ie. INTERNET people seem to have
very long toes.  Now THAT characteristic (long toes) does NOT seem
to be as general ON BITNET, in fact, quite rare in my experience.
 
I've seen enough postings on BITnet that are outright flames but
where the flame is ignored or joked about, with the meat of the
posting being taken into serious consideration.
I've also seen way too many VERY POLITE postings on the Internet
where the ONLY result is nasty and uncalled-for flames.
 
Now can we quit feeling that our precious toes are being stepped on?
Just because a product is "free" does NOT mean its lesser points
cannot be criticised.  Also, contributions to products can be made
in numerous ways, one of which is pointing out certain aspects of
those products.  Most of us contribute at whatever level we can.
 
I've seen people contribute with "only" a fanfare-trumpet-hurray
posting to some list, thanking the implementor(s) of a product!
But a "thank you for a job well done" is not a "contribution",
right?  And documentation (of which user help-files & such) are of
course totally irrelevant, and one shouldn't waste any time on such
useless crap, right?  Because who the hell cares what a USER thinks
of a product?!  And why should one EVER take the trouble to even
consider that a USER might think differently than the implementor(s)
of some product?
 
My suggestion:
Pull in those toes and try to think like a USER.  Just once...
If you can't, hey that's OK (nobody's perfect), talk with a GOOD
teacher, or to people at the help-desk who have a good reputation
interfacing with users (or ditto computer consultant at the CC).
They'll be glad to help you out; it's part of their work.
 
 
>Whining about it on every forum you can find (how many lists did this
>message hit?  4 + a Usenet newsgroup? ) doesn't achieve a single thing.
 
Duane gave me permission to post it, and *I* decided to post it
to 5 discussion centers.  Some people have already responded with
positive comments and requests for suggestions and more details.
IMHO a good start towards improvements related to MLMs in general.
 
 
>>LISTSERVER is also not as robust as the Revised LISTSERV.   IT
>>appears that processing a list with slightly over 400 subscribers
>>puts a strain on the server.
..[Might be more hardware- than software-related]..
>Did you stop to think that LISTSERV has been hand-tuned for a
>specific set of architectures, while Majordomo, Listserver and
>other Unix MLMs span a huge segment of the Unix spectrum?  (That
>is NOT intended as any sort of slam on Eric -- LISTSERV is a great
>piece of work)
 
So what are we all gonna do?  Sit on our halos, feel we've done our
best, stick out our long toes, and leave it up to the poor USERS to
figure out how to talk with all these differing MLMs?
Or are we going to try to figure out how to improve all these MLMs
even more?
 
 
>Feeling *much* better now,
>--Wes
 
Really?  (grin)  Or have I managed to find some toes/feet not yet
stepped on?  >;->
 
 
Regards.
$$\

ATOM RSS1 RSS2