[comp.sys.3b1] premature dropping of unix-pc.*

gst@gnosys.svle.ma.us (Gary S. Trujillo) (02/15/91)

In <1991Feb12.072818.24706@i88.isc.com> botton@i88.isc.com
	(Brian D. Botton) writes:

> Hi Netland Kit Builders,
>   This will probably be my last multiple group posting, I can't tell you how
> glad I am to have the new groups, ;-)...

I don't want to pick on Brian, but I selected this posting to follow up since
Brian has been a regular poster for a while, and makes explicit mention of
his dropping unix-pc.general from the Newsgroups line of his posting.

My opinion is that not only should the unix-pc.* newsgroups be kept around
for a few months, but that we should all continue to cross-post to them for
this same time period (say through the end of April).  Some of us are in
the position of getting only a partial news feed, and having to negotiate
with a busy system administrator at a net-neighbor site in order to pick up
the feed of the new comp.*.3b1 newsgroups.  After several weeks of poking
and prodding and gentle reminders, I finally succeeded myself, but I'm sure
I've missed quite a bit of traffic in the process.

I thought the whole point of having a few months of keeping the unix-pc.*
newsgroups around was to permit people to switch over gently and gradually,
without missing anything.  If people stop cross-posting to those groups,
the whole scheme falls apart.  Sure, it provides additional incentive to
pick up the new newsgroups, but just because they're carried in the main-
stream hierarchy is no guarantee that they're going to be easy for every-
one to pick up, in fact, I can think of some folks who are at the end of
a UNIXpc-based chain who are dependent upon net-neighbors who currently
have no access themselves to the new newsgroups.

So - I make an appeal:  Please continue to cross-post to unix-pc.general and
unix-pc.sources (I don't care so much about unix-pc.uucp and unix-pc.test)
at least until the end of April.

There, now I feel better!  :-)

Gary

P.S. to people in this position - did you know that unix-pc.general has
     dried up to a trickle, while comp.sys.3b1 is going like gangbusters,
     and that there have been a number of goodies posted to the new
     sources newsgroup???  I'll post a list in unix-pc.general and in
     comp.sys.att to let you see what you're missing.
-- 
    Gary S. Trujillo                            gst@gnosys.svle.ma.us
Somerville, Massachusetts              {wjh12,bu.edu,spdcc,ima,cdp}!gnosys!gst

crtb@helix.nih.gov (Chuck Bacon) (02/15/91)

Trujillo's comment is just another of those I've seen bemoaning the
slowness of some neighboring news admin to adopt comp.*.3b1.

My own relation with news is that I'm mighty glad that there's a
fairly competent admin on my local nntpserver (I get it all at work).
I got comp.sys.3b1 quite early, and I've seen unix-pc.* slow to a
halt (I think things are getting forwarded (?) to comp.*.3b1).

The biggest surprise is that after a year or so of seeing an average
of 0 to 10 articles per day in unix-pc.*, now on comp.sys.3b1, I'm
seeing never less than ten, up into the forties per day!

In other words, I think that news admins take comp.* more seriously
than other newsgroup names.  Of course, it's just possible that
a whole lot of people are seeing and posting to the new groups,
who never even knew about unix-pc.*.

Finally, I must agree with Gary Trujillo's posting, that the old
groups shouldn't be abruptly abandoned.  Cross post to unix-pc.general
for those who don't get comp.sys.3b1 yet!  Let the agreed-upon time
schedule play out.

Oh yes.. does the readership of comp.*.3b1 feel that perhaps we
should consider _selling_ unix-pc.* (5 count 'em! 5) to the people
who, stuck with Intel boxes, would like a real operating system?
I briefly got the impression that some of them were in fact taking
it over 8-)

--
Chuck Bacon - crtb@helix.nih.gov - 301-496-4823
	"After all, computers have rights too!" - Ernst Bacon, 1898-1990

kak@hico2.UUCP (Kris A. Kugel) (02/15/91)

In article <987@gnosys.svle.ma.us>, gst@gnosys.svle.ma.us
(Gary S. Trujillo) writes:

> Some of us are in
> the position of getting only a partial news feed, and having to negotiate
> with a busy system administrator at a net-neighbor site in order to pick up
> the feed of the new comp.*.3b1 newsgroups.  
> . . . I'm sure I've missed quite a bit of traffic in the process.
> 
> If people stop cross-posting to those groups,
> the whole scheme falls apart.  

Evidently, that was happening even at hico2, where I pay real bucks
to get remote feeds.  Not every administrator has been swift to
add the new groups to the distribution, and I think even some
who wanted these changes were under the impression that our problems
would be "fixed".  Others put all unix-pc.test groups in their "alias"
file (what happens when you have one half of a dual-posted article
aliased to the other half?)

In reality, I think we won't get proper distribution of comp.sys.3b1
and comp.sources.3b1 until:

	1. we kick and scream
	2. we figure out where the stoppages are
	3. we get a couple more "create group" messages sent around
	4. we get our groups into the regular "checkgroup" message

> So - I make an appeal:  Please continue to cross-post to unix-pc.general and
> unix-pc.sources (I don't care so much about unix-pc.uucp and unix-pc.test)
> at least until the end of April.
> 

NO NO NO NO don't crosspost to unix-pc.sources - we already had that
problem.  [ --- but it was such a LITTLE articles.  *sigh* ]
I suspect that the cross-postings dissappear on systems that alias the new
group.  I think the solution to the comp.sources.3b1 lack-of-distribution
is a proper archive (accessible by anom. uucp), so we won't miss anything.

What I would have LIKED to have done, is to be able to compare
which machines articles posted to unix-pc groups alone got to,
and which machines comp.sys.3b1 articles got to.  Then compare
the lists to see which machines don't pass articles in comp.sys.3b1
(that were passing the old pc groups) to see who to talk to.

Why are we aliasing unix-pc.test ANYWHERE?  That would have been
a good place to post copies of a test message, and there's really
no purpose for the messages in the group except to test posting
to the unix-pc distribution.

*sigh*  Well, I'll do what I can to identify the holes with the
articles that are currently on hico2.  If anybody has about a
month's worth of articles WITH headers on their machine, maybe
they can check this too.  I'll come up with another script
(probably with another bug :-) :-( ) that'll generate a list
of machines (sort of like that I used to post the usage analysis
a while ago)

Meanwhile, check to make sure the machine YOU are on is carrying
the new groups.  If it's not, get your administrator to fix it.
                               Kris A. Kugel
                             ( 908 ) 842-2707
                      uunet!tsdiag.ccur.com!hico2!kak
                        {daver,ditka,zorch}!hico2!kak
                      internet: kak@hico2.westmark.com

tale@rpi.edu (David C Lawrence) (02/17/91)

In article <1066@hico2.UUCP> kak@hico2.UUCP (Kris A. Kugel) writes:

   In reality, I think we won't get proper distribution of comp.sys.3b1
   and comp.sources.3b1 until:

           1. we kick and scream
           2. we figure out where the stoppages are
           3. we get a couple more "create group" messages sent around
           4. we get our groups into the regular "checkgroup" message

  5. You just sit back with the patience you displayed so well while
suffering under the distribution of unix-pc.* trying to get these
groups created and just let time do its thing.

4 and 5 will do the most toward that end.  1 and 3 (1 mostly) will do
the most to piss people off, and 2 will likely waste more of your
producutivity than it will help the overall picture.  It generally
takes in the area of two months or so for a new, mainstream group to
level off near the propagation percentages it will tend to have during
the course of its existence.  Part of this is because some admins wait
until Spaf's next checkgroups or Active Newsgroups comes out.  Another
part is just slowness of some admins and feeds to arrange themselves.

   I suspect that the cross-postings dissappear on systems that alias the new
   group.  I think the solution to the comp.sources.3b1 lack-of-distribution
   is a proper archive (accessible by anom. uucp), so we won't miss anything.

On C News systems doing refiling the cross-posted article will appear
in the group as many time as it is being filed there; ie, a
cross-posting to comp.sys.3b1 and unix-pc.general will be seen twice
(with just one inode being used) in comp.sys.3b1 on C News sites doing
such refiling.  The problems with this are that most newsreaders do
not do Xrefs until after you leave a group, and that for people
looking at Subject summaries they will see the same article in the
summary multiple times.  This is the way C News is though, at least
for now.  It also does not rewrite the header to change the group
being refiled in any way, unlike B News and aliasing.
--
    (setq mail '("tale@rpi.edu" "uupsi!rpi!tale" "tale@rpitsmts.bitnet"))

mvadh@cbnews.att.com (andrew.d.hay) (02/18/91)

In article <1066@hico2.UUCP> kak@hico2.UUCP (Kris A. Kugel) writes:
[]
"file (what happens when you have one half of a dual-posted article
"aliased to the other half?)

you get *both* articles, and one doesn't get junked when you read the
other (at least at our site...).

-- 
Andrew Hay		+------------------------------------------------------+
Ragged Individualist	| 			JAAAAAAANE!		       |
AT&T-BL Ward Hill MA	|	    HOW DO YOU STOP THIS CRAZY THING?	       |
a.d.hay@att.com		+------------------------------------------------------+

jeffrey@sci.ccny.cuny.edu (Jeffrey L Bromberger) (02/19/91)

In article <1991Feb18.113841.19560@cbnews.att.com> mvadh@cbnews.att.com (andrew.d.hay) writes:
>In article <1066@hico2.UUCP> kak@hico2.UUCP (Kris A. Kugel) writes:
>"file (what happens when you have one half of a dual-posted article
>"aliased to the other half?)
>you get *both* articles, and one doesn't get junked when you read the
>other (at least at our site...).

We're running Cnews here, and we get 2 copies of the article.  There's
only one on the disk (the multiple copies are linked), but it's still
scary to see 100 new messages in comp.sys.3b1 (cause of crossposting
to c.s.att and u-p.general)!

j
-- 
Jeffrey L. Bromberger
System Operator---City College of New York---Science Computing Facility
jeffrey@sci.ccny.cuny.edu			jeffrey@ccnysci.BITNET
	Anywhere!{cmcl2,philabs,phri}!ccnysci!jeffrey