BIOSCI newsgroup creation/termination policy - LONG!

Dave Kristofferson kristoff at GENBANK.BIO.NET
Wed Feb 26 18:52:54 EST 1992

Point One: Do we have a quorum on BIOFORUM?

Before beginning this debate I want to address the "quorum" question
as many e-mail readers have *not* requested BIOFORUM.  Over the past
month there have been at least two calls to BIONEWS explicitly to
direct e-mail users to subscribe to BIOFORUM if they wish.  The last
message a few days ago also informed people of the impending start of
this discussion today in case they were interested.

Currently at GenBank there are 212 on the BIONEWS e-mail list and 41
on BIOFORUM (USENET news readers undoubtedly get both bionet.announce
and bionet.general).  I do not have statistics at the other two BIOSCI
sites, but IRLEARN is typically comparable to our level and Daresbury
a bit smaller.  I'd hazard a rough guess that we may have an audience
of about 100 or so on all three BIOFORUM e-mail lists and my last
estimate of bionet.general USENET readers (Oct. 91) was roughly
500-600.  This should be sufficient to get a decent response from
BIOSCI readers on this forum although we run a bit of a risk in being
"non-representative" here compared to the proverbial silent majority
still only on BIONEWS/bionet.announce.

Point Two: Newsgroup Creation Policy


As Rob Harper noted in a recent posting, the current policy for
newsgroup creation (40 YES votes within 60 days) was adopted by the
BIOSCI managers back in the "good old days" (i.e., slightly later than
yesterday 8-) when there were fewer readers, and we didn't want to make
it impossible for newsgroups to be formed.  Thus the bionet USENET
domain has never been required to conform to mainstream USENET
newsgroup creation rules because they are sufficiently stringent to
make the likelihood of our creating any group in bionet quite small.

For example, for mainstream USENET domains such as comp, first there
is a two week discussion period on a charter for the group before a
vote is taken.  Voting lasts for 30 days versus our current 60 day
period.  A final tally is posted to the net along with the e-mail
addresses of the voters.  Then there is a five day waiting period
while objections can be raised which might invalidate the vote, etc.,
etc., etc.  Voters can vote either for *or* against the formation of
a newsgroup; we only take YES votes on bionet.

To establish a USENET newsgroup, a proposal needs to receive 100 more
YES votes than no votes.  I don't think that we have ever even reached
100 votes on any proposal with perhaps a couple of exceptions.

POINT FOR DISCUSSION: Since the establishment of the 40 vote in 60 day
policy, I think that it is safe to say that the number of users on the
net has at least doubled.  I am proposing that we double the number of
required votes for newsgroup creation to 80 AND shorten the voting
period to 30 days.  Most votes come in within a week of the
announcement of the vote and then again shortly after the second call
which currently occurs about 30 days into the 60 day period.  Nowadays
a 60 day period only delays the inevitable.  I believe that we would
get essentially the same response by posting a call for votes followed
two weeks later by a reminder, and then end the voting period in a
month.  Also note that, since these proposed rules still differ from
mainstream USENET rules, it would be necessary to restrict the call
for votes to BIONEWS/bionet.announce.  Lobbying other USENET
newsgroups to find voters would be grounds for cancellation of the
vote.  Finally, each proposal needs to have a newsgroup charter and a
person designated as the discussion leader (moderated newsgroups can
also be proposed) who is responsible for maintaining a minimum level
of activity on the newsgroup (see Point Three below about newsgroup
termination policy).

WHY TIGHTEN THE REGULATIONS?  BIOSCI "history" includes several
examples of newsgroups that were approved by the voters with some
degree of enthusiasm only to fall shortly later into disuse.  The
bottom line is that this creates essentially needless work for those
of us who run this system in addition to adding unused
newsgroups/directories to news systems around the world.
Unfortunately I have not kept totals for every vote over the last
several years, so I do not have a "scientific" basis for the 80 vote
level suggested above.  I would have preferred to compare voting
totals with the subsequent usage level of the newsgroup, but can't do
this.  The number suggested above is based simply on a rough estimate
of growth and is open to debate (and, of course, later revision if our
initial conclusion proves to create more problems than it solves).

Point Three: Newsgroup Termination Policy


I repeat again the stats for 1991:

                      Usage of BIOSCI Newsgroups
		     1/1/91 - 12/31/91 inclusive
         ranked in order of number of messages posted per day

Newsgroup Name             Messages Posted      Messages per day*
--------------             ----------------     ----------------
BIONEWS                       1379                     3.78
BIO-SOFTWARE                  1179                     3.23
METHODS-AND-REAGENTS           796                     2.18
BIO-INFO                        19                     1.58
NEUROSCIENCE                   305                     1.11
EMPLOYMENT                     392                     1.07
ARABIDOPSIS                    264                     0.94
GENBANK-BB                     335                     0.92
MOLECULAR-EVOLUTION            304                     0.83
BIO-JOURNALS                   212                     0.58
BIONAUTS                       203                     0.56
COMPUTATIONAL-BIOLOGY          135                     0.51
HIV-MOLECULAR-BIOLOGY           10                     0.50
PROTEIN-ANALYSIS               152                     0.42
SCIENCE-RESOURCES              149                     0.41
HUMAN-GENOME-PROGRAM           114                     0.31
BIO-MATRIX                     102                     0.28
POPULATION-BIOLOGY              92                     0.25
AGROFORESTRY                    76                     0.21
AGEING				50                     0.14
EMBL-DATABANK                   48                     0.13
GENETIC-LINKAGE                 35                     0.10
CHROMOSOME-22                   28                     0.08
GENOMIC-ORGANIZATION            27                     0.07

Totals:                       6406                    20.19

* several newsgroups came in to existence during the course of 1991.
The statistic is calculated only over the number of days that the
newsgroup was actually available during 1991.

All newsgroups in this list from AGEING down have less than one
message posted to them a week.  When I posted a warning to AGEING and
CHROMOSOME-22 that they might be shut down, a brief flurry of activity
occurred on both and has since died down again.  GENETIC-LINKAGE has
shown a few small signs of life recently, but GENOMIC-ORGANIZATION is
critically ill.  EMBL-DATABANK has a low volume level but is used by
the EMBL databank to make announcements to the community (and since
they are our collaborators at GenBank it seems to be worthwhile to
keep 8-).  I have had discussions with GDB at Hopkins about taking
over the GENOMIC-ORGANIZATION newsgroup, but there has been no recent
movement on this point.

Of course, once these groups are established there is no absolute need
to eliminate them if they are inactive.  My hope is that a tougher
newsgroup creation policy will prevent future examples like these
groups.  However, we have received complaints from some systems
managers in the past about creating too many unused newsgroups (of
course the complainers are comparing our volume to comp, rec, and alt
domains), so I also feel a certain responsibility not to let these
drag on forever.

POINT FOR DISCUSSION: I propose to set a newsgroup termination level
of one message a week.  This will be determined at the end of each
calendar year and any newsgroup that has fallen below 52 postings for
the year will be terminated.  The current "laggards" will be given
another lease on life to get their usage up by the end of 1992 except
for GENOMIC-ORGANIZATION.  If this newsgroup can not find a sponsor
such as GDB by June 1 I will propose its elimination.  Any group
subject to elimination will be alerted by a posting to the newsgroup
in question.  If at least 80 votes for continuation of the group can
be garnered within two weeks based solely on that single announcement,
the newsgroup will be maintained for an additional two months during
which it must maintain an average usage of one message every three
days.  If it fails to meet this test during its "probationary" period,
it dies a horrible death, no further questions asked.

So, if you have actually read to this point 8-), these are my policy
proposals; feel free to comment, dissect, modify, spindle, fold and
mutilate as needed.  If we come to a rapid consensus, then let's just
do it, drop all of this bureaucracy, and get back to biology.  If this
ignites a raging controversy we may need to have another ballot in a
couple of weeks between any factions that develop.

Please post your responses to any of the following addresses or to
bionet.general in USENET (not to me personally please):

Address                               Location        Network
-------                               --------        -------
bioforum at               Ireland         EARN/BITNET
bioforum at              U.K.            JANET
bioforum at              U.S.A.          Internet/BITNET


Dave Kristofferson

More information about the Bioforum mailing list