woods@hao.UUCP (Greg Woods) (02/10/84)
I am not going to create subgroups, because the Rich Rosen's out there
will be too quick to say I am being selfish by doing so. I still support
the idea, but not for the reasons Rich thinks. I do *not* want to create
a net.music.dead . Yes, I like the Dead a lot. But the reason for creating
subgroups is *not* so I can go off in a corner and forget about all other
types of music (some of which I *do* like, believe it or not! :-), but rather
so others don't have to exercise their 'n' keys too much. The Dead,
*regardless* of my personal taste, Rich, clearly do not generate enough
discussion to warrant a group of their own. Rock music, on the other hand,
clearly does. In fact, net.music is so overwhelmed by rock articles that
we might as well be calling it net.rock . I *don't* get a chance to read
about any other types of music, even if I wanted to, and neither does anyone
else.
I think this discussion is important because it goes beyond the net.music
group. Other groups (like net.religion with the recent proposal for a subgroup
for Judaism) could benefit from this. Rich Rosen and I are clearly at total
opposite ends of this issue, with most everyone else somewhere in between.
This issue needs to be resolved, because it will come up again and again,
if not in net.music then somewhere else. But it should't be resolved with
a war between Rich and myself where I create a group and he rmgroups it (I
wouldn't put it past either of us, from what I've seen so far!) Most people
seem to agree with Rich so far, but I'm still getting mail. I will post
a summary of the responses I have gotten when the flood dies down.
P.S. Net.records is dead. Both the sub-groupers and anti-subgroupers
seem to almost unanimously agree on that. As soon as I figure out how,
I am going to rmgroup net.records .
GREG
--
{ucbvax!hplabs | allegra!nbires | decvax!stcvax | harpo!seismo | ihnp4!stcvax}
!hao!woodsmp@whuxle.UUCP (Mark Plotnick) (02/13/84)
One problem with the newsgroup hierarchy system right now is that it's
only superficially a hierarchy. The only thing it buys presently is
protection against the 14-character-filename limit and a way to
selectively unsubscribe to topics of more limited interest. But you
may not even see the groups and subgroups presented in a treelike order
unless someone is keeping your active file alphabetized. And, even if
you type U to every net.micro subgroup, you'll still see every new
subgroup that pops up (unless you've discovered how to edit your
.newsrc file).
The current debate has produced some arguments based on misinformation,
pessimism, and a good bit of elitism. Don't forget that:
- not everyone is on a free hardwired 9600baud terminal
- not everyone can spend 5 hours a day reading netnews
- not everyone can tell whether an article is worth
reading by merely reading its title (especially if it's as
meaningful as "Re: Orphaned Response - (nf)" )
- not everyone is willing to read through 100 articles (or headers)
in order to find one that interests them. They should still be able to
function productively on this net.
- the creation of one subgroup is not necessarily going to lead to the
creation of 20 others. More to the point, the creation of one subgroup
under a given group doesn't warrant the creation of every other possible
type of subgroup under that group.
- the creation of a new subgroup to serve a special-interest segment
is no more discriminatory than creating a new department
in a department store that's in a shopping mall. (hmmm, does this
make us the shoppers or the management?)
- the active file is not too large (yet). Clean out the newsgroups
that were created by mistake and you'll have lots of room (I'm
amazed that net.db keeps popping up).
Forming subgroups is fine if done correctly and in moderation, allowing
for the fact that the software needs a little more work. Those people
who would read everything in net.foo will probably read net.foo.all.
Those who would never read net.foo (e.g. because net.foo is a
high-volume newsgroup filled with flamers) may be happy to read and
maybe even participate in net.foo.subgroup1. In such a case, how
can you object to creating such a subgroup?
Finally, ponder this: if there isn't at least one subgroup under a
top-level group, then perhaps that group doesn't belong at top-level in
the first place, and should be a subgroup of something else. Witness
net.{unix,unix-wizards,bugs,usoft} which will forever be separate
because of their heritage.
Mark Plotnick
WH 1C-244 x6955