🚀 go-pugleaf

In article <3hl8gr$p6p@panix.com> geert@panix.com (Geert K. Marien) writes:
> What I would like to do is find out, for the people I have now on the 
> list, what THEIR propagation is.  I.e. the people that will switch.  I am 
> really interested in this (maybe we can talk off-group and you can show 
> me how to find out these rates, please?) as it might help me see if what 
> people are telling me in the survey is accurate.

I would be happy to help off-line, but since I have a public answer
for another question I will publicly answer this too: unfortunately,
there is no really good method for finding out what you want to find
out.  Your survey is probably the best realistic way that there is,
but it does have some weak spots.  What percentage of the subscription
base has replied?

In general, I would be very surprised to find a site that took bit.*
that did not also take both misc.* and rec.*.  So I just took another
rough look at the hundreds of sites fed by UUNET.  Only 20% of them
take bit.*, compared to 40% that take misc.* and 37% that take rec.*.
Of those that take bit.*, only 4.5% (ie, about 1% of the total number
of feeds) do not take both misc.* and rec.*.  One of them is pretty
clearly using us to get only non-maintstream hierarchies, as they
don't take any of the Big Seven + alt.  Looking at the others who take
bit.*, I'm not to sure what their motivation is but it is a pretty
even mix between whether they don't take misc.*, don't take rec.*, or
don't take either.  However, combined they are less than 1% of the
total number of feeds and statistically insignificant.

> Lastly, are bit. articles distributed all over, or would a site out in 
> the wild frontier need a specific feed site?  By this I mean, are the 
> articles transmitted to them and they just choose not to pick them up, in 
> general?

Generally the situation is that articles do not get transmitted to a
site unless they want them, because it is a big waste of resources
otherwise.  bit.* is available around the world, but if a site's
existing neighbors are not carrying it then they would have to arrange
a feed of the group from some site that is getting it.  They should be
able to find a site within their locale that carries bit.*, but it
will be harder than finding rec.* or misc.*.