Do we support 2.2.1 or not ?
rra at stanford.edu
Thu Oct 7 09:15:23 UTC 1999
Hans Lambermont <Hans.Lambermont at garg.mpn.cp.philips.com> writes:
> Though, when i look at INN from a distance nowadays, i see a very
> enthusiastic group developing it, but at the cost of stability. And i
> think the focus shifted too far to development. I deploy INN for about 5
> years now for a major worldwide company's intranet, and stability is my
> major concern. Perhaps it is the combination of my platform (BSDI) and
> INN, i don't know, but i find the stability lowering over the years. I
> think that is sad.
Well, some of my opinion is probably revealed by the fact that my reader
machine is running 1.7.2-insync-1.1d, which was a remarkably stable
version of INN and would last quite a while.
I think the main reason why stability went down for a while is that the
introduction of the storage API and the conversion of INN to autoconf were
both really major structural changes to the package, and it takes a while
to finish those kinds of changes. We've just now, in the current
development version, gotten all the way to a pure storage API system, and
we still haven't completely gotten rid of subst. There have also been
several drastic overhauls of the overview code, and that's caused some
I think INN 2.3 is shaping up to be another version that will be quite
stable for a while, and we're almost there in terms of finishing up the
structural changes that INN needed.
Russ Allbery (rra at stanford.edu) <URL:http://www.eyrie.org/~eagle/>
More information about the inn-workers