[Fwd: [Mod_gzip] [ANOUNCEMENT] mod_gzip 1.3.19.2a documentation update]

Looks like the cluebat landed and is actively working.

So, should we post a reference in the FAQ?

Rob

-----Forwarded Message-----

> From: Michael.Schroepl@telekurs.de
> To: mod_gzip@lists.over.net
> Subject: [Mod_gzip] [ANOUNCEMENT] mod_gzip 1.3.19.2a documentation update
> Date: 13 Sep 2002 15:12:17 +0200
>
>
> Hello all,
>
>
> > I'm proud to announce a new version of mod_gzip ...
> > So I approved a Sourceforge project
> > (http://sf.net/projects/mod-gzip/) ...
>
> ... and to round up the whole thing:
>
> http://www.schroepl.net/projekte/mod_gzip/
>
> has experienced a major update and should now be in sync
> to (hopefully ;-) all the changes of mod_gzip 1.3.19.2a.
>
> Document sections that have changed from the last version
> (mainly because of the new mod_gzip release, but even some
> more) are highlighted with a yellow background color - so
> you can quickly look over all the files and find out what
> is different now.
>
> > So, any criticism is welcome!
>
> Which of course applies to the documentation as well as
> to the source code.
>
> We want to encourage you to at least thoroughly read the
> changes and then decide what to do with the new version.
> At least consider using it because of the "Vary:" header
> it will now create - but be sure to read the documenta-
> tion about this feature carefully.
>
> Also, the feature of using static precompressed files
> (that until now has been awkward to use at least) may
> become a potentially useful thing from now, as mod_gzip
> 1.3.19.2a will at least detect outdated static versions.
> The next step will be to automatically update them - and
> then we are already close to automatically create them
> in a similar way mod_proxy is maintaining its cache tree.
> gzip_cnc has proven that it will save a lot of compres-
> sion steps (only about 3% of all requests currently lead
> to actually gzipping the content of my domain), so a
> mod_gzip that conditionally compresses only outdated
> versions may well save a lot of CPU time, and then even
> use the maximum compression level (although the additio-
> nal effect compared with the current mod_gzip compression
> will only be marginal).
>
> Some of the changes are rather in their starting phase -
> for example, the logic for calculating the 'perfect'
> "Vary:" headers will have to be improved in forthcoming
> releases (which is already described in my documentation
> pages - but will need major code restructuring).
>
> >From my point of view the current release in the first
> place should show that mod_gzip is alive and kicking,
> that now again there is a C programmer able to under-
> stand and change the source code (which is Christian,
> not me!), and that there is a new home page for mod_gzip
> that is waiting for your visit.
>
> I encourage everyone to use the SourceForge functions to
> contribute to the project.
> You can file bugs there, and post feature requests - any
> ideas about what would make the project better can be
> stored there and will be more transparent than buried in
> the depth of the mailing list archive.
> Especially the testing of features and configuration on
> different platforms can easily be done by a lot of people.
>
> And we would definitely need help to create DLL/DSO binary
> download versions for 'exotic' platforms, if users want
> those versions.
> There are already other mod_gzip downloads on the net, like
> the port to OS/2 by Andrea Brancatelli - we will try to
> bundle all these spin-offs (at least link to them if we
> get to know about them).
> So if you have seen something about mod_gzip that is mis-
> sing in my links list, please tell us ...
>
> Regards, Michael (and Christian, of course ;-)
>
>
> _______________________________________________
> mod_gzip mailing list
> mod_gzip@lists.over.net
> http://lists.over.net/mailman/listinfo/mod_gzip
>