> In a message dated 01-02-21 22:47:06 EST, Jim J. writes...
>
> When the patches were originally sent, they were sent in a combination
> of sequential patches. The second depended on the first, the third on
> the second, and so on. So that the tenth patch relied on the first nine
> being applied or it wouldn't apply.
That's right.
It was/is one of the most complicated 'series' of patches that anyone
has ever submitted. The requests for compliance with Apache's existing
coding styles and methods were not unreasonable but the author
seemed to take those 'requests' as 'rejection'. Pity.
As someone who can be immediately identified as an often vocal
critic of Apache ( because I believe all great efforts can benefit from
healthy critique ) this might sound strange but I believe recent comments
directed against 'The Apache Group' regarding the SGI patches are
mis-informed, myopic, and unfair.
What the complainers either do not realize or are conveniently forgetting
is that at the very moment the 10 patches were submitted the group
had ALREADY decided to embark upon and focus their efforts on
finishing Apache 2.0, which actually stands to provide a much greater
performance increase for Apache than a bunch of patches that
optimize loops and wedge hashing tables onto sequential lookups.
This decision took the best talent that Apache has and the people
that know the 1.3 codebase best 'away' from a primary focus on the
1.3 codebase. The very talent that was needed to review and
approve the compilcated 1.3 based performance 'tweaks' did
not have the time to review these patches properly because of the
new focus on 2.0. That's simply the way it was ( and still is ).
So lighten up, everybody.
Give Ryan Bloom and the rest of the Covalent boys and William Rowe
and Greg Stein and many, many others the credit they deserve for a
TREMENDOUS effort to improve this server base and stop whining
about some weekend-warrior hacking based on running the code
through standad optimizers and seeing where all the looping takes place.
BTW: Did you know that standard Apache modules 'break' when running
with the new Red Hat Linux/Mandrake Apache-AdvancedExtranet
Server that has the SGI patches applied?
RCI just released a new version of mod_gzip ( 1.3.17.2a ) which follows
the Apache module design to a tee and the ONLY version of Apache
it will not run with is the Mandrake Apache rewrite from Red Hat
Linux with the SGI patches applied.
That ought to tell you something about these patches that people
think are so 'wonderful'. They break things that aren't supposed to break.
It looks like the decision to suspend review of the patches until
it can be done properly was the right thing to do.
Yours...
Kevin Kiley
CTO, Remote Communications, Inc.
http://www.RemoteCommunications.com/
http://www.RemoteCommunications.com/rctpd/ - Free IETF Encoding Server
http://www.RemoteCommunications.com/apache/ab/ - Free Enhanced ApacheBench
http://www.RemoteCommunications.com/apache/mod_gzip/ - Free Content
Acceleration module for the Apache Web Server
