O.K., I'm sorry. I would like to explain myself though. No, "I" don't
have the problem I state in my reply to Alan. I do think that there is
a large percentage of "Linux administrators" out there that either
don't have the time or skill to patch the existing stable kernel with
the new RAID 0.90 patches -- especially when they are not released on a
regular basis for each new stable kernel. Therefore, when I said "we
won't get the RAID code" and "I guess I can look forward to" I was not
referring to me in particular. I could have used better terminology or
expressed myself better.
I was trying to show my frustration or failure to understand why the
new RAID code is not included. I still do not believe that it can be
properly termed "developmental" or "experimental" anymore, and hence
there should be no objection for including it in a stable kernel
release.
M Carling, in another email you state:
"Right now, most investment banks on Wall Street are Solaris shops. Even
with Solaris, they pretty much all roll their own distributions--based on
Sun's distribution of course. They standardize on a set of patches, and
gnu-ify the tools, as well as adding their own libraries. If Linux gets
adopted on Wall Street, I expect it will be the same: They will take a
distribution and then heavily modify it and forever diverge from it."
So I again fail to understand what the issue is for putting the new
RAID code in if these Wall Street guys are going to tear it apart and
create their own "distribution" or "custom kernel" anyway. If they are
going to do that, then they surely would be capable of taking a 2.2.12
kernel and patching from there on without the RAID code, which I doubt
would be in their best interest anyway.
Also, I think we need to take a closer look at how kernel releases are
actually being utilized. If we could have some accurate statistics I
truely believe that we would find a small percentage of users that are
directly using kernel releases. Most users are probably using releases
from their distribution, which pick and choose which features they want
to include in their kernel anyway. If we want to think about renaming
kernel releases possibly "development" "stable" and "production" would
be more appropriate, with individual distributions making the decision
on what is "production" quality (as they seem to do now anyway).
Again, Sorry if I sounded a bit harsh or offensive in the previous
posts. I was not having a very good day and it was inappropriate.
fwr
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/