Re: Rejecting Korporate Linux
From: student (student_at_linuxbox.anom.com)
Date: Sun, 12 Oct 2003 06:36:21 GMT
On Sat, 11 Oct 2003 18:08:41 GMT, Burgess Meredith <email@example.com> wrote:
>I suffered through the interminal drivel, that the Dynamic Dummies, Alan
>Connor-Man, and the Boy Blunder had to say,
>> It will allow the participation of cultures that are actually different from
>> this one, rather than just our de facto colonies, to the great benefit of
>> all concerned.
>There will be two types of corporate linux.
>M$ style - bought from redHat and tuned by 'consultants'
>Good for system admins.
>Larger corporations will soon figure that they
>can higher their own kernel and system developers
>to make their GM, Boeing, or Wal*Mart Linux (internal)
>do exactly what they want it to do.
Type II is Just Not Going To Happen. Having a high-dollar stable of kernel
developers to do OS research is the last thing the typical Fortune 500
Soap Manufacturer is going to spring for. The reasons are too numberous
to go into at the moment.
As for a distribution, the systems engineers tend to be happy with it
because they can submit a laundry list of the politically correct hardware
de jour and get a simple yes ("we support it") or no ("wait for next
release") from the distribution owner. The suits like it because they see
hope of a standards-based architecture, containable and accountable
support (through support contracts), and a because of a belief that a
distribution is simply going to be more complete and better tested than
anything they can cook up and sustain in-house.
Bottom line to corporate computing is a stable platform with decent
performance, which supports all software and hardware requirements.
And responsive support from the vendor. The future of Linux distributions
depends critically on how well they perform ( and in some cases whether
they can improve their performance) in meeting _all_ of these requirements.