SuSe and Compaq

[Paul_Melson@keykertusa.com]

>The second point, is RedHat is in the business of selling services not
>software. They have (IIRC) a 100 million dollar contract with IBM to
>develop GCC for the PPC processors. (IE the Apple processors, Apple is
>developing a Altivec for GCC which should be released shortly.)
>They have various other large contracts for porting from corporations as
>service, and support.  Im not saying they don't make any money on
>their distribution, but it is not their main income. It just forms
>the basis for their services, and support division.

     You're right, but I'd wager that IBM's interest in a compiler
     for Linux/PPC is because their RS/6000 systems are also based
     on PPC/RISC processors (the 604e and 750, respectively).  I'm
     all for that if it eventually means the availability of tools
     like `mksysb` and `smitty` for Linux/x86. :)

     At this point, really, RedHat does themselves a service by not
     charging outrageous sums of money for the distribution.  The
     cheaper it is, the more people will use it.  The more people
     that use it, the more companies interested in supporting it or
     porting their products to it.  It's a very similar strategy to
     what Microsoft used in the European and Asian markets 8-10 years
     ago.  (Please forgive me for using that particular comparison,
     but it is relevant.)


>Another point, is Linux is not going to _replace_ AIX, TruUnix64, etc.
>anytime soon, frankly it is not good enough at this point and the second
>reason is simply accountability it might work fine but if it doesnt
>the accountants want to blame someone.

     IMHO, it's not as much of an issue of being good enough, as
     it is simply being able to get support from your application
     vendor/author.  Typically, companies don't say, "We need a
     UNIX server to do DNS, FTP, and maybe a basic web server -
     let's spend $40K+ on a Sun E-class."  And you *DO* see Linux
     and/or *BSD boxes in those roles, and they really are good
     enough.  Where Linux has a long way to go is in terms of
     getting third-party applications that companies need.  This
     is where AIX, Solaris, HP/UX, etc. have a huge head start
     (and that's really all it is - they've just been around longer).


>The last point, is simply Linux IS doing really well in the embedded
>market. I found one Linux Box I would give my parents, and that is a TiVo
>(its even PPC!). The parts the keep Linux off the desktop are still the
>lack of good applications for non-programmers and the gui to match it.
>Which isnt helping any commercial dist.

     I am of the opinion that several of the GNU window managers
     available for XFree86 are far superior to CDE or OpenWindows,
     (or Win32 or MacOS) and that it won't be difficult to get
     people to see that in the open market.  But until the major
     graphics, engineering, and design applications make their way
     to Linux, there will be no need for people to use any of the
     window managers for Linux anyway.

     Anyway, that's my $0.02.

PaulM