🚀 go-pugleaf

In <3330c3a3.0@atheria.europa.com>, bknotts@europa.com (Brian Knotts) writes:
>In <5gnuil$1hb4$1@news-s01.ny.us.ibm.net>, baxte17@ibm.net  (Kevin J Baxter) writes:
>
>>If we want IBM to better support OS/2, then PSP must 
>>sell its own hardware (much like AIX is sold with RS/6000s, OS/400s
>>is sold with AS/400 and OS/390 is sold with S/390s).
>
>Well, that's what the whole Power Personal thing was initially,
>but, as I understand it, it was originally part of RS/6000, then
>PSP got hold of it.
>
>They more or less abandoned it before it had a chance to sink or
>swim, because of the whining attitude of PSP (as you note, *not*
>the major money makers within IBM). I say, to hell with PSP; IBM
>should revive the Power Personal for Warp 5.0.
>
>The original concept of a portable, microkernel-based OS/2 (or OS/2-
>like operating system) ON ITS OWN NON-INTEL HARDWARE PLATFORM, was
>an excellent one, IMO.

I won't give any credit to the Power Personal Series as a kind of
pro-OS/2 exclusive platform.  Far from it, the PReP platform that the 
Power Personal Series was based on was conceived as a fair multi-OS 
venture that included PPC NT, PPC Solaris, PPC Netware, AIX and even
MacOS. The great irony was that Power Personals could not sell because 
it cannot run *MacOS*, which had evolved to require so many
proprietary hardware dependencies, that MacOS cannot run on a 
common multi OS platform. (For the same reason the MacOS on Intel
project got skunked in the first half of the nineties.)  Unlike the 
Intel platform, where OS/2 can use Windows and DOS software 
compatibility as a take off point, neither PPC OS/2 nor Power Personal 
is compatible to the base of Mac software, whose actual numbers and 
their quality is far more impressive than many would believe.  One can 
blame that IBM itself, which shipped more than 2.4 million PowerPC 
processors last year, was the last company who was able to control and 
exploit their own technology.  Apple literally ran away with control of 
the PowerPC market.  Right now, no one can sell a PowerPC system (other 
than AIX) that can't run MacOS, and the two has become intertwined.
 
Furthermore, the more OS/2 PPC remains in hiatus, the more IBM delays in
coming out with their own advanced *desktop* PowerPC OS (AIX and OS/400
are basically back end server OSes), the more IBM's control of that 
market withers away, as OpenStep and BeOS---both of whom has a take no 
prisoners approach to technology and object orientation---competes for 
the predomination of the PowerPC platform.  It's one thing when you 
take OS/2 to argue against Windows 95 and NT on technical merits, but 
it's another ballgame when it's against OpenStep and BeOS.  Furthermore,
Rhapsody is an advanced OS that is Unix in the core---it's one thing for
AIX to compete against PPC NT, but again, it's another ballgame if it 
takes on Rhapsody.  The fact that Apple has dropped its AIX support show
that it has motives to that effect.

Lastly PowerPC is practically an open standard.  It appears anyone can 
make it, just as Exponential (30% owned by Apple) has shown that IBM 
does not monopolize the PowerPC.

>
>It would have allowed OS/2 to have its own, unique marketplace, where
>it wouldn't be in direct competition with Windows (yes, I know NT
>was there on PPC at the time, but NT was hardly used back in '95;
>heck, it's still not used that heavily today).
>

For OS/2 to have its unique marketplace, it must have its dedicated 
hardware platform.  If PPC OS/2 is released next year, hypothetically 
speaking, on the PPCP Mac platform, it still has to compete against 
Rhapsody and BeOS, both like PPC OS/2, highly advanced microkernel based
OSes with extensive object orientation in their architecture.  (PPC 
OS/2, Rhapsody and BeOS are actually distant cousins---they all have 
roots in MachOS.)  The marketing on Rhapsody will undoubtedly 
aggressive, and so will BeOS be.  For PPC OS/2 to succeed, first it must
show that it has a superior kernel over that of Rhapsody and 
BeOS---something which it might really have, if the rumors about 
IBM Microkernel v2.0 is true (1024 CPU SMP support, will support more 
threads and loads, real time support.)  Then second, IBM must craft some
cheap hardware that can outprice PPCP Mac clones and won't run Rhapsody 
or Be.  Since Mac clones are easily hitting the $1200 mark now, the new 
IBM PPC OS/2 hardware has to be cheaper than that.  Third, it must
show it has better Java support and performance, better than any OS 
from both Apple and Microsoft, and Java helps it integrate better with 
other IBM platforms.  Lastly,  it's a machine that can be sold easily to
its big corporate customers.

In conclusion, an IBM PowerPC Network Computer that runs the Java NC 
version of PPC OS/2.  That's for the bottom end.

As RS/6000s gets squeezed by Intel NT in one side, AIX has the 
capability to run on low cost PPCP platforms.  Unfortunately, AIX is too
expensive and cumbersome for the ordinary user.  Thus IBM needs a lower 
cost desktop OS to protect its stake in the PowerPC market.  Right now, 
the PowerPC is on the comeback and may still actually threaten 
Intel---Mac clone vendors have raised the MacOS share of the overall 
market from over 7% in October 1996 to over 11% in January 1997, 
accounting for over 61% in new retail growth.  That's 1% per -month-
folks.  I find even greater irony that IBM will be making Apple's 
180-240MHz 2400 Powerbook subnotebook---IBM Thinkpads with an Apple 
label.  Not to mention that it's been sublicensing MacOS to 
potential Taiwanese Mac clone makers, including Acer and Tatung.  If IBM
does not put a stake on the PowerPC market with a PPC OS/2, it will
become a Mac clone vendor in the PowerPC market.

Rgds,

Chris

Famous People on the Day Windows 95 is Launched---
Neil Armstrong---"One Small Fortune for Bill Gates, 
One Giant Leap backward for Mankind."
President Roosevelt---"This date shall live in infamy."
*** crobato@kuentos.guam.net *** TKS for the Contributions.