[Am-info] Re: adopting alternative OSes

[mike]

and, for what it's worth, knowing the kernel wouldn't help you do=
 
anything else with the system (unless you want to write=
 drivers)... 
it's Mach surrounded by a BSD layer and a Cocoa layer.  Cocoa is=
 based 
on NextSTEP, so if you're familiar with THAT, it will help you=
 out.  
Since I didn't see that listed in your (rather obnoxious and 
self-righteous) list of OS's you're familiar with, you might have=
 
slightly more work ahead of you.  That doesn't even begin to=
 address 
using the UI, knowing about how system preferences are stored,=
 the 
NetInfo database, etc., etc., etc.  My point is that a kernel=
 does not 
an OS make.

I might sound self righteous, but I assure you, that with the=
 exceptions of all the times I have been wrong, I am always=
 right.

I was speaking about Operating system API's and how they are able=
 to speak with the operating system from an applications point of=
 view.... however....  The kernel for the most part is dealing=
 simply with scheduling and i/o.  (unless you are Microsoft and=
 it becomes a toilet to throw everything in....)

Most of my experience is with the post 1980 era of application=
 programming.  However I also worked closely with another chap=
 who specialized in device drivers, so many days and nights were=
 spent helping each other with hurdles.  These were in almost all=
 the aforementioned operating systems (except I think to my=
 knowledge Gerry did not do any work in QNX or in MUFO) Most of=
 the work in the 90's was in OS/2.  

Also, speaking from experience here, lots of people say that they=
 can 
do a better job at marketing something than another company with=
 
billions or even just millions of dollars for their product of=
 choice.  
I've worked with groups that have said, if we only had the=
 marketing 
dollars.  Often, it doesn't matter... the product itself can get=
 a 
temporary boost from improved marketing, but in the end it's=
 about the 
product or some social/business/non-technical issue.

Trained monkeys could have done a better job....  The problems=
 were obvious.  Yet no-one had the balls to answer any of them. =
 they feared for the middle management jobs they were holding. =
 The PC Hardware division felt nothing but distain for the=
 software division. I wished Akers would have split the company=
 up.  At the time there were rumours I would be working for the=
 software company. I was hoping....

WARP had the marketing dollars.  What it didn't have was OEM=
 deals.  I 
would suggest that some manufacturers might have been nervous=
 having 
their operating system (which is as critical to the computer as=
 it's 
CPU, maybe more so from a consumer or business user's standpoint)=
 come 
from their largest competitor in the hardware market.

and the separate software company would not have had that=
 problem.

And (putting my flame retardant suit on), Warp wasn't that=
 special.  It 
was, from a user's perspective, the same damn thing as Windows. =
 I know 
I spent most of my time in games, WordPerfect, or connected to a=
 
mini-computer in college, so I lived in DOS and was perfectly=
 happy.  
What made people choose OS's is the same thing that makes people=
 choose 
today... Word ran best in windows and it came with their=
 computer.   
That was the reason why OS/2 never even came up on the radar=
 screen on 
my campus, and in my private life.

You lived a sheltered life.....Many high-schools here in=
 Southwest British Columbia switched entirely over to Warp 3.  At=
 that time the only thing Microsoft had that could be called an=
 operating system was NT 3.1.

OS/2 1.2 was essentially aiming for the same market as windows=
 3.0.  But it preceded Windows3 by 2 years.....
Word was available for OS/2 1.2, Wordperfect, Lotus 123, Excell,=
 Ventura Publisher, Pagemaker, Micrographics, Describe, And many=
 others.  In fact I can't think of a gui based application that=
 was not available for OS/2.  (note this was 1990), and the=
 operating system I used then was called....  MS OS-2.  The=
 device drivers we were writing then (Actually my good friend=
 Gerry) were Video drivers that operated in a bi-modal fashion,=
 that is to say they were written to operate in real mode DOS as=
 well as protected mode, concurrently. Later when OS/2 2.0 came=
 out the focus of course switched to writing video drivers for=
 the 32 bit operating system. Clients included SMOS, S3, Epson,=
 Trident, Chips & Technology.  Later projects included SCSI=
 drivers, and a tape backup solution for OS/2.  (Fastback)

So please understand I do recognize an operating system as a=
 collection of API's, But I also understand the Kernels position=
 within that operating system.  Very little of the programming=
 work was done in what many here would call the GUI.


On Sunday, April 27, 2003, at 12:31 PM, Mitch Stone wrote:

> I don't question for a moment doubt IBM's mismanagement of Warp=
 -- in 
> fact from what I have heard (mainly on this list), this seems=
 to be 
> one of the great missed opportunities in recent technology=
 history. My 
> point being still that it takes much more than a few millions=
 and a 
> good product to crack the Microsoft hegemony in the OEM market=
 (ask 
> Jean Louis Gasse). You can't say that IBM did not try at all --=
 they 
> had the product, the PC manufacturing arm, the cash, and most 
> importantly, the brand name. Their failure to put all of these=
 
> elements together suggests that the problem of succeeding in=
 this 
> market is greater then you may assume. Rightly or wrongly, IBM=
 
> calculated that they were not likely to succeed.
>
> MacOSX is not a "derivative" of the Mach kernal, it uses the=
 Mach 
> kernal. A point of order, for whatever it's worth...
>
> On Saturday, April 26, 2003, at 10:41 PM, mike wrote:
>
>> Well then.  Perhaps I should have made the point clearer for=
 you.  I 
>> think I could have done a thousand times better at 
>> marketing/supporting Warp than IBM ever did..   And unlike=
 many who 
>> did nothing, I was involved with Warp at the time with IBM.  I=
 did 
>> what I could but I was limited by management.  If we would=
 have been 
>> let loose to do what we wanted in 1992 and 1993/94, I am sure=
 Warp 
>> would have passed the 25% market share....  and at that point=
 it 
>> would have surpassed critical mass. Akers wanted to split the=
 company 
>> up and let the software division go on its own, much like the=
 Lexmark 
>> experience.  However Gerstner came in and put a stop to it. =
 In 
>> hindsight that was a big mistake on Gerstner's part.
>>
>> But yes I think if I had a few millions and the Warp=
 development 
>> team, I could put something together.....  At least as good as=
 the 
>> Linux stuff that is around.  And at least as good as the Mac 
>> alternative.  and without the burdon of IBM I think it could=
 be sold 
>> to OEM suppliers.
>>
>> But of course you are more knowledgeable about operating=
 systems than 
>> I am....  I am limited to knowledge of Dos, CPM, CPM86, PCMos,=
 MUFO, 
>> Windows, OS/2, BEos, SCO Unix, And Linux.  I am not too=
 familiar with 
>> the MAC's system 10.  But since it is derivitive of the Mach=
 kernal, 
>> I suspect I could become familiar with it in a few weeks.
>>
>> Anyhow it is fruitless to be skeptical of another persons=
 dreams.   
>> Unless you are prepared to change it to a reality and write=
 the 
>> cheque....
>
>   -----------
>   Mitch Stone
>   mitch@accidentalexpert.com
>
>   In America you can go on the air and kid the politicians,=
 and
>   the politicians can go on the air and kid the people.
>   -- Groucho Marx
>
> _______________________________________________
> Am-info mailing list
> Am-info@lists.essential.org
> http://lists.essential.org/mailman/listinfo/am-info
>
------ sujal shah ------ sujal@sujal.net -----

                  http://www.sujal.net

_______________________________________________
Am-info mailing list
Am-info@lists.essential.org
http://lists.essential.org/mailman/listinfo/am-info