NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/os/os2/advocacy/11551

Internet Message Format | 1993-01-04 | 6.4 KB

Path: sparky!uunet!ogicse!sequent!muncher.sequent.com!furballs
From: [e-mail] (Paul Penrod)
Newsgroups: comp.os.os2.advocacy
Subject: Re: FCC will proclaim Microsoft is run by Communists! : )
Message-ID: <[e-mail]>
Date: 4 Jan 93 07:51:38 GMT
Article-I.D.: sequent.1993Jan4.075138.13422
References: <[e-mail]> <[e-mail]> <[e-mail]>
Sender: [e-mail] (usenet )
Organization: Sequent Computer Systems, Inc.
Lines: 120
Nntp-Posting-Host: crg8.sequent.com

In article <[e-mail]> [e-mail] (Jeff Sicherman) writes:
>In article <[e-mail]> [e-mail] (Jay Maynard) writes:
>>In article <[e-mail]> [e-mail] (Phil Lafornara) writes:
>
>   [ deleted ]
>>
>>>    You bring up functional specifications below - the fact is,
>>>MS-DOS doesn't have a rigid one that would allow a clone
>>>to be written.
>>
>>By default, then, MS-DOS' published API qualifies, especially since MS will
>>tell you every time you turn around that that's all a programmer should
>>use...never mind that MS breaks that rule regularly...
>
>  Although I don't know if it has *legal* significance, I would expect that
>there are areas of the API that MS could reasonably violate or not document.
>In particular, with the utility programs and drivers that are released when
>new versions of DOS itself are released. One could argue that these are
>really part of DOS whether they reside within the same code file or not and
>so they could share 'special' and intimate knowledge of its internals.
>
>   The question then arises whether if WINDOWS makes use of such knowledge,
>whether it is really an extension of DOS both in a technical and a *legal*
>sense (since legal issues are at the root of this whole discussion) or it
>is really a separate product as well as being an 'environment'.
>
>>
>>>Digital Research must have done a lot of
>>>reverse engineering, with quite a bit of guesswork.  They
>>>got very close - they didn't quite make 100%.
>>
>>They made 100% to MS-DOS' published API. Since MS says that that's all an
>>application developer can depend on, that should be enough.
>
>  As I understand it, the problem was actually that Windows circumvented
>as established standard (DPMI) in some way - or short circuited DOS interfaces
>with it - with no obvious gain (to outsiders, anyway) - than DR-DOS would no
>longer work properly with it.
>
>>>I'm sure that
>>>as further incompatibilities are revealed, they will fix them
>>>as well.
>>
>>That should read, "as further MS violations of their own rules are revealed,
>>DR will add compatible unpublished interfaces as well."
>
>  I think that calling them violation of their own rules is a little strong.
>MS has never, to my knowledge, and often quite the contrary, made any promises
>about how DOS works internally or how reliable certain 'discovered', rather
>than documented, interfaces may be. I think the issues of fair dealing  are
>a little more subtle than that and depend upon different expectations when
>an entity effectively controls part or all of a particular market.
>
>>>But I don't expect to see "DR-DOS" applications
>>>coming out any time soon, and until then, the people who
>>>buy DR-DOS to run their MS-DOS apps are basically taking the
>>>gamble that it's going to work.  It's a good bet, but it
>>>isn't 100%.
>>
>>If MS followed the rules, it WOULD be 100%. MS has the power to break DR-DOS
>>at will, and all external indications are that they have done exactly that.
>
>    I personally don't see how they could break it for long, given the nature
>of todays s/w inplementation and distribution methods, without causing an
>awful lot of problems for themselevs and for many, if not most, DOS-based
>applications. I think it is indicative of that that they narrowly chose to
>'break' it by and with a product that is in many respects and O/S itself and
>one which people expect to work differently and perhaps be incompatible.
>
>   With the release of Windows for Workgroups and the folding of DR into
>Novell, this all is now really part of a bigger battle for the Network OS
>of the future.
>
>
>--

I agree that it's a battle of egos and pocketbooks for the next
killer NOS that will dominate the market place; but I seriously
doubt that much will come of it other than more attractive
packaging of the sand in the can, with your choice of Provo or
Redmond to send the check to...

Now as to the issue of DOS. As someone so aptly put it several
months ago, DOS is nothing more than a boot sector virus. I would
be a bit more polite and recategorize it as a 16 bit program loader
with features. What I think needs to be taken into account here, is
the amount of effort required to emulate or modify DOS. Having done
both to some degree over the years I can honestly say that any good
programmer with enough moxy on the ball should not have any
problems. DOS aint that complicated folks. IF Novell can't get it
right first pass, then that's their fault, not Microsoft's or
IBM's. Frankly I am amazed these guys even know how to use a
disassembler and walk a thread sometimes, from some of the reports
I get from engineers who have to deal with them.

So the changes come out a bit late. BFD. IF I had a product that a
competetor was successfully emulating, I would not readily hand
over the keys to the castle either; but I would not stop them.
Instead the smart money is on licensing the technology for an
established period of time and then throw the doors open and allow
feedback from the competitors supporting the standard if you will.
This creates a win-win situation for all involved. The product
becomes standardized, the competition markets successfully as well
and gives you the ability to grow the product, while the consumer
benefits. What the major players (Novell, IBM, Lotus, Microsoft,
etc.) fail to realize, is that their product is not a necessity of
life and that ego and greed only bring about their downfall,
because in their books, the bottom line is more important than the
customer. IBM is relearning this lesson a bit late. The rest of the
aforementioned tend to see only the wallet, and not the entire
customer.

JMHO...


--
--------------------------------------------------------------------
            Bureaucracy: noun, plural - Bureaucracies.
         The process of turning energy into solid waste.
---------------------------------------------------------------------