Re: No Microsoft Breakup

On Monday 10 September 2001 04:28 pm, Jim Damato wrote:
> My glib message about finding a new term to replace "operating system" was
> a reply to the discussion YOU and Leif were having, where YOU explained
> what YOU (and your dad) thought an operating system was, not anything said
> in the courts.

Leif only jumped in at all because he wanted to argue a point in the
discussion between you and I. After watching you and I discuss the point, he
sent a post asking, "Is OS/400 and OS?"

My discussion with you has more to do with statements you made. Like in your
message of 6 Sept 2001 at 18:45 where you said, "If they make it a part of
the OS, it's a part of the OS. Period." or your post of 7 Sep 2001 at 11:20
where you said, "...there are OSs of varying thickness...".

I mention that because you seem to be trying to say you have nothing to do
with the discussion about OSs. (refer to quote above)

But what difference does it make? If you don't want to talk about OSs, why
did you send this post? This is actual confusion on my part. (I note this
because of other comments in your post).

> The past nine or ten messages on this thread had been yours and Leif's
> opinions.  It's confusing when you come back with something about the
> definition used by the court.  I think "Huh? what court definition?  I
> didn't respond to anything about the court -- Chris hasn't even brought up
> anything about the court using anyone's definition of an OS.  How could he
> think I'm responding to something that wasn't even said?"  Try to think
> from my point of view.  For me, this came way out of left field.

Okay, the quote from my message commenting about the court needing a
definition.

        "I didn't make up the definition of OS. In fact, it wouldn't be
relevant at all except that it does make a difference if a court is making a
decision based on that definition. The court found that there really was a
definition and it wasn't what the vendor said it was."
(note I corrected the error in the original post where I forgot to put the
n't on the end of wouldn't)

Obviously I wasn't very clear about what I was saying here. Let me clarify:
        I don't care what you think an operating system is. The definition of an
operating system wouldn't be relevant at all in this discussion if it were
not for the fact that the court needed to define it. After all, the case is
about a company that is said to have a monopoly in the desktop operating
system market. So, knowing what an operating system is becomes important.
        I did not make up the definition of "operating system." I have posted 
before
and I believe I mentioned in my email direct to you that you can just look it
up and I am using the Encyclopedia of Computer Science as a reference as
well. I do not think this work is "the last word" in computing or anything.

        I will add this, IE is not compiled as part of Windows' kernel. It is
compiled seperately and runs using both unique and shared dlls. The operating
system is functional without IE and its unique dlls. IE also runs on the Mac.
To me, all these things are clear indications it is not part of the OS.
        Your statements that you feel it should become the default interface 
for the
OS seems to indicate that you also feel the GUI is not part of the OS. I
agree. I agree that it would be great if we as users could opt to use a
browser (IE or one of the others) to be that user interface.

> I followed you down this "what is an operating system?" road, but now I
> think that it's beside the point.  Even if Microsoft really brought the

Then we agree.

> code for IE down to the machine layer and bound it to the Windows OS (under
> anyone's definition) I'm guessing that you and the courts would find it to
> be a questionable, unfair practice, and I'd find it to be a promising
> technical and competitive move for the operating environment and the
> company.

Well, the courts don't really ask me much. ;-) Personally, I would think it
would be a dangerous and unnecessary change. After all, compiling that much
high level function into the kernel would tend to increase the instability of
the OS as well as introducing many, many possible security holes. Since the
browser function is so high level it will no doubt require changes on a
regular basis, and the kernal should not be changed often.

> Where I was coming from on this whole thread was that we see other examples
> of integrated products, or parts of operating systems, or bundled
> middlewear or whatever.  If bundling/integrating the browser is a
> monopolistic practice then I would have thought that bundling UDB with
> OS/400 is a monopolistic practice.  You've said (I hope I'm paraphrasing
> correctly) that bundling the browser with Windows is monopolistic because
> there are no other system alternatives.  To me, that's saying that there's
> already a monopoly, so bundling the browser extends or exploits the
> monopoly.  I wonder if we agree that bundling/integrating by itself is not
> cause for an anti-trust suit.

Well, we were quite in agreement until the last sentence. I may just not
understand what you meant. Of course there would not be any problem at all if
Microsoft were not a monopoly. But even though they are a monopoly, it is not
illegal for them to bundle things. But the reason for the lawsuit is the fact
that they bundled this specifically with the intent of driving another
company out of business using their monopoly control of the market.

Bundling is absolutely legal, for Microsoft or anyone else. Anti-trust suits
are only necessary when some behaviour of a monopoly company is hurting the
consumer. In this case, by eliminating innovation and choice in the market.

> If Linux or Mac/OS or OS/2 had thrived as viable, competitive alternatives
> to Windows, would it still be wrong for Microsoft to impose their standard,
> integrated/bundled browser on their platform?

Above.

> Given the current situation of Microsoft dominance, what will happen
> when/if Microsoft SQL is integrated into the Windows platforms?  If it
> severely impacts Oracle's market presence, is there cause for an anti-trust
> suit?

Microsoft doesn't dominate the server market, so I don't think their monopoly
problems extend there. In this case, while it may just be that some people
don't install NT because they prefer to use Oracle. So again, bundle away.
But if Microsoft does become a monopoly and then chooses to eliminate Oracle
by bundling there would be a problem.

> Hypothetically,  what if IBM decides to embed source/object cross-reference
> into SEU and PDM and provide it for "free" as a part of that set of
> licensed products?  As a result, competitive products from Hawkeye, ASC,
> etc. become moot.  Is there cause for an anti-trust suit?  Does it matter
> if IBM did this to logically extend the licensed product, or if they did it
> because they wanted ASC and Hawkeye out of their way in the middleware
> market?
>
> What do you think?

Again, IBM does not have a monopoly. Their product is a server. Users can
move to other platforms. I do think that they would "get away with it" but it
would also be cutting their own throat. The fewer options available for users
on the iSeries, the less attractive it is, and these days the server market
is pretty competitive.

If, on the other hand, they were a monopoly, this would be a problem.
Existing software bundling would also have to be re-examined.

Keep in mind that this is an example of the anti-trust laws working. When IBM
signed their consent decree they were a midrange monopoly. While I am a big
fan of the IBM midrange (A big, big fan ;-) ) I do think that users have a
greater variety of choices today than they would have. I also think the
AS/400 has gained a lot because IBM has needed to keep it ahead of the
competition.

> -Jim
>
> James P. Damato
> Manager - Technical Administration
> Dollar General Corporation
> <mailto:jdamato@dollargeneral.com>

--
Chris Rehm
javadisciple@earthlink.net
If you believe that the best technology wins the
marketplace, you haven't been paying attention.