|
|
|
| Authored by: foulis on Friday, May 10 2013 @ 08:11 PM EDT |
<p
align=right><b>PLAINTIFF'S<br>EXHIBIT</b><br><u>
;4029</u><br>Comes v. Microsoft</p>
<b>From:</b> Richard Fade<br>
<b>Sent:</b> Friday, January 15, 1999 12:18 PM<br>
<b>To:</b> Joachim Kempin; Allen Wilcox (LCA)<br>
<b>Subject:</b> FW: Compaq: Need Input</p>
<blockquote><b>ATTORNEY CLIENT
COMMUNICATION</b></blockquote>
Joachim I cannot add much to this thread as Rod's comments on his performance on
promoting our properties (how many) are about what was verbally agreed to in the
discussions. I also had a conversation with Carl that David Heiner had at some
point reviewed the language in the license agreement. Allen may want to apprise
you of that.<br>
richard</p>
----Original Message----<br>
<b>From: Laura Jennings</b><br>
<b>Sent:</b> Thursday, January 14, 1999 1:54 PM<br>
<b>To:</b> Paul Maritz; Bill Gates; Steve Ballmer; Joachim Kempin;
Carl Sittig; Richard Fade; Allen Wilcox (LCA)<br>
<b>Cc:</b> Greg Maffei; Carol Whitaker (Davies LCA)<br>
<b>Subject:</b> RE: Compaq: Need Input</p>
Thanks. These two reasons are extremely helpful. We'll stay the hard-core
approach on #1 and #2 and leave the old digital stuff out of it. I expect if we
push hard enough tonight this will escalate by Monday if not before.</p>
----Original Message----<br>
<b>From:</b> Paul Maritz<br>
<b>Sent:</b> Thursday, January 14, 1999 12:50 PM<br>
<b>To:</b> Bill Gates; Laura Jennings; Steve Ballmer; Joachim
Kempin; Carl Sittig; Richard Fade; Allen Wilcox (LCA)<br>
<b>Cc:</b> Greg Maffei; Carol Whitaker (Davies LCA)<br>
<b>Subject:</b> RE: Compaq: Need Input</p>
Only thing that I ask is that we not roll in the Digital payments into the
"high end" discussion - it can only result in them asking for more in
situation where we have enough at stake already. My view is in fact that we
should keep them out of both discussions, make deals on the merits, stay hard on
the point that this is money owed to us no matter what, and later as an
accounting measure we can agree to trade one off before the other.</p>
----Original Message----<br>
<b>From: Bill Gates</b><br>
<b>Sent:</b> Thursday, January 14, 1999 12:46 PM<br>
<b>To:</b> Laura Jennings; Steve Ballmer; Joachim Kempin; Carl
Sittig; Paul Maritz; Richard Fade; Allen Wilcox (LCA)<br>
<b>Cc:</b> Greg Maffei; Carol Whitaker (Davies LCA)<br>
<b>Subject:</b> RE: Compaq: Need Input</p>
Attorney client</p>
This all sounds like the Rod gets rich plan to me.</p>
I think Compaq is making a big mistake selling the bounty they get from their
machines to an outside company. It will make them non-competitive in the
consumer PC space.</p>
I see no reason not to be hard core about the first two points outlined
here.</p>
I have never heard about these Digital back payments. It may be that those
should be kept out of this negotiation and folded into the negotiation Paul is
having with them on high end things. We have some things they owe us related to
Tandem as well. The total number of all these things is really large.</p>
My biggest beef relates to point 1.<br>
<p
align=right><b>MS-PCA-2606018</b><br>HIGHLY<br>CONFID
ENTIAL</p>
<hr>
<br>
We agreed to let them have some flexibility for a year to see how they fulfilled
point #1. We were clear that if point #1 was not fufulled[sic] in a big big way
that we would stop letting them do anything. Unfortunately our mistake in
letting them do something was passed on to other companies losing massive value
for us and creating a precedent we will be struggling to back out of for a long
time.</p>
My view is that they should use our branded communications services and do some
other things just to deal with point #1. Not unbranded.</p>
We should be hard core on this one.</p>
Point #2 is just a money point. They owe us and they have to pay.</p>
'Rose is going to testify in the antitrust lawsuit but the positions we are
taking here are very reasonable.</p>
Lets now give in on #1 without getting something large. If they have to delay
their offering that is too bad. We WILL be locking down the Windows agreement
more over time and with new products.</p>
If I need to call Pfeifer on this I will.</p>
----Original Message----<br>
<b>From:</b> Laura Jennings<br>
<b>Sent:</b> Thursday, January 14, 1999 10:05 AM<br>
<b>To:</b> Bill Gates; Steve Ballmer; Joachim Kempin; Carl Sittig;
Paul Maritz; Richard Fade; Allen Wilcox (LCA)<br>
<b>Cc:</b> Greg Maffei; Carol Whitaker (Davies LCA)<br>
<b>Subject:</b> Compaq: Need Input</p>
<b>ATTORNEY CLIENT COMMUNICATION</b></p>
<u>Summary:</u> Compaq plans to file to take AltaVista++ public next
week, details of their plans included below. Greg and I are actively engaged in
dialog with Compaq on two fronts: 1. a commercial agreement for this new company
to use Hotmail, Messenger, Passport as their communications infrastructure and
2. an equity investment of 5-10% in this new company. There are three
outstanding issues we are familiar with regarding Compaq which could potentially
influence our negotiating position, and we'd like input from those of you more
familiar with these previous deals on how to use them, if at all, in these
negotiations. -- thanks.</p>
The new company will contain:<br>
<ul>
<li> AltaVista</li>
<li>Shopping.com, which they announced their intent to purchase last week
for $220 million</li>
<li>"Traffic and promotion" from a "major media
player", detail of which are sketchy at best, but Rod is currently holed up
in NYC completing these negotiations and promises more info to us
tomorrow.</li>
<li> Exclusive rights to all internet-related activities on Compaq
hardware - desktop real estate, keyboard links, etc. - for a period of 3 years.
This new company can decide to direct all traffic to altavista or to sell the
links/position to third parties.</li>
</ul>
Rod will run this new company. Compaq is going to keep a controlling interest,
and sell minority percentages to their major media partner and their
infrastructure provider, the latter of which we hope is us. (Rod told us Yahoo
will get the infrastructure deal if not us, and they are negotiating
simultaneously with the two of us).</p>
The three outstanding leverage points Greg and I know about with Compaq
are:</p>
1. "Prominent" Internet Partner contractual language from Compaq's
Windows license, per Carl Sittig:</p>
<blockquote> In the Windows negotiations last Spring, MS made a large
concession to Rod regarding ISP referral server provided that
"<tt>(1) MSCORP is promoted as COMPANY's prominent partner in the
internet and (2) MSCORP's www-based services (i.e., Expedia, CarPoint,
Sidewalk.com, MSNBC, Internet Gaming Zone, MSN Investor, etc.) are promoted (A)
in such ISP sign up wizard:(B) on COMPANY's Presario internet home page: and (C)
in other Presario internet activities."</tt> This was also a
discussion during a Compaq Exec review with BillG, Rod, Eckard,
et</blockquote>
<br>
Given that a big benefit that Alta Vista will have is the Compaq distribution
rights, if MS were to aggressively assert its rights with Compaq under this
contract, it would surely affect AV's valuation. Rod's</p>
<p align=right><b>MS-PCA
2606019</b><br>HIGHLY<br>CONFIDENTIAL</p>
<hr>
<br>
current strategy seems to be to ignore this aspect of our existing agreement.
What Greg and I need to know from folks who negotiated this or have subsequently
discussed it with Compaq execs is how strongly to interpret and assert our
rights under this agreement. Unless we hear from someone that the intent as
discussed during the negotiation phase was softer, we will push very
aggressively a stance which says MSN properties are already guaranteed MFN
status in the new AltaVista portal for as long as it serves as the Presario
default or is linked in any way to Presario-linked internet activities, and that
regardless of how the infrastructure commercial discussions go, Microsoft plans
to vigorously defend these rights.</p>
2. Search contract (I already own, included for completeness)</p>
AltaVista signed a one-year, $18 million distribution deal with MSN this summer.
Compaq has subsequently made it clear they want out of this deal. We have
indicated a willingness to work with Compaq to address their needs - we'll take
the payment in PCs, we'll take it in more aggressive cross-distribution than we
currently are already entitled to under the clause above, we'll change the
payment structure, add international, etc., - but we won't just let them out of
the contract. Compaq is currently in breach of the significant terms of the
existing search distribution contract, and failed to make the second payment of
$4.5 million in December. We have not yet taken them off of our site nor taken
legal action against them since we were engaged in dialog to both rectify their
failures and explore a potential new infrastructure deal, but if we were
prepared to do so, this could be a second strong card in our hand which might
make it more difficult for Altavista's public offering.</p>
3. Digital back payments</p>
This one we understand the least well. Greg believes that Digital owed us back
payments of $40 million at the time of the Compaq acquisition, and that this
matter is not yet settled. One option would be for us to agree to take this $40
million and the $18 million above in the form of equity in the new
company.</p>
In summary, Greg and I want to take the position in our calls this evening and
tomorrow that we are already in a very strong position on Compaq's internet
activities and that not resolving their existing outstanding issues with us (#1
and 2 above) will make the spin-off more difficult. We are willing to transfer
some of their outstanding obligations into equity in the new AltaVista company,
but if we find ourselves locked out of AV's plans, we intend to vigorously
pursue our rights per the first two contracts above, and they are better off
negotiating with us on these issues directly than trying to ignore them. This is
hard ball, they will likely escalate to Steve/Bill, but we believe now is the
time for us to stand firm.</p>
<p align=right><b>MS-PCA
2606020</b><br>HIGHLY<br>CONFIDENTIAL</p>
[ Reply to This | Parent | # ]
|
- Comes 4029 - Authored by: PJ on Friday, May 10 2013 @ 10:23 PM EDT
| |
| Authored by: Anonymous on Sunday, May 12 2013 @ 06:13 PM EDT |
http://groklawstatic.ibiblio.org/pdf/iowa/www.iowaconsumercase.org/011607/4000/PX04372_B.p
df
<p>
Plaintiffs' Exhibit 4372_B<br />
Comes v. Microsoft
</p>
<p>
<b>From:</b> Brad Silverberg<br />
<b>To:</b> Marianne Allison (w-maria)<br />
<b>Cc:</b> Bill Neukom (billn); Debra Vogt (debrav)<br />
<b>Subject:</b> FW: AARD drivel<br />
<b>Date:</b> Tuesday, August 03, 1993 10:59AM
</p>
<p>
what do you suggest we do? should we send to DDJ? Schulman? edits you'd make?
please advise.<br />
----------<br />
</p>
From: Bill Gates<br />
To: billn; bradsi<br />
Subject: RE: AARD drivel<br />
Date: Tuesday, August 03, 1993 9:12AM
</p>
<p>
This all seems fine to me. Has it been sent to Schulman?
</p>
<p><b>Privilege Material Redacted</b></p>
<p>
----------<br />
From: Brad Silverberg<br />
To: Bill Gates<br />
Subject: AARD drivel<br />
Date: Monday, August 02, 1993 1:15PM
</p>
<div style="margin-left: 1em">
<p>
From aaronr Mon Aug 2 11:35:34 1993<br />
X-MSMail-Message-ID: 15DF233B<br />
X-MSMail-Conversation-ID: 15DF233B<br />
X-MSMail-WiseRemark: Microsoft Mail -- 3.0.729<br />
From: Aaron Reynolds <aaronr@microsoft.com><br />
To: bradsi<br />
Date: Mon, 2 Aug 93 11:32:50 PDT<br />
Subject: AARD drivel<br />
Cc: aaronr
</p>
<p>
I read with some interest the "Examining the Windows AARD Detection
Code" article in the September issue of Dr. Dobb's Journal. I was
generally impressed with the technical aspects of the article, and
with
a few of the opinions expressed, however the opinions about the
purpose
of this code and the reasons for its existence were not correct. I
should know, I was one of the primary people involved with this code.
</p>
<p>
I preface my remarks with the following commentary. The rest of this
is
a complete discussion of the reasons for this codes existence, nothing
is left out, this is the complete story. You probably aren't going to
believe it though, so to some extent I am probably wasting my time
explaining it to you. In this age of sensationalist journalism, the
background of this code isn't "juicy" enough, so people will probably
use the simple tactic of dismissing it as untrue or incomplete. There
seems to be a curious assumption running around the world: "Microsoft
is a malevolent, Machiavellian organization which is always doing
secret evil things according to hidden agendas and you can't trust
anything they say." Not much I can do about it if you are pre-disposed
to disbelieve what I say.
</p>
<p>
Windows is tight coupled to the underlying MS-DOS operating system. It
relies on a number of very precise behavioral characteristics of
MS-DOS
which have nothing to do with the INT 21h API. I should know, I am the
person who designed most these characteristics in both MS-DOS and
windows, and spent a long time figuring out all of the subtle issues
relating to them. The reliance on these characteristics puts windows
into a very different class of program than any other MS-DOS
application. Because of this tight coupling, an MS-DOS "work a like"
must have exactly the proper behavior or all sorts of subtle and not
so
subtle problems will occur.
</p>
<p>
Microsoft does not test windows on anything other than Microsoft's
MS-DOS. We do not consider it our mission to test windows on MS-DOS
"work a likes". That is a problem for the MS-DOS "work a
like" people
to worry about. If they want to do all the work, and certify their
products for use with windows, fine, we are not going to do it for
them. This should not surprise anyone. Look at the market. How many
other MS-DOS program developers spend time testing on MS-DOS "work a
likes"? Testing is a very expensive and time consuming enterprise. It
is important not to waste time testing something when the possible
additional sales you might get won't even pay for the testing. Sorry
that this is not as "warm and fuzzy" as some people would like it to
be, staying in business is like that. If you are an MS-DOS "work a
like" and you are trying to catch up with a dominant market leader,
you
have to work very very hard and you are not being very smart if you
expect that market leader, who is your main competitor, to do a bunch
of work to help you out. You have to do that work yourself.
</p>
<p>
I know of 7 MS-DOS "work a likes". On all of these but one, windows
will not even start. On the one that it does manage to start on,
depending on which mode of windows you run, it has some other more
subtle problems. I am purposely vague here because I am in a difficult
position. If I name names, I or my company will probably get dragged
into court, so I will not name names or be more specific, sorry. "But
wait!!!!" You said you didn't test on MS-DOS "work a likes", how
do
you
know this? During the windows 3.10 betas we got a few bug reports
about
windows not working correctly on some MS-DOS "work a likes". So it
seems that a very small percentage of the market may have some
problems
if trying to run windows 3.10 on an MS-DOS "work a like". In order to
be fair and up front with our windows users it might be a good idea to
disclose to them in a timely fashion, before they might encounter some
possibly data corrupting problem, that they were running the windows
product on a non-Microsoft MS-DOS on which Microsoft had not done any
testing. This is what the "AARD" code is for. It detects whether the
DOS it is running on is Microsoft MS-DOS. If the DOS is not Microsoft
MS-DOS, a disclosure message will be displayed to the user that
windows, I include all windows components in this, is being run on a
DOS that it has not been tested on.
</p>
<p>
"But wait!!!!" That is not the form of the message that was in the
windows 3.10 betas! That is correct. The message that was in the betas
was crafted carefully to produce a desired effect: A report back to
Microsoft that the message had been displayed. This code was added
very
late in the beta cycle, we were extremely concerned about it having
some subtle bug in it and/or it "misfiring". For this reason we had a
very strong desire to hear about every single occurrence of this
message in the beta program so we could follow up and confirm that in
fact a non-Microsoft-MS-DOS was being used and the code was working
properly. This is why the magic word "error" was used. This is the
only
reason why the word "error" was used. And based on the statements in
Mr. Schulman's article, this strategy was successful beyond our
wildest
expectations. It is still generating "bug reports" a year and a half
after it was disabled!!!! Look at the message: "Please contact the
Windows 3.1 beta support." Do you still think that is what the message
was going to be in the final product if we had left it enabled? Of
course not. if you can change one part of the message, can't you
change
all of it? I think so, don't you?
</p>
<p>
Mr. Schulman seems to be trying to make a big deal out of several
factors which aren't very interesting. I presume that this is mostly
because he doesn't understand the reasons. "The effect of the AARD
code
is to create a new and highly artificial test of [MS-]DOS
compatibility." This code is purposely asking this exact question:
"is
this Microsoft's MS-DOS that Microsoft has tested windows on?" The
strange things the code has to look at to answer this question is to
some extent a commentary on the quality of some of the "work a
likes".
Mr. Schulman goes on at length about how this code is "obfuscated and
encrypted" and that this is somehow an indication of malicious intent.
He then at the end explains completely the exact reason for it!! "An
indication that the AARD code's obfuscation is successful is the fact
that Novell's most recent version of DR DOS fails the test..." That is
the reason for the obfuscation, the complete reason. This code is
likely to be targeted by the "work a likes", which defeats the code's
purpose to disclose to the user that windows is being run on a DOS
that
Microsoft has not tested it on. I am not ignorant enough to think this
task is impossible, the intent was simply to make it difficult. Since
the primary tool used for figuring things like this out is a debugger,
it should surprise nobody that one of the obfuscations is to try and
disable a debugger. "Anyone with a copy of Windows 3.1 can hex dump
WIN.COM [...] and see the error message [...] and the AARD and RSAA
signatures." Welcome to the wonderful world of "fix paranoia".
This
code was added to the betas very late, the last large beta in fact.
The
decision is then made to not do this. I will not waste time going into
the details about why this decision was made. It should be obvious at
this point what the reasons were. Now we find ourselves between the
rock and the hard place. We don't want this disclosure message in the
product, but we want to make the minimal possible change so that the
change does not destabilize the product and require us to do another
large beta to make sure that the disable didn't break something
(probably due to some weird side effect). Leave all the code and the
message in, even run the code, just don't display the message. By the
way, Mr. Schulman's analysis of WIN.COM brings up an interesting
point.
He goes on about how code was added to look at a byte to see whether
or
not to display the disclosure message. This code was added after the
beta went out, but before the decision to remove the disclosure was
made. Unlike SETUP and MSD which are not frequently run things,
WIN.COM
is run every time the user runs windows. A user who has decided
windows
works OK on the MS-DOS "work a like" he is using might tend to get a
little bit annoyed at having to press a key to dismiss the disclosure
message and continue every time windows is started. For this reason it
was decided to add a command line switch to WIN.COM which would
disable
the message and continue. As I recall the byte variable was added to
WIN.COM, but the code to parse the command line switch was under
conditional assembly and was not assembled into the final product. I
agree this sounds a little odd, but that is my recollection of the way
it happened. This meant that when we decided to not print the
disclosure message at all, all we had to do was change the initial
value assembled into WIN.CNF so that the default value was "don't
display the disclosure message and continue". This meant that the
source code DIFF for WIN.COM was a one byte change which is about as
minimal a change as you can get. By the way, don't ask me why the code
and message are completely removed from HIMEM.SYS and MSD.EXE because
I
don't remember although I suspect that the reason was we decided to
make an unrelated change to these components in this time frame and
decided that removing this too would incur minimal additional
destabilization risk.
</p>
<p>
I also note in passing: "Its presence in five otherwise-unrelated
programs also suggests a fairly concerted effort, as it is unlikely
that five so different programs are maintained by the same person. In
fact, the programs probably fall under the domain of several different
product managers or divisions." I agree that a concerted effort was
involved, the rest is meaningless and I am at a complete loss as to
what point Mr. Schulman was trying to make here. The AARD code was all
written by one person for one product, windows 3.10. He is correct
about the fact that these five different programs were the
responsibility of different people, but what does that mean? "Here is
a
module with a routine named xxxxx in it, call the routine and look at
this to see whether the disclosure message should be displayed."
"...given the effort required to write this tricky code." About one
week, all by one person, a large part of this time actually being
testing as opposed to writing. How much effort is required depends
very
much on what the knowledge base is of the person doing the work. By
the
way those signatures, "AARD" and "RSAA" were debugging aids
that would
have been removed if the disclosure had not been disabled. Since it
was
disabled, and the whole thing became uninteresting, the signatures got
left in. They probably should have been removed from the beta too, oh
well.
</p>
<p>
As I said above, what is going on with this code is probably just not
Machiavellian enough for many people to believe it. All we were
interested in doing was disclosing to users in a timely fashion that
they were running the windows 3.10 product on something on which
Microsoft had not done any testing. It seems that even this is
something that you can't do because somebody else who is trying to
leverage 10 years of your hard work by copying it feels they have a
right to expect you to waste a lot of your money doing all the testing
for them for free too. As we observe, something as innocent and well
meaning as this does nothing but generate a lot of complaints about
the
fact that you are being "unfair" to your competitors. Apparently a
lot
of people feel this is more important than us being fair to our users.
This is an opinion I refuse to agree with because it fails to serve
the
most important people, the users of our products.
</p>
</div>[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Sunday, May 12 2013 @ 06:15 PM EDT |
http://groklawstatic.ibiblio.org/pdf/iowa/www.iowaconsumercase.org/011607/4000/PX04372_C.p
df
<p>
Plaintiffs' Exhibit 4372_C<br />
Comes v. Microsoft
</p>
<p>
<b>From:</b> Brad Silverberg<br />
<b>To:</b> Jonathan Roberts (Xenix)<br />
<b>Subject:</b> FW: IBM: Ready To Launch OS/2 Without Windows<br
/>
<b>Date:</b> Wednesday, October 27, 1993 10:02AM
</p>
<p>
ibm's royalties do not go away. even in the most favorable interpretation they
still owe us I believe $3.50.
Plus there are a host of other issues where they may be inducing people to break
their license agreement.
</p>
<p>
<b>Privilege Material Redacted</b>
</p>
<p>
----------<br />
From: Newswire Mailing<br />
To: execnews <br />
Cc: newswire<br />
Subject: IBM: Ready To Launch OS/2 Without Windows<br />
Date: Wednesday, October 27, 1993 7:29AM
</p>
<p>
****IBM Ready To Launch OS/2 Without Windows 10/26/93
</p>
<p>
WHITE PLAINS, NEW YORK, U.S.A., 1993 OCT 26 (NB) -- Unhappy at
paying royalties to rival MICROSOFT Corp., IBM is preparing to
launch a version of its OS/2 operating system without built-in
code to run applications written for MICROSOFT's Windows
operating environment.
</p>
<p>
The Windows-less OS/2 would let users run Windows applications
provided they already had Windows installed on their PCs. OS/2
would establish where the Windows code was stored and would call
on it directly when asked to run a Windows application.
</p>
<p>
This is a practical idea largely because many personal computer
vendors are bundling their hardware with copies of Windows. That
means a fair number of buyers of OS/2 already have Windows on
their PCs. By not duplicating the code in OS/2, IBM would not
only avoid paying royalties to MICROSOFT for its use, but reduce
the size of its OS/2 package and possibly save disk space for
users who don't take the trouble to remove the pre-installed
Windows files when they install OS/2.
</p>
<p>
The company should also be able to pass on to customers some of
what it saves on royalties. This would free buyers of PCs with
Windows bundled from paying twice for the code -- once in the
price of the PC and once in the price of OS/2.
</p>
<p>
Sources said an announcement can be expected by mid-November. The
Comdex/Fall trade show, which begins in Las Vegas Nov. 15, would
be a likely venue.
</p>
<p>
An IBM spokeswoman said the company does not comment on
unannounced products.
</p>
<p>
(Grant Buckler/19931026/Press Contact: Rob Crawley, IBM,
512-823-1779; Mee Lin Sit, IBM. 914-251-5996)
</p>
<p>
NBviaNewsEDGE<br />
Copyright (c) 1993 Newsbytes<br />
Received via NewsEDGE from Desktop Data, Inc.: 10/26/93 15:46
</p>
[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Sunday, May 12 2013 @ 06:17 PM EDT |
http://groklawstatic.ibiblio.org/pdf/iowa/www.iowaconsumercase.org/011607/4000/PX04373.pdf
<p>
PLAINTIFF'S EXHIBIT 4373<br />
Comes v. Microsoft
</p>
<p>
<b>From:</b> Bill Pope<br />
<b>To:</b> bradc; bradsi; keithw; rogersw<br />
<b>Cc:</b> billn; collinsh; debrav; pamk; vstevho; victorr;
w-coll<br />
<b>Subject:</b> RE: Final Draft: OS/2, Windows 3.11
Q&A<br />
<b>Date:</b> Friday, March 04, 1994 4:45PM
</p>
<p>
<b>Privilege Material Redacted</b>
</p>
<p>
----------<br />
From: Rogers Weed<br />
To: Brad Chase; Brad Silverberg; Keith White<br />
Cc: Bill Neukom; Bill Pope; Collins Hemingway; Debra Vogt;
pamk@or.wagged.com; Victor Raisys; Colleen Lacter<br />
Subject: Final Draft: OS/2, Windows 3.11 Q&A<br />
Date: Fri, Mar 4, 1994 3:47PM
</p>
<p>
Here is a draft of this that includes some comments from Collins and BradC
on tone. I'd like to send this out for posting on CIS first thing Monday at
latest, so please comment quickly.
</p>
<p>
Thanks...Rogers<br />
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
=<br />
Windows 3.11 Q&A
</p>
<p>
1. What is Windows 3.11? How is it different from Windows 3.1?
</p>
<p>
Microsoft Windows 3.11 is a "refresh" release that implements
packaging
changes designed to reduce counterfeiting. The release also includes some
updated device drivers and a few bug fixes to address issues such as
diskless workstation support.
</p>
<p>
2. Why were these changes made?
</p>
<p>
It's been nearly two years since Windows 3.1 was introduced to the market.
These new drivers and bug fixes will improve the experience of a customer
buying
Windows for the first time. Also, these changes had already been
incorporated
in our Windows for Workgroups 3.11 release.
</p>
<p>
3. Who should update to Windows 3.11?
</p>
<p>
There is no new functionality in the 3.11 release of Windows and all of the
drivers being added or refreshed have been available on the Windows Driver
Library up on CompuServe and Microsoft's Download Service. So unless a
customer is having a problem related to one of the areas covered by the
fixes mentioned above, there should be no need for them to update to this
release. lf a customer does need to update they should call Microsoft at
(800) 871-3270 for more information.
</p>
<p>
4. What is the difference between Windows 3.11 and Windows for
Workgroups 3.11.
</p>
<p>
Windows 3.11: Windows 3.11 is simply Windows 3.1 plus the additional
video and printer drivers and the five bug fixes described above.<br />
As mentioned above, there is no new functionality or performance
improvements, features, etc.
</p>
<p>
Windows for Workgroups 3.11: WFW 3.11 offers a number of enhancements
to our Windows 3.1 product. WFW 3.11 includes enhanced performance,
from 50-150% faster disk I/0 and up to 100% faster network access. It
also includes integrated networking, Microsoft Mail, Microsoft
Schedule+, Microsoft At Work Fax, and Microsoft's Remote Access
Services, and a number of other enhancements for both standalone and
networked Windows PCs.
</p>
<p>
5. Does Win 3.11 work with OS/2 for Windows?
</p>
<p>
No it does not. From what we have been able to learn without the
benefit of source code, which IBM refused to provide Microsoft, OS/2
for Windows patches Windows in memory at fixed address locations. Such
a design makes OS/2 for Windows extremely fragile because it depends on
Windows code being frozen over time.
</p>
<p>
Microsoft encourages developers to write well-behaved Windows apps using
documented API's and good programming practices. By establishing these
ground rules
we can move the platform forward and maintain support for these
applications.<br />
Windows 3.11 did not break any well-behaved applications that we are aware
of.
</p>
<p>
Though we regret the inconvenience to customers, IBM has to be responsible
for writing well-behaved applications -- they have taken shortcuts with OS/2
for
Windows that have put their customers in a bad position. We presume that one
or more
of the bug fixes incorporated in Windows 3.11 (which were included in
Windows
for Workgroups 3.11 released November 6, 1993) changed the fixed address
locations that are patched by OS/2 for Windows. As a result, OS/2 for
Windows
no longer runs correctly.
</p>
<p>
6. Did Microsoft do this to thwart IBM?
</p>
<p>
No we did not. We have been talking publicly about this release since August
of last year -- well before IBM even announced their OS/2 for Windows
product. Negotiations with Novell and a desire not to interrupt the
Christmas selling season caused us to delay, but Windows has been on the
market for nearly two years now and we felt that customers would benefit
from a refresh of the product.
</p>
<p>
7. What should customers do that have purchased OS/2 for Windows and want to
update to Windows 3.11?
</p>
<p>
IBM has the means at its disposal to permit purchasers of OS/2 for
Windows to use it with Windows 3.11. First IBM can release a new
version of OS/2 for Windows that patches Windows 3.11 at the
appropriate address locations in memory. Second, IBM can provide
customers with the KERNEL, USER, and GDI modules from Windows 3.1 that
do not contain the bug fixes which apparently are the source of the
problem. Alternatively customers can purchase the full OS/2 2.1 product from
IBM.
</p>
<p>
8. What about customers that buy Windows 3.11 and then want to run OS/2 for
Windows?
</p>
<p>
We think IBM bears the responsibility for designing OS/2 for Windows in
such a way that virtually guaranteed it would break if Microsoft made
any changes in Windows. Nonetheless, in an effort to assist our mutual
customers, Microsoft will provide purchasers of Windows 3.11 who
experience difficulty running the product with OS/2 for Windows with
the unimproved Windows 3.1 files which we modified in Windows 3.11 to
fix various bugs. Our Product Support Services will distribute these as
soon as a diskette can be made available.
</p>
<p>
9. What exactly are the code changes made to Win 3.11 and what do they do?
</p>
<p>
First the following drivers have been added or refreshed in the Windows 3.11
release:
</p>
<p>
- HPLJIV driver<br />
- 256 Color generic SVGA video driver - which will support TSENG ET4000
chipsets and V7, but not ATI Ultra or S3 chipsets.<br />
- As well as the following drivers:<br />
PSCRIPT.DRV UNUDRV.DLL<br />
PSCRIPT.HLP UNIDRV.HLP<br />
HPPCL5MS.DRV HPDSKJET.DRV<br />
HPPCL5E.DRV HPPCL.DRV<br />
HPPCl.5E.HLP PAINTJET.DRV<br />
HPPCL5E1.DLL L1OO_425.WPD<br />
HPPCL5E2.DLL L300_471.WPD<br />
HPPCL5E3.DLL L300_493.WPD<br />
HPPCL5E4.DLL L5OO_493.WPD<br />
HPPCLSEO.DLL L200230&.WPD<br
/>
FINSTALL.DLL L33O_52&.WPD<br
/>
FINSTALL.HLP L630_52&.WPD<br
/>
EPSON24.DRV EPSON9.DRV<br />
V7VDD.386 V7VGA.DRV
</p>
<p>
In addition, the following 5 files were changed:
</p>
<p>
1) krnl386.exe: Minimal kernel changes to more gracefully shut down the
Windows system after installing some 3rd party hardware (which call for
Windows to shut down before returning focus to program manager).
</p>
<p>
2) gdi.exe: Changes which assist ISVs currently writing graphics
applications, especially those that call the 'draw rectangle function'.
This will minimize the number of API calls the programmer needs to
write and maximize the system resources available on the system when
using graphics intensive applications.
</p>
<p>
3) commdlg.dll: Changes which allow Windows apps on diskless workstations
to use common dialog boxes. (Formerly available as a PSS application note)
</p>
<p>
4) pscript.drv & unidrv.dll: Updated to eliminate a font enumeration
conflict that occurs with many word processors when 60 or more fonts
are used in a single document. (Formerly available as a PSS application
note)
</p>
<p>
5) vtda.386: The updated vtda.386 solves a very small percentage of
timer related problems that occur when starting an MS-DOS based
application under Windows on large Novell Netware networks. (Formerly
available as a PSS application note)
</p>
[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Sunday, May 12 2013 @ 06:18 PM EDT |
http://groklawstatic.ibiblio.org/pdf/iowa/www.iowaconsumercase.org/011607/4000/PX04499.pdf
<p>
PLAINTIFF'S EXHIBIT 4499<br />
Comes v. Microsoft
</p>
<p>
File : c:winmailfeb10.fld
</p>
<p>
####################################################### 18<br />
From bradc Tue Feb 4 11:20:33 1992<br />
To: lorisi<br />
Cc: bradsi mackm sharonh<br />
Subject: FW: RE: borland and dri<br />
Date: Tue Feb 04 11:19:08 PDT 1992
</p>
<p>
please set-up a one and a half hour meeting to brainstorm on ms-dos
5.x strategies. include everyone in the mail below plus anyone mack
would like to add (tomle?) unless bradsi wants to add or subtract
people
</p>
<p>
thanks<br />
>From bradsi Mon Feb 3 18:32:47 1992<br />
To: bradc mackm richt<br />
Cc: davidcol johnlu richf russs<br />
Subject: RE: borland and dri
</p>
<p>
Date: Mon, 03 Feb 92 18:30:05 PST
</p>
<p>
yes, we need to get together to discuss. i agree that we don't want
to negatively affect msdos6. but realistically, msdos6 is still
quite a ways off. we need to have a close look at it, too, to see
what's essential and what's not. i presume msdos won't be until
mid-to-late '93. i don't see how we can hold the fort with msdos 5.0
</p>
<p>
until then. if dri was smart, they'd put us on a release treadmill.
get some new utils, call it 6.5. get a new round of reviews. do it
again 6-9 months later. each round of review would say, dri was
already ahead, how they are farther ahead. once you lose a lot of
ground it is very very hard to pick up. we need to stop losing
ground while making progress towards the ultimate "major leap"
msdos6. clearly the trick is to not make kernel changes and just add
bought utils.
</p>
[ Reply to This | Parent | # ]
|
|
|
|
|
