Summary: Antitrust exhibits about Microsoft DOS (MS-DOS) versus DR-DOS

IN MICROSOFT'S path of destroying superior (in quality) competition, illegal things were done. For instance, how about "Bambi" and the "Alien OS"? Today we present fuller portions from the same type of case.

In Exhibit PX_9923 (1989) [PDF], which is available as text at the bottom, Microsoft is seen analysing ROM DOS, although its people stated in public that they only ever tested MS-DOS. The message is sent to:

Mr. Leonard Liu President Acer Incorporated 602 Min Sheng East Road Taipei 10445, Taiwan R.O.C.

The next exhibit, Exhibit PX00993 (1991) [PDF], shows that Microsoft was indeed testing operating systems other than MS-DOS, which contradicts claims that were made at the time. See for example:

Does anybody know how to detect dr dos 6.0? Bambi will not run properly on dr dos 6.0 because of a quirk in their device driver handling, so we need to detect them.

And also:

What do we think? Should we test further with the patched Bambi to see if there are any more incompatibilities???

For more information about "Alien OS", see this famous antitrust exhibit, the DOS index, and DRI index. ⬆

---

Appendix: Comes vs. Microsoft - exhibit PX_9923, as text

---

Plaintiff's Exhibit 9923 Comes v. Microsoft

Microsoft Corporation 1601 NE 36th Way Box 97017 Redmond, WA 98073-9717

Tel 206 882 8080 Telex 160520 Fax 206 883 8101

Friday, September 22, 1989

Mr. Leonard Liu President Acer Incorporated 602 Min Sheng East Road Taipei 10445, Taiwan R.O.C.

FAX: 011 886 (02) 715-1950

Dear Leonard,

It was good to talk to you on the phone the other day. I hope we can see each other again soon. Maybe Comdex?

Attached to this note is your personal copy of our analysis of the ROM versions of DRI's product versus Microsoft's. I think you'll see that we have a large advantage over DR-DOS in the size of the operating system kernel: this is surely critical in machines designed to ship with 256k memory. There are several other important strengths mentioned in the document.

What I really want to tell you - on a confidential basis - is the answer to the question you raised when we talked. You asked "Is this ROM DOS strategic for Microsoft?" The answer is a strong "yes".

On the 26th of this month, we are going to announce officially the ROM DOS product. That announcement, which will be made in Europe first, will be accomplished by Psion's announcement that they have licensed Microsoft ROM DOS. Psion, a British company, is the maker of the Psion Organizer which is probably the largest selling pocket computer company in the world: it's a hit in Europe. Their current technology is eight bit, but they will be moving to 16 bit and MS-DOS. Psion will also announce that they will use our Flash File System, which supports the FEPROM technology that Psion has designed into its system for mass storage. We will also announce that Poqet, a California company, has license Microsoft ROM DOS and will be incorporating it into a future version of its pocket PC. We expect similar announcements from Japanese companies in the near future, for pocket PCs, embedded control machines and small personal computers.

Our US press release is scheduled for October 2nd, as we believe this date will get us better coverage in the US.

This product is by no means "something that we have dreamed up to just propose to Acer". It has been in development for a year. It and its future versions are a key business line for us. It is our plan to expand the range of our offerings as 16 bit processors move into new markets.

I'm sure that Alan Sugar would not mind me telling you about his decision for Amstrad. He decided to stay with MS-DOS (they use the disk version). In 1985 when he first came out with his low end 512k machine, he shipped both DRI's then-current product and M-DOS: his market research showed that virtually all users chose to use the Microsoft product, rather than risk the compatibility questions that DRI's operating system products raise. His more recent machines dropped the DRI products entirely. Armstrad is a

Microsoft Confidential

X 517975 CONFIDENTIAL Page 1

---

company that ships most of its machines into the home and small business market; but some of their machines go into businesses that use networks, or that use Windows. It only takes a couple of reports about non-compatibility to give the kiss of death to a PC: we've seen that on the hardware side as well in as in the operating system area.

Your concern about Basic supporting Hercules is legitimate. We actually have two solutions to that. We can license our famous Quick Basic Interpreter to you at very little cost, or with a little engineering work we can provide a version of GW Basic that supports Hercules. Quick Basic is one of the products that we are evaluating for localization into Chinese. While localization will not be completed in the time frame of the shipment of the Third Wave machines, it would surely make more sense in the future to move from the US version of Quick Basic to a Chinese version of the same product. Also please remember that Microsoft is very committed to future versions of Quick Basic; Borland has announced that they are not planning further Turbo Basic work.

Regarding the commercial terms we have proposed, I'd like only to point out that we have already agreed, contrary to our usual practice, to license these new machines for our ROM DOS without any increase in financial commitment. This recognizes the large volume of business that Acer already does with Microsoft - business that we very much appreciate. I am sure that once you have had a chance to look at the data, you'll agree that the slightly higher cost of including the real MS-DOS product will ensure a more successful product.

Microsoft cares very much about your decision on this operating system requirement. There are many exciting developments due at the lower end of the market in the coming year or two, including the optical or multimedia PC that Ken Tai and I talked about last winter. We want to see our relationship cover the high and low ends of the personal computer and workstation markets. If there is any other information you need, don't hesitate to have your people call us. As you know I am ready to receive your call at any time. As leading companies in the industry we would be foolish not to work as closely as possible together.

Yours sincerely,

Microsoft Confidential

X 517976 CONFIDENTIAL

Page 2

---

Appendix: Comes vs. Microsoft - exhibit PX00993, as text

---

Plaintiff's Exhibit 993 Comes v. Microsoft

CONFIDENTIAL

Non-responsive Materials Redacted

X0592196

From chuckst Wed Nov 6 18:38:46 1991 To: philba Cc: scottq Subject: Scott's 9/30/91 mail re: DRDOS Date: Tue Feb 23 17:07:06 PDT 1993

>From scottq Mon Sept 30 13:57:38 1991 To: dosdev Subject: detect dr dos 6.0 Date: Mon Sep 30 13:50:35 1991

Does anybody know how to detect dr dos 6.0? Bambi will not

WinMail 1.21 philba Tue Feb 23 16:49:36 1993 Page 160

MS-PCA 1143047 CONFIDENTIAL

---

run properly on dr dos 6.0 because of a quirk in their device driver handling, so we need to detect them.

Scott

CONFIDENTIAL

From chuckst Wed Nov 6 18:40:05 1991 To: philba Cc: scottq Subject: More Bambi/DR-DOS mail Date: Tue Feb 23 17:07:11 PDT 1993

From chuckst Sun Sep 29 17:16:46 1991 To: mikedr philba scottq Subject: Bambi on DR-DOS 6.0 Date: Sun Sep 29 17:16:39 1991

I tracked down a serious incompatibility with DR-DOS 6 -- They don't use the 'normal' device driver interface for >32M partitions. Instead of setting the regular START SECTOR field to 0ffffh then using a brand new 32-bit field the way MS-DOS has always done, they simply extended the start sector field by 16 bits.

This seems like a foolish oversight on their part and will likely result in extensive incompatibilities when they try to run with 3rd part device drivers.

I've patched a version of Bambi to work with DRD6, and it seems to run Win 3.1 without difficulty. This same problem may have caused other problems with Win 3.1 and the swapfile under DRD6.

It is possible to make Bambi work, assuming we can come up with a reasonably safe method for detecting DRD6. The runtime hit would be minimal in time and space, although we would have a couple of instructions in the main code path for checking the 'special' DRD6 flag.

What do we think? Should we test further with the patched Bambi to see if there are any more incompatibilities???

Non-Responsive Material Redacted

X0592197

WinMail 1.21 philba Tue Feb 23 16:49:36 1991 Page 161

MS-PCA 1143048 CONFIDENTIAL

Credit: wallclimber and RCH.