Fixing the Bugs
Austin, TX (Oct. 23, 1998) -- Today, I return with an unusual volley to yesterday's article. In haste to get the column out to you on Thursday, and due to the fact that my column ran long, some edits were performed on it that just didn't sit right with me. In fact, on a few, I was downright disappointed and even flamed Tom a bit (woohoo!).
That was a relief to me yesterday -- flaming has its merits -- but today, as a service to you, I'd like to provide a few corrections to yesterday's article. I think they're important clarifications if, together, we're to understand how the WinTel-based personal computer industry really developed.
1) Let's start at the beginning. I need to make it clear that I don't believe Microsoft has ever "enterprisingly grabbed new markets." Truly, Microsoft has never done this. What the company has excelled at is knowing how to hold onto the customers it already has, as a way to incrementally build for the future. The critical point I wanted to make is that Microsoft recognized the benefits of backwards compatibility early on, and they understood both how to innovate and when not to innovate. In essence, Microsoft was extremely and successfully conservative! While they were behind the times in 1980, they hooked up with a company that had an extremely conservative customer base (Big Blue) and turned that into an advantage. Patience, persistence, and business conservatism were some of their main strengths.
2) Next up, I need to be clear that IBM did not outsource all of its parts. For example, the "Basic Input-Output System" (BIOS) ROM code that controls much of the low-level hardware was IBM's own, proprietary, copyrighted property. It thought that ownership of this single component would be enough to protect its system from being copied. At the time, IBM didn't think there was enough demand for PCs to interest any company, let alone those with the resources to create their own compatible version from scratch. And, in a way it was right; before it entered the PC market, there wasn't very much demand for personal computers. IBM massively underestimated its own impact as a standardizing and legitimizing force.
3) It should also be clear that the other PC makers did not go to Microsoft and Intel with the simple message of "Help us compete with IBM." Their message really was, "Hey, we'd like to buy your products." The distinction is important because Microsoft and Intel weren't organized to battle with IBM; in fact, they maintained good working relationships with IBM for many years. Intel maintains vestiges of that partnership to this day. All either company was doing with other partners was maintaining their right to sell products on the free market.
4) We have to be clear on the anti-piracy mission at Microsoft. Anti-piracy was in part a convenient excuse to squelch competitors. Several exclusive provisions in their contracts had nothing whatsoever to do with piracy. Microsoft's success in powering its way onto personal computers was never "unintentional;" it knew exactly what it was doing.
5) Let's go to the Commodore for the next two alterations. Former Commodore users know that it wasn't that the Commodore 64/Commodore 128 thing was a "pain in the arse," it's just that the C64 programs could not access the new C128 functionality. The old programs could not co-exist with new ones. You were forced to run one mode or the other. Because of this, the programmers could not incrementally expand into the new features without completely rewriting their products. There were no gradual transitions. It was all or nothing.
6) How about that awesome computer the Amiga? The Amiga wasn't so much the "straw that broke the camel's back" for Commodore. People bought C-128s to use as C-64s -- there was even that little switch to make it boot straight into C-64 mode. But when they moved to Amiga, there was no carry over between machines. The added value of the Amiga -- the faster CPU and extra RAM -- was wasted. Essentially, the Amiga had no value as offspring to the Commodore 64; it was a brick wall in the upgrade path; users had to start over. Ground zero. No brand. No history. And even though Commodore planned to provide a software C-64 emulator, they let the shipping date slip until years after the Amiga product first hit the market. C-64 users felt that they had been abandoned.
7) Let's jump forward to the end of the article. In the final paragraph of yesterday's article, the word simplicity was inserted to explain one of Microsoft's aims. I need to clarify something about that. Simplicity actually has very little to do with Microsoft. Backwards compatibility is a valuable thing for users, but it's not a simple one for programmers. It leads to unnecessary duplication and extremely complex code that you can't change for fear of breaking something.
This is a topic I look forward to covering in more detail -- namely, how the massive bloating of modern Microsoft's operating system is bound to cause problems. Why? Because their approach now means that the code is growing exponentially larger. Each new release must support everything from the previous releases no matter how irrelevant, or even no matter how dangerous they could be to the present application. Fools need to know that Microsoft didn't understand backwards compatibility "best," it merely understood it first, learning from the mistakes of others. Their style of backwards compatibility today could present them with real problems in the future.
8) And now... I'd like to make one final point that speaks to some e-mail I've received about my articles. Let's agree to agree that computers have a long and rich history that a single article (or even a weeklong series) cannot hope to cover definitively. The title of yesterday's article was a little misleading (that was Tom's title! He he.). What I offered in the column was a very simplified overview of the history of the WinTel PC during the 1980s (back when Microsoft built its empire with the enthusiastic support of an industry that grew up with it). It was an attempt to provide some context and to share a few of my reasons for the general discomfort I feel with Microsoft today.
Perhaps I should return to this subject in three weeks to provide more in-depth coverage, without cramming so much into one article that Tom has to weed-whack it for the website. What do you think, Fools? Let me know on my perennial (if intermittent) hangout, the Cash-King Strategy board.
Have a great weekend, Fools. I do find this subject fascinating because 1) I'm a programmer professionally, and 2) hundreds of billions of dollars in shareholder wealth was created by an industry that many did not expect. Understanding it could lead to a lifetime of riches.
- Rob Landley (Oak)
Day Month Year History C-K -0.85% 1.90% 7.63% 7.63% S&P: -0.72% 5.28% 6.43% 6.43% NASDAQ: -0.52% 0.00% 1.65% 1.65% Cash-King Stocks Rec'd # Security In At Now Change 2/3/98 24 Microsoft 78.27 106.38 35.91% 2/3/98 22 Pfizer 82.30 103.50 25.76% 5/1/98 37 Gap Inc. 51.09 56.75 11.08% 8/21/98 22 Schering-P 95.99 100.31 4.51% 2/13/98 22 Intel 84.67 87.25 3.04% 6/23/98 34.5 Cisco Syst 57.56 58.69 1.95% 2/27/98 27 Coca-Cola 69.11 70.25 1.65% 2/6/98 56 T. Rowe Pr 33.67 30.63 -9.05% 5/26/98 18 AmExpress 104.07 87.63 -15.80% Foolish Four Stocks Rec'd # Security In At Value Change 3/12/98 20 Eastman Ko 63.15 73.50 16.39% 3/12/98 20 Exxon 64.34 71.25 10.75% 3/12/98 15 Chevron 83.34 80.94 -2.89% 3/12/98 17 General Mo 72.41 60.31 -16.70% Cash-King Stocks Rec'd # Security In At Value Change 2/3/98 24 Microsoft 1878.45 2553.00 $674.55 2/3/98 22 Pfizer 1810.58 2277.00 $466.42 5/1/98 37 Gap Inc. 1890.33 2099.75 $209.42 8/21/98 22 Schering-P 2111.7 2206.88 $95.18 2/13/98 22 Intel 1862.83 1919.50 $56.67 6/23/98 34.5 Cisco Syst 1985.95 2024.72 $38.77 2/27/98 27 Coca-Cola 1865.89 1896.75 $30.86 2/6/98 56 T. Rowe Pr 1885.70 1715.00 -$170.70 5/26/98 18 AmExpress 1873.20 1577.25 -$295.95 Foolish Four Stocks Rec'd # Security In At Value Change 3/12/98 20 Eastman Ko 1262.95 1470.00 $207.05 3/12/98 20 Exxon 1286.70 1425.00 $138.30 3/12/98 15 Chevron 1250.14 1214.06 -$36.08 3/12/98 17 General Mo 1230.89 1025.31 -$205.58 CASH $48.07 TOTAL $23452.29 *Please note: On 8/4/98 $2,000 cash was added to the