Looking Back 25 Years

Twenty-five years ago IBM announced the Model 5150 Personal Computer. For those why missed that event, this was a a $1500 and up-priced system with an Intel 8-bit CPU. Capability-wise, I believe it sufficient to say that today it would not be considered even for a doorstop.

More interesting to us was the software side of the story. IBM wanted to sell hardware and found a Harvard drop-out named Bill Gates that said he could supply the operating system on the required timeline. Gates bought the software from a local Seattle company and then customized it to meet the needs of IBM.

While Gates only paid $50,000 for that software he was clearly already a good businessman and licensed it to IBM such that he could also sell it to others. That decision, of course, meant that many other companies could start with IBM's open description of the PC, build their own systems and license the same software.

That's what happened but let's look at what might have happened if either IBM or Microsoft had made some different decisions. That is, if IBM had decided to make the PC proprietary or if Microsoft had either given IBM exclusive rights to the software or had decided to give it away the way Linux is given away--that is, source code and the freedom to change it.

Apple already had the Apple II which is certainly a contributing factor for IBM to quickly want into the market. The Apple II can be a case study for the closed hardware and closed software approach. It is reasonable to assume that IBM (and MS-DOS) would have ended up with small market share as well if someone else came along with open hardware--something that was extremely likely.

What if IBM had had exclusive rights to the operating system software? It seems that there are two possibilities: either IBM could have sold rights to other computer manufacturers or another operating system would have been developed. In the first case the real difference is that software profits would have shifted from Microsoft to IBM. Possibly, if IBM had elected to open up information on the software the same as was done with the hardware, other applications software developers could have ended up with greater market share. In the second, we would likely have had competition in the operating system software early on.

Let's look at one other possibility--free operating system software. Assuming this meant free as in freedom--source code available to all--this would likely have created a lot of incompatibilities as each hardware vendor decided to make their computer "better"--meaning different--and customized the software to only work on their system.

But, there was operating system competition (CP/M already had been shipping on some systems including those made by NEC). Also, there were free applications. Possibly the most famous was PC Write by Bob Wallace, an ex-Microsoft employee. It was relatively common for retailers to include PC Write on computers they sold as a free add-on.

So, what happened? That is, why did competition go away in software but not hardware? All along, IBM was being "open". The lock-in was coming from the software vendor. Initially that was just the operating system but, as Microsoft grew, so did their offerings. Their office applications had a market advantage because of inside software knowledge.

Everyone knew what the hardware did. That was IBM being open. But, vendors did not know what was inside Microsoft software. Thus, it was easy for Microsoft to write "better" applications programs because they knew the secrets of the operating system including what would be different in the next version.

How does Linux make this different? Like the openness of IBM's hardware design, Linux offers this openness in software design. Thus, software applications developers start out with equal knowledge of what it takes to write a good application. That doesn't mean that everyone will implement a good application but it does mean it is possible.

In addition, starting on this equal footing has encouraged vendors to adopt open standards for information interchange. For example, the OASIS standard for document formats means that you can move documents between OpenOffice.org, Kword and AbiWord. Again, Microsoft is dragging its feet on OASIS format support because they lose their inside knowledge advantage.

In any case, let's thank IBM for bringing us the PC--the most common Linux hardware platform. And let's hope it doesn't take another 25 years to get a full set of compatible applications programs in place.

fyl - Mon, 2006-08-14 15:05.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Looking Back 25 Years

Good article to make you reflect and think about the recent history of computing. However, I would like to pick at the tone a little...

>All along, IBM was being "open".

This is a very generous statement.

IBM used existing parts and designs in building their first home computer where ever it could in order to be cheap and quick, not in order to design an open platform. When they agreed to Gates's request that Microsoft keep ownership of the software they agreed because they did not see it affecting anything.

The BIOS chip was made as proprietary as possible in a quick time frame so no one else could make an IBM compatible computer. Compaq reverse engineered how the BIOS chip worked and then went to Microsoft putting together the pieces needed to make an IBM clone. The IBM compatible market came about because of luck, skilled lawyers and sneaky/clever Compaq developers, not because IBM was being open or even vaguely helpful

If you think that the advent IBM compatible was a good thing, then you should be thanking coincidence, not IBM.

Linux has been available on virtually every micro-computer platform there has been since it was first written, and Unix has been available on pretty much every single home desktop computing platform in the last 25 years. This leads me to believe that the most common Linux hardware platform would always be whatever the most common hardware platform in general would be in any given theoretical history. I find it hard to believe that if IBM hadn't become the most popular but something else did that Linux would some how not be on that platform and therefore we should thank IBM.

So although the article makes for good thought and reflection, I do not think anyone owes IBM any gratitude.

There is free as in speech, free as in beer, and free as in IBM forgot their beer on the counter and the bartender said it was OK for you to drink it.

Davros (not verified) - Tue, 2006-08-15 05:04.

Well Said!

ROFL! Well said!

The thought that IBM was "being open" the whole time indicates a definite mis-remembering of all the proprietary pieces IBM put in their desktops. Can you say "micro-channel" RAM? And if I remember correctly, IBM had proprietary connectors for their internal components and peripherals as well. IBM was in it to make money, not give anything away. And they only got out of the desktop market when the "IBM-clone" market grew large enough to eat away a sizable portion of their market share.

I agree with you that the article makes for good reflection. But I think we should be thanking the ingenuity of Compaq engineers more than coincidence.

The gratitude we owe IBM is for making the PC a household name, thus generating enough public interest to grow the market, thus sparking rival companies to compete in that market segment. Face it, until IBM got interested, products from companies like Apple, Atari, Commodore, etc., had limited exposure and were more a novelty than the "next big thing."

Sandivar (not verified) - Fri, 2006-08-18 10:20.