Don't Fall Into The Sea
I read "Great Ideas Get Lost In The Sea Of Incomplete Documentation" \[Sept. 18, p. 68\], and you're absolutely correct. However, the problem shouldn't be stated as "We are not documenting our work well enough" (to paraphrase your ideas), because this is a symptom. The problem should be stated as "We are not developing our products correctly."
The solution to the incomplete/missing documentation has been available for decades. But the development community has been misled about how to do development and about the existence of a really excellent development methodology that solves the problem. Actually, you didn't go far enough in your complaint. There's a lot more to be said on the effects of inadequate documentation.
Telematics Communications Group, Motorola
Let's Move Analog To A New Level
I'm an analog designer from way back in the vacuum tube days, and I say that it's about time these devices became available \["Moving Programmable Analog To A Higher Design Platform," Oct. 2, p. 60\]. We analogers like to tweak things and come up with our own neat little circuits. But, as you know, there are plenty of plug and chug operations that we also need to use for a successful design. I think programmable devices would help me to simplify my designs, so that I would have more time to have fun tweaking. I, for one, would like to know more about developments in this area. Even though I'm not a digital designer per se nor a neophyte, I'd like to experiment with these devices.
Out-Of-Date Before It Hit The Page
It appears that you reported on some Intel Vaporware in the Oct. 16 issue. Timna is dead!! \["Revamped Microarchitectures Let CPUs Deliver Top-Notch Performance," p. 25\] I was just published in another magazine for suggesting what to do with 1 million recalled PIII motherboards. My suggestion was that they be given away as controllers in a NASA-sanctioned Martian Lawn Dart Contest. A better suggestion was that they be used as skeet targets at an AMD picnic. Intel is no longer king of the microprocessor hill and their news releases should receive some scrutiny.
You got me on this one. Unfortunately, that article was written about 10 days before Intel announced the plug would be pulled, and by then it was too late to extract the out-of-date information from the article. So, the article had to be printed as originally written. But I'm glad to say that this kind of occurrence is the exception rather than the rule with material published in Electronic Design.—Dave Bursky, Editor-In-Chief
A Gamble Of Psychology
I enjoyed "Economics For Engineers Is A Branch Of Psychology" \[Oct. 2, p. 160\]. It kind of ties into an observation that I made about the stock market when I initially started to invest in the market. As an engineer, I started by researching companies' profitability, product lines, etc., in an attempt to find some logical relationships to develop an "engineering" understanding of the market.
One of the first things that struck me as very odd is that there doesn't seem to be a relationship between how much a company makes and the value of its stock (at least for the small investor). There's no discernable feedback loop between profits/losses to the company and value of the stock—except in some sort of odd psychological sense. The "value" of a particular stock, in terms of whether a company is making money or not, is disconnected for me because I get very little, or no, direct return on my investment in the stock. (I don't get an important part of the profit distributed to me.) The only return on investment that's significant to me is that which is based on the increase in the price of the stock, which is only based on what others are willing to pay, not necessarily on how well the company is doing.
People seem to pay some attention to the health of the underlying business, but it isn't clear why this matters, except that it gives a hint about what value others like myself will artificially place on the stock. That value is based upon what we think others will be willing to pay in the future, not on how much the company makes. The actual health of the company only made any noticeable difference to my investments when it went out of business. This was not a good event from the point of view of the value of my stocks. It seems like an "all or nothing" concern rather than one based on details about the health of the company under consideration.
If a person buys some or all of a small company, the value of that purchase is directly related to the profitability of the company because the profits go directly back to the owners. If my company makes $500,000 a year in profits and can distribute $300,000 a year to the owners, then the value of the purchase can, in part, be determined on the time value of those distributions. An extremely weak or nonexistent return of this type exists for purchases of stock on the stock market. It's interesting that we can sustain a process like the stock market which supposedly means that we are purchasing ownership in the company, while the direct benefits of ownership play little or no role in the value of that ownership. It makes investment in the stock market a gamble of psychology, not an engineering study.
On page 80 of "Tool Set Optimizes And Validates Designs Based On The PCI-X Bus" \[Oct. 2\], I think that you meant 1.066 Gbytes/s rather than 1.066 Gbits/s for PCI-X throughput.