Letters to the Editor - Letter to the Editor

Computer Technology Review, Sept, 1999

Griping lack

For immediate release: I'm really hurt (and I'm not exaggerating). I just finished reading Mark and Hal's June column ("The Editors Gripe ... Again," page 8). Unfortunately for us PR types, we must admit to an element of truth in your editorial--but you'll never obviate our optimism. From now on, I promise to temper my optimism with a little objectivity. Print that! Thanks for the insight, guys.

Sarah Huntzinger Optimum Networking

I read Mark and Hal's June column ("The Editors Gripe ... Again," page 8). Now I have a designer's gripe and a couple of questions. I have designed a system to reduce speech to about 1200 bits per second. Not only is it demonstrable, but no one who has been shown the technique disputes its capabilities. Moreover, with some additional developments to the system, I am sure I can get full duplex speech down to 300 bits per second.

However, no one seems to care, which begs the question: Is there a market for this technology, such as cell phones, Internet telephony (the technology would require less than 3Kbit reserved bandwidth), radio communications, or any place that speech is stored or transmitted? The responses I get from companies include, "We are only interested in basic technology" (I thought this was basic technology) and "We have wide bandwidths so we don't need narrower ones."

Is there any way to get past these idiot gatekeepers who are intent on protecting their company from outside technology? And how do you get to the people who could appreciate new approaches?

The basic problem seems to be that if you are not known or do not have the access or it wasn't invented here, your technology is questionable at best. Do you think that this technology has a future and if so, do you have any ideas on how to get it out there?

Gene O'Quinn

I'm not an engineer or a hacker--I'm more of a "sports-writer" than a "ballplayer," if you know what I mean. So I can't draw any conclusions about the viability or originality of what you've done, or about the response you've gotten.

My advice, for what it's worth, is this: If you have invented any technology that works--even if it works only on your test bench--the first thing you need to do is apply for a patent. It takes time, but at least you'll know if anybody has done it before (in which case, you can find out what happened to it, whether anybody has licensed it, etc.).

Also, I don't know if you work for a big company, but the employment contracts at many companies require you to submit any invention of yours--even something you did exclusively on your own time--to that company. They'll want first right of refusal; and only if they really don't want it will they formally release it to you. Good luck.

--Hal Glatzer Editor-at-Large

Y2K: Not The End Of The World As We Know It

I believe that the hysteria over so-called Y2K problem with computers is overblown. A few new Federal laws, combined with some simple actions by computer users, can provide a common sense solution. Disaster should be obviated.

First, let's divide all vulnerable computers programs between those that depend on date comparisons and those that don't. One example of a non-dependent program is your word processor. Another is a blood-testing device in a hospital lab. Each of these tasks can be performed typically without reference to dates prior to the current day. Next, assume that the hardware or operating systems running such programs will crash on or after January 1, 2000.

You can simply reboot such computers, set the date to January 1, 1990, and get ten more years of usage from the particular software, If commercial data were distributed from that computer, it superficially would appear to be ten years old.

That's where the Federal government comes in. They should pass a law that states, for commercial purposes, any document prepared after 1/1/00 that is dated exactly ten years earlier, and issued in good faith, is valid and required to be accepted by other parties as if the date were current.

Next, consider those programs that are dependent on date comparisons. One example is your cable TV bill, which runs from month to month. Presumably, the big problem inside that billing software is to shoehorn a four-digit date into the space allocated for a two-digit date. This required modification of the program and the data is the essence of the Y2K bug.

However, a simplistic solution would eliminate that big hurdle: just change all the dates stored in the computer to dates ten years earlier. Then, run the old program under protection by the aforementioned Federal laws and receive another ten years to permanently resolve the problem. Admittedly, this may take a programmer to accomplish. Yet one has to believe that changing dates is a lot easier than revising programs.

Variations of this law could address specific situations. For example, if day of week information is important, one could use 1972, as it is exactly the same as the year 2000.

In conclusion, Y2K fears are not justified. A simplistic solution will address the unresolved problems that could crop up this New Year's Eve.