Electronic Design

# Bob's Mailbox

Dear Bob:

This letter is in response to your August 20, 1992 article titled "What's All This Spreadsheet Stuff, Anyhow?"

There is nothing wrong with the Spreadsheet!  The problem is with the user of the spreadsheet.  Back when we trained our engineers to use slide rules for their calculations, the first thing we learned was about accuracy.  We all learned how to do complex calculations but understood the limits of the slide rule's accuracy.

Unfortunately, in business school there is no caution about the accuracy of the trusted spreadsheet.  The problem with the spreadsheet is less with the tool than the person using the tool.  The spreadsheet is an outstanding tool when put to use by a person who understands the limits and proper application of the tool.

V.A. Claude, Orlando, FL

Yeah, nobody does a sanity check on their computers-sigh!-RAP

Dear Bob:

I do not quite agree with your views about spreadsheets.  I have been using spreadsheets (computer spreadsheets) for the last 5 years and I have found numerous uses, which could not be done by hand calculation.  The following are some examples.

To look at parametric distributions of a particular product, frequency distributions and trend charts can be produced very easily by using spreadsheets.  These by the way, give a precise picture of what the product does, and this can not be seen by just looking at numbers.  To fine-tune a process to get optimum performance and yield, such data is not only useful but necessary.

The distribution plots and trend charts make a lot of business decisions very easy, and no one's personal judgment and experience is necessary.  The data speaks for itself.  The distributions can also be used for competitive product analysis, which tell you exactly where you stand with your competitors.  This type of data analysis gives you a direction for improving the product to make it more competitive or even to set goals for new products.

The statistical data analysis becomes very easy with the use of spreadsheets.  Pareto analysis of failures points to the most critical problem and also indicates all other problems in order of importance.  The simple data sorting of failure categories would be an enormous task if one wants to do it manually.  Also, the facility of importing data from automatic test equipment simplifies the analysis process.  The Paretos are not only useful in technical analysis, but in business planning as well.  The volumes of numerous products can be compared at the same time and forecasts made based on the facts.  The products that are not selling well are also identified at the same time, and the sales force can be given the directive to focus on those.  The same technique can be used to monitor expenses and control spending on all of the things involved in managing a business.

Thus, there are plenty of very good uses for spreadsheets, in technical as business areas.  And spreadsheets do not endanger corporate health at all.  The complaint in your article could be applied to any machine.  We must remember that machines can not think.  If a human being makes a mistake, they will repeat it or may magnify it.  But, there has to be a human error involved.  If there is no human error, there is nothing to complain about-but there is a lot to benefit from today's machines (including computer spreadsheets).

Abhijit Tulpule

Oh, yes, you are surely right, spreadsheets do things that otherwise would be ABSOLUTELY IMPOSSIBLE; no judgment necessary.-RAP

Dear Bob:

I always enjoy your column and have been tempted to stick my oar in before now.  But I don't see how I can let this Spreadsheet subject go by without some comment.

First, your examples concerning ROI and frequency optimization bring just one word to mind:  GIGO.  Sure, you can get incorrect answers with a spreadsheet-but I can get answers, probably even MORE incorrect, with pencil and paper.  Let's not blame the poor spreadsheet because someone feeds it garbage.

Second, what kind of accuracy do you get from a slide rule?  Better than a spreadsheet?  What KIND of spreadsheet?

Finally, you are right about the spreadsheet only thinking (my apologies) linearly.  Computers in just about any application aren't worth a toot at being creative, which sounds like what you and Schrage were talking about.  But be reasonable-my hammer doesn't do a real good job of removing chips from a circuit boards, either, but I still find uses for it      !  Maybe it's a question of the right tool for the right job...

Keep the columns coming!

Bill Aull, Technical Director, WLTX TV, Columbia, S.C.

That reminds me of the old saying, "To a man with a hammer, everything looks like a nail."  To a man with a computer, everything looks like-a computer problem!-RAP

Dear Bob:

I read your article concerning spreadsheet mentality with great interest.  What you have observed is considerably more significant than you may realize.

The spreadsheet permeated the real estate valuation industry originally as the Ellwood Method about 30 years ago and more recently (15 years) as Discounted Cash Flow Analysis.

A fundamental mathematical formula expresses the relationship of the three variables involved when valuing an income stream:

Y = R + A, where:

Y = Discount or Yield Rate (annualized rate used to equate future income to present value); R = Capitalization Rate (rate of annual income); A = Growth Rate (rate of annual growth).

The appropriate Discount Rate for valuing an income stream equals the Capitalization Rate plus the Growth Rate.

Spreadsheet valuation programs let the user treat these three interdependent variables as if randomly independent.  Once two variables are input, the third is, of course, determined.

If the Discount Rate is mathematically understated, or the Growth Rate is mathematically overstated, the value produced by the spreadsheet is mathematically inflated.  In either case, the present value of increasing income is overstated.

Milton Friedman argues academically that the possibility of an exception (i.e. a property could outperform the market) means that the formula for valuation of an income stream is not necessarily hard and fast.

How would you like to be a stockholder in a Savings & Loan where every loan was predicated on the collateralized asset outperforming the market?  This is what happened and half the Savings & Loans are no more.

There have been those in my industry who have specialized in such distortions, which is THE reason for the Savings & Loan crash.

Your observation was particularly interesting to one who has been making a similar argument in a seemingly unrelated field for the past twenty years.

Michael D. Hartnett, Tampa, Fla.

Yeah, everybody making the SAME assumptions and using the SAME spreadsheet-right or wrong-is scary all by itself!  "Every loan must outperform the market"...Wow!-RAP

Dear Bob:

You've done it again-written about something I just have to comment on even though I'm supposed to be doing real work.

I, of course, refer to your heroic efforts to maintain charge on your camcorder battery while in the field.  Your existing solution is ingenious and reliable, but an hour's worth of whirring noise while sitting around a wilderness campfire must take a little bit of the charm out of the experience.

My thoughts are as follows:

1.  I think you dismissed the option of using a set of D cells too quickly.  Good Alkaline D cells have 14 Ah of capacity-about 3 to 4 times that of lead or Nicad.  Although we need to confirm (by testing) whether you can draw 1 A from a D cell for the whole 14 hours, it seems reasonable to expect a week of shooting at one hour a day.  A second set of cells will carry you for another week, at a total weight comparable to your generator set.  A drawback, of course, is the extra wiring from your belt pack to the camcorder.

2.  High rollers (like networks) are said to use primary lithium batteries, which should double or triple the life of alkalines.  This option will probably not appeal to your penurious nature.

3.  If you really prefer recharging the built-in battery, what about those thermoelectric generators the Soviet Army was supposed to have used?  Campfires (or lanterns) are always available on camping trips, and this would be a silent and effortless option.  I'm sorry I don't know more specifics, but the modules used for automotive coolers might well supply useful power at a reasonable voltage.

4.  If you aren't already aware, the biggest problem in field use of camcorders is the prevention of internal dust and especially internal condensation.  A "dew" episode will, at the minimum, trigger a protective shutdown that lasts an hour or more and may well drain your battery (the camera is supposed to stop and warm itself to drive off the moisture).  Any attempt to bypass this protection will result in severe head clogging and tape damage (I spent several hours with the hair dryer on a cruise ship hand drying the tape inch by inch after taking my camcorder out of an air-conditioned cabin into Gulf Stream humidity, and misinterpreting the "Dew" warning as a low battery indication that I apparently overrode by changing batteries).  While your treks don't seem to feature air conditioning, cameras carried next to perspiring bodies can suffer similar problems.  Make sure your fanny pack is sweat-proof.

5.  One more tip-are you sure you want an hour a day of footage?  Even if it's all gorgeous, consider how long it will take you to watch the final product; maybe you should be a little more judicious on the trail.  I took almost 6 hours of footage on a truly unforgettable Orient Express train tour from Paris to Istanbul in 1984.  It's all good stuff but hard to show to anybody due to its length.  You would have been impressed by my efforts.  This was the pre-camcorder era-I lugged a portable deck whose "compact" camera alone was bigger than most of today's camcorders.  The train was advetised as using authentic '20's era cars, so I suspected that ac power might not be available.  They did say that each compartment had a small fan.  I built a "universal" charger using a capacitor to limit the current at 220 V ac to 100 mA, so I could get something off any ac voltage (primarily for hotel room use).  However, I correctly surmised that the train's lighting power would be dc, because old train cars used storage batteries recharged by belt-driven generators running off the wheels.  Having thoughtfully provided myself with a meter, tools, and a dropping resistor, I determined that the fan was powered by 28 V dc and could readily be tapped at a screw terminal block, like the ones in your old VW.  I was not only able to recharge my spare battery, but was able to power my "12-24 V" 7-in. TV, so we could see the "dailies" in the comfort of our compartment.  A 16mm film crew had taken the trip earlier, and a network crew had made at least a partial trip, but I expect I was the first consumer video shooter to go the distance.

Good luck with your efforts-modern camcorders are wonderful memory-savers, although not as robust as film cameras.

Patrick H. Quilter, Vice President of Engineering, QSC Audio Products, Costa Mesa, Calif.

Thanks for the tip on alkaline cells.  I'm investigating thermoelectric generators (heavy and inefficient?).  I love your story of 28 V dc on the train!-RAP