Winning and Losing

FRAMINGHAM (01/31/2000) - "Winning is a habit. Unfortunately, so is losing." - Vince LombardiWinning at anything changes everything. It confers strength, validates and creates opinion. But winning at, say, basketball, is not the same as winning the Miss America Pageant (a fabulously silly competition, but work with me here folks). In basketball, winning is about who is faster, who has more skill and who is a better athlete. In short, it is definitive. Yes, I know you can argue about the referee's calls, but you have to admit that most matches are fairly judged.

But Miss America ("she's studying to be a veterinarian and wants to work for world peace and split the atom") is a silly win. It is one beauty against another, and may the best legs win. Ridiculous. What happens when Miss Rhode Island (the former Miss Plumbing) loses? She is crushed because she was judged on subjective criteria and she didn't measure up. What cruelty, what a blow (I'm actually being serious).

"Winning is not everything. It's the only thing." - Vanderbilt coach "Red" Sanders.

Now, I just read "King of the NOS hill" by John Bass and James Robinson in last week's Network World. A good solid piece of research, but I have a problem with the underlying philosophy. That philosophy, a common computer industry one, is that it is meaningful to say that one product in an area as complex as network operating systems is better than another, even though the winner is only marginally better and the judgment highly subjective.

Before any of you who know my feelings about NetWare and Windows even think of suggesting that I'm piqued because NetWare came in second to Windows 2000, let me say that I think Windows 2000 is a fine piece of code (and from what I've seen, a big improvement over previous Windows releases).

But my complaint is that these sorts of comparisons are fundamentally misleading. Even though you, dear reader, are more than smart enough to know that the authors' conclusions are biased from their experiences and that your own mileage may vary, it is still the case that when an august journal declares "a winner," one's perception is subtly - but powerfully - altered. It is hard not to take such accolades seriously even though the winning margin might have been razor thin.

For example, as my evil twin Gearhead noted last week, if you scan the press, you'll find that the opinion of most product reviewers is that it is sayonara to Windows CE and long live PalmOS. While Palm Inc. devices have garnered the lion's share of the market, I doubt whether that is because the Palm is a better machine than its competition.

Back to the NOS comparison. These kinds of reviews are good for feature cataloging, but testing always has its limits. In the case of NOSes, there is one attribute that never gets reviewed because it is so hard to evaluate in a lab: stability. Products in the lab never really appear as they do "in the wild."Take any product and test it outside of real-world conditions and what have you got? Sure, there's going to be some useful data and some valuable insights, but not true facts. For that, you need to survey the market and establish real-world performance - for new market entrants such as Windows 2000 that is an impossible goal.

So, despite Windows 2000 winning, it is patently the fact that the win, as is the case with any win where hugely complex products are being evaluated, is artificial. I have a request: Can we have product reviews that don't try to find a winner? Let's just provide enough data to make judgments, and let's let the network managers pick their own winners.Champions to nwcolumn@

Join the newsletter!

Error: Please check your email address.

More about Island

Show Comments

Market Place