> This message is being posted at the request of StatSoft management.
> Please note, the person who signed this letter was not with the company
> when I was there, and as far as I know, we have never met.
********
OK folks. Now you get to see StatSoft at work. Note the nature of the
response. They do not respond to content. They do not account for their
behavior. Instead they attack and make suppositions.
>
> **** INCLUDED TEXT BEGINS ****
>
> The tone and language of Mr. Velleman's note is unprofessional and
> regrettable; however, most surprising is the NAIVETE of its main
> premise.
I'll appeal to the net. Have I been unprofessional?
>
> (1) Does Mr. Velleman really believe that all you need to do in order
> to succeed in the software industry is to tell buyers that your product
> is free of important weaknesses and limitations which are present in the
> competing products, and this will immediately sway buyers' opinions and
> will ruin the competition (even if what you say are "half-truths,
> innuendo, and outright lies" as he alleges)?
I never claimed to believe any such thing. I merely noted that Statsoft
does tell buyers that competing products are bad and does so with pamphlets
whose tone is unprofessional and whose contents are often false. I appealed
to readers on the net to look at the materials for themselves and judge
them accordingly. As it happens, I do believe that half-truths, innuendo,
and lies can sway the opinions of buyers who have not seen the products
described, but that is really not the point. The point is that in my
opinion a professional should describe why his or her product is good. He
should not denegrate the competition's product.
>
> (2) Does Mr. Velleman really believe that StatSoft, Inc. has achieved
> its phenomenal growth rate, extremely high level of customer
> satisfaction, and its excellent reputation in the scientific community
> simply by distributing inaccurate statements about the competition?
> (e.g., see the StatSoft "Feedback from the Users" brochure for an
> authorized reprint of the results of a survey recently conducted by
> SCIENCE of a random sample of its readers regarding how StatSoft and its
> products are perceived by the scientific community.)
Of course I dont believe StatSoft has succeeded "simply" by distributing
inaccurate statements. Indeed, I think they have succeeded in spite of it.
I think it is unprofessional of them to do so.
>
> (3) Does Mr. Velleman really believe that those (in fact, probably the
> most sophisticated and critical) users of scientific software are that
> silly and completely irresponsible in forming their opinions and making
> their buying decisions?
No. Nor did I ever say such a thing. (Readers, please note the technique of
StatSoft's response. I seem to be accused of calling all of you silly and
irresponsible. Did I do any such thing? If anything, StatSoft has
criticised those of you who use non-StatSoft products for not seeing the
gross failures they claim to find in other products.) I do believe that
many users of statistics software are not sophisticated in the details of
statistical computing, and may not be knowlegeable about all of the
alternative packages available. One of the advantages of statistics
packages is that they save the user from needing to know the inner details.
However, users can be mislead by pamphlets that lie.
>
> It is hard to believe that this is really what Mr. Velleman thinks. It
> is understandable, however, that he is upset that weaknesses of his
> software are being publicized.
Ah, now comes the counter-attack. Note that StatSoft has done nothing to
respond to my simple allegation that their pamphlet is an unprofessional
negative review of their competitors containing many errors and lies.
Actually, in carrying on this discussion, I have aired many of StatSoft's
views about the supposed weaknesses of Data Desk.
>
> StatSoft, Inc. has always shown a great concern for the truthfulness and
> accuracy of all of its promotional materials.
Even if they were accurate, they would be inappropriate.
> Every comparative review
> distributed by StatSoft comes with a postage-paid FEEDBACK POSTCARD
> containing a rating form and space for comments, where the readers are
> asked to rate and comment on the fairness, and completeness (e.g.,
> possible errors, wrong emphasis) of the reviews. To encourage readers'
> responses, StatSoft, Inc. makes a charitable contribution (to a cause of
> the respondent's choice) for every returned card, regardless of its
> contents and whether it is signed or not. In this way, StatSoft, Inc.
> has collected thousands of ratings and opinions from the readers of the
> comparative review, many of whom are experienced experts in the use of
> statistics software. Also, thousands of respondents volunteered to
> authorize StatSoft to quote their opinions complete with their names and
> affiliations. The authors of those comments represent the most
> prestigious universities, research institutions, and organizations from
> around the world.
However, most of the responders are not expert in all of the software
"reviewed". This is a fine example of misusing statistics. The issue here
is not whether the readers *think* the reviews are accurate. The issue is
whether the facts are actually true. And, again, the more important issue
is whether one should try to build up one's own product by denegrating the
products of one's competitors.
>
> If Mr. Velleman is interested in knowing what other (unlike him, truly
> unbiased) readers of StatSoft reviews think about their fairness and
> completeness, we encourage him to read the brochure "Feedback from the
> Users" containing several hundred opinions (complete with names and
> affiliations of respondents) from readers who authorized StatSoft to
> publish those opinions. (Note that those eight pages of dense, 3-
> column, single-spaced quotes were received over only a 2-month period;
> if StatSoft decided to publish all quotes, the brochure would look like
> a book.)
Anyone on the net want to comment on self-selection biases?
>
> Even if this brochure, published over a year ago (and including
> responses collected over only a two-month period) represented all
> positive feedback about the comparative materials that StatSoft has ever
> received, it would still be justified to say that StatSoft's evaluations
> and opinions about the reviewed software packages are consistent with
> independent opinions and overall evaluations of a large number of
> knowledgeable software users representing a wide variety of scientific
> disciplines and affiliated with prestigious research, educational, and
> industrial organizations.
Although they differ quite markedly from the opinions published by
qualified reviewers in major publications. Those reviewers seem to think
that many of the products that StatSoft denegrates are fine, capable
products. Indeed, they often judge other products to be better than
Statistica.
> Mr. Velleman should apologize for the tone and language of his note not
> only to StatSoft but also to the numerous software users who publicly
> confirmed the accuracy and fairness of StatSoft's opinions and
> encouraged StatSoft to continue distributing those materials by
> confirming that they contribute to the "raising of software quality
> standards" (see the first scale of the rating form).
I'll appeal to the net on this one, too. Anyone out there think I owe
StatSoft an apology? Please let me know directly. I'll tabulate for the
net.
> By the way, our full-time staff members at our ASA booth were quite
> flattered to hear that they can still pass as "students."
Those full-time staff members told me that they knew nothing about the
contents of the material they were distributing (to which I have been
objecting). Sorry if I assumed that this was because they were not
full-time StatSoft folks.
-- Paul Velleman