Background:
InfoWorld is a powerful publication that claims to be "The Voice of
Client/Server in the Enterprise". Every year, InfoWorld conducts their
Readers' Survey Awards. They allow all of the readers of their print
edition to vote for products in different categories. This was done
by providing ballots with the publication. For a number of years, OS/2
continued to win Product Of The Year. In 1996 (the vote for 1995
readers' choice products), this ballot was moved to the world
wide web to
their InfoWorld Electric readers.
Once again OS/2 won Product Of The
Year. InfoWorld has told me that they believe these polls to be valid
and that they meet InfoWorld's high standards.
In 1997 (the vote for 1996 products), OS/2 won by an overwhelming
margin. Editor In Chief, Sandy Reed, chose to invalidate the poll and
explained her reasoning in this editorial.
InfoWorld then activated a
discussion forum on their web site to discuss this. I do not have an
archive of that particular forum so you'll have to trust my memory on
this next comment. Ms. Reed told us that she invalidated the poll
because the results did not correspond to market share (please remember
this when seeing the results of her new poll). While I agree Ms. Reed
had good reason to suspect the poll, I do not believe she had reason to
invalidate it. Rather than investigate the matter for "stuffing" by
looking at who voted and how often they voted, she simply considered it
invalid. She then refused to release the results to back up her claims.
InfoWorld conducted another poll
via telephone. In this poll Windows NT
cleaned up (remember the market share issue from the previous
paragraph). Ms. Reed produced this editorial
where she spent an
excessive amount of time telling us how poorly OS/2 did (unlike those
other years of polling that meet InfoWorld's high standards). InfoWorld
then made this forum
available to discuss the issue. In the forum, it
was revealed that 300 subscribers (of over 300,000) the print edition
were called. It was also revealed that any subscriber that worked for a
hardware company, worked for a software company, or considered
themselves an IT professional was disqualified. Also note, InfoWorld
Electric subscribers were not called (unless they happened to be a
subscriber to the print edition as well). As you can imagine, this
created quite a stir in the new forum (which I do have an archive of).
To date, InfoWorld has refused to release details (exact instructions
given to the polling firm and exact results received) of the new poll.
On a personal note, I believe InfoWorld has a strong Microsoft bias.
While they are quick to criticize Microsoft, they at least report on
them. On the few occasions that other platforms are reported on, they
are compared to Microsoft. As an example there were two press releases
published by InfoWorld Electric
during the week of the forum. This one
for IBM's BlueBird should be good news. However, InfoWorld takes it
upon themselves to talk about the narrow focus (which is unjustified).
They then attack it for not being Windows NT compatible. They also be
sure to note that Microsoft will be releasing Windows NT 5.0 with the
Zero Administration Kit sometime in the future. On the other hand this
press release published by InfoWorld Electric for the coming
Windows NT
5.0 Zero Administration kit makes no mention of Blue Bird. It makes no
mention of not being OS/2 32-bit compatible. It makes no mention of
it's narrow focus (NT has a smaller installed base than OS/2). A bias?
I think so.
On another note, I find it interesting that although NetScape Navigator
wins the Internet catagory (with 4.6 times as many votes), they feel
it necessary to tell us
"browsers are on the verge of
disappearing into the operating system, so surfing the World Wide
Web soon won't be any more significant than surfing your hard drive."
A very subjective (bias?) opinion thrown in to make Navigator's win
over Windows NT to be meaningless.
InfoWorld wanted Windows to come out on top no matter what.
They did it.
The Archive:
Disclaimer:
My comments are mine and mine alone. You are free to reproduce them as
you wish. The InfoWorld forum required a username and password (easily
obtained via a 30 second WWW form) to access. I don't know who owns the
rights to that text. Most entries were not made by InfoWorld employees
so I don't see how they could have rights to it. When signing up for
InfoWorld Electric,
nothing was mentioned about the secrecy of forums or
of data that you might find in there. All other text from InfoWorld did
not require a password to access so I would consider it distributable in
this context.
Unless noted otherwise, all files from from InfoWorld Electric
were
downloaded directly from InfoWorld Electric.
These files were modified
to the extent to make the links between forum entries work (they were
originally run through a CGI parser). The filenames were also changed,
but the numbers in the filenames were not changed.
The forum archive is valid from when it was opened to the afternoon of
June 20, 1997 (about a week). I would have done it later, but sometime
that evening, several forum entries were lost. Since the forum was
dying out, I felt it better to keep the earlier one.