On Sun, Apr 01, 2001 at 08:48:28AM -0400, rpjday wrote:
> i can't figure out the point of a poll like this. any company that
> has a clue should be following trends and market shares and be able
> to figure out which OSes are doing well and which ones aren't.
Answer if the query is legit:
It's hard to figure market share on Linux, when any and everyone can build,
download for free, or multiply install a single purchased set. Actually
asking the people who install and use operating systems is a critical
adjunct to outside observation of perceived trends--kind of like actually
carrying out experiments, instead of just doing the gedankenexperiments.
> anyone who hasn't yet figured out that linux is a player deserves
> to go under.
These guys are pretty deeply buried in the Microsoft market. It looks
like this came directly from them, instead of being forwarded by a
(normal) list member. Their faith and strong inclination will be,
of course, to see Win* as the clear winner.
HOWEVER, their normal membership is going to vote overwhelmingly in
favor of any and all M$ products--well, actually, I don't even know of
Microsofties who'll vote for Win/ME. But Whistler, yeah, they're going
to say that it's the be- and end-all.
So conducting a poll like this on only their own mailing list is going to
give 'em the result they'd like--but it'd be SO single-source that nobody in
their right mind would ever consider it believable.
Pumping a request here will result in a set of votes that are just as strongly
biased in the direction of Linux, giving them the appearance of credibility.
I'm certain their confidence is that there are MORE voters available in the
Microsoft camp, so they'll "win" handily, but look much more legit.
> polls like this strike me as a way for someone to get market
> research for free by getting someone else to do it.
I don't have a problem with free market research--data's where you
find it. This, though, is absolutely untrustworthy, and a waste of
everyone's time. Worse, it may actually get play in the popular press,
legitimizing garbage data. They want a more believable poll, they should
hit unaligned lists, publications, and associations. That means don't
send a poll to members of Uniforum, MCNEs, or the RedHat list.
As you imply, if you REALLY want something meaningful, have Time'n'Talent
or some other market survey firm actually conduct a blind market survey.
> my $.02, for what it's worth.
Fair; I only threw in $.01, myself.
Cheers,
--
Dave Ihnat
[EMAIL PROTECTED]
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
