Re: Someone else's similar questionnaire

[Roger Dingledine <arma@mit.edu>]

In message <36091E17.A82BC1AD@csrlink.net>, dloss@csrlink.net writes:
>You may already know about this, but I just saw it on
>comp.os.linux.setup.  This URL is to a survey by a marketing research
>group which is trying to profile the on-line linux community:
>
><http://www.webcmo.com/mis/S967328.HTML

This is a good survey. It's not focused on what we're focused on,
but we can take some good lessons from their demographics section.

In particular, we might put at least half our demographics at the end.
Get the neat questions at the beginning.

Also, I think we should consider avoiding the "which magazines/websites/
ftpsites/whatever do you go to?" types of questions. They will get us
more information about the linux user community, but apparently there are
already surveys out there that are doing that. We could reduce the number
of questions we ask by focusing on "what do you want your computer to *do*",
which really is quite different from "do you read slashdot?"

A valid answer is "but if we ask those questions all in the same place, then
we can correlate things better, and also verify their results." But I think
we should quit trying to be too ambitious, at least for right now.

(Argue with me! :)

--Roger