Kevin,

>First...here is the complete sentence.
>
>If you randomly poll 485 professional developers and ask questions like "what languages/tools do you seriously work with", and "which languages do you think should have been on the list that weren't", you'll likely see results similar to these
>
>Again, I didn't say the site conducted a random poll. What I'm saying is that I strongly believe a random poll would likely yield similar results.

The site polled 485 of it's readers. It's a far jump to conclusion to say that extends to the whoel world of programmers, but you are free to draw that conclusion if you so choose.

>And the survey results shows the bias you'd expect based on the sampling of people that took the survey.
>
>How do you know that?

We don't really know bupkus about the surveyed people. Since the site has a focus on VS and VS.Net (based on the articles, advertisers and sponsors) it's a safe bet that most people going there, and by inference the ones that took the survey, are VS and VS.Net biased.

> We don't know how many used Fox (or other tools) in the past, but no longer.

Fairly irrelevant point, if someone has moved on from a language why would they want to see it wasting bandwidth on a site they are visiting to get information pertaining to their current choice of language(s)?

>But let's assume you're correct: if you went to a Fox user's group or some other Fox organization, you'd get Fox biases. Question: what (for you) would constitute a sampling that didn't have self-contained "biases".

Just about any place that doesn't have a site bias. A survey taken by ACM would be more unbiased, but even it would have some biases toward mainframe environments. PC-SIG, Dr. Dobbs?

If I walk into random rooms of 485 people in France, Russia, Germany, China and Brazil and survey the "spoken language of choice" do you think I'll get the same results? And those results can hardly be applied to the 6+ billion people of the planet.

>T-SQL is in a different category because of the specialization. T-SQL development has 2 general camps: DBAs, and application developers who work in at least one other language. T-SQL is tied to the database.

Baloney. You said the site covers SQL-Server, but there's not one SQL-Server related count on the list.

Or maybe VFP is just too specialized to be on the responders languages of interest lists.

>I studied statistics in school, and I've read "how to lie with statistics" - but thanks for the recommendation.

Then you should know better.

>Maybe you can attend an MSDN event and see how many are actively using Fox, as a % of the other tools mentioned?

I do regularly attend MSDN events, I did 2 weeks ago to see some really cool stuff coming in SQL2005.

Despite your evident belief, I'm not mono-lingual, I'd guess that I've spent more time doing C and C++ than you've spent in C# or VB.Net. I don't particularly care what language any individual developer uses. I don't particularly how many developers use language X or language Y. It's more or less a pointless argument to fight X over Y.

I didn't post in the thread to bang the drum for any particular language. I just wanted to point out some fallicies in the randomness of the survey and thus the inability to extrapolate the results.