Re: A few NG related questions

Hi John, you said...
> My thinking was that amount of traffic is a good indicator of the
> popularity of a product. I don't think anybody can dispute that ORACLE
> is a much more powerful product than SQL server. But if more people
> visit SQL server NGs than ORACLE, and if more now than before, it
> kinda makes you wonder about the trend.
> Kinda like the 'terror future market' put out by the Pentagon, well,
> sort of...
>
>
> John
>

Yes, you are almost correct. The amount of NG traffic is an indicator, though I don't think I can say it's a "good" indicator. For statistics to be of any value (as if they ever can be of value), you have to gather your statistics from a "representative sample" of the population. I'm not sure what makes us regulars hang out in NGs, but somehow I can't imagine that we're a "representative sample" of the database user population.

If I do a Google search for "Oracle" and note the number of hits, then do a search for "SQL Server" and note the hits ... can I then say with any confidence which is the most popular? I don't think so.

If I compare the number of licenses issued by Oracle to those for SQL Server, can I say which is most popular? I don't think so.

If I compare Larry Ellison's personal worth to Bill Gates's personal worth, can I draw any conclusions? Well, other than the fact that Larry doesn't seem to like to be second ... not in business nor in yacht racing.

When I was in college, I remember one of my sociology professors trying to convince me that the interviews of college women that appear in Playboy magazine were representative of the feelings of college women in general. Uh-huh. Sure.

Statistics are interesting ... as long as you don't put too much faith in them. It's a good thing we don't have any statisticians lurking in this NG or I'd be in deep ... er, trouble.

-- 
[:%s/Karsten Farrell/Oracle DBA/g]