Date: Mon, 10 Oct 2005 09:49:15 -0600
From: Michael Fuhr <mike@fuhr.org>
To: Greg Sabino Mullane <greg@turnstep.com>
Cc: pgsql-www@postgresql.org
Subject: Re: User survey suggestion: add CVS as a choice
Message-ID: <20051010154915.GB99013@winnie.fuhr.org>
References: <20051010143800.GA98669@winnie.fuhr.org>
	<b6f5c6edd058f377fc6a33fd04945804@biglumber.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b6f5c6edd058f377fc6a33fd04945804@biglumber.com>

On Mon, Oct 10, 2005 at 03:14:35PM -0000, Greg Sabino Mullane wrote:
> >> In either case, I think the handful of people who use cvs rule it out
> >> as an answer, IMO.
>   
> > Is CVS usage really that low?  I don't know, not having seen any
> > activity logs.
> 
> Don't know. I'm just guessing based on the approximate number of posters
> to hackers and patches. I guess it can't hurt to add it, but I still
> think it will be an outlier. Personally, I prefer to use the tarballs
> so that I can help spot any bugs that may have been introduced in the
> packaging process.

So that's one data point.  Shouldn't conclusions about the population
in general be based on survey results instead of on a priori assumptions?
Otherwise what's the point of the survey?

-- 
Michael Fuhr