On Friday 01 September 2006 11:14, Paul Smith wrote: > It would be interesting to know how much large are the communities of > each distribution. Can the figures be found somewhere? > Equally interesting would be where the sample was taken from - IOW what target audience is likely to have replied - and how big the sample was. I suspect that the sampling setup would have a great effect on the results, which is why statistics are rarely useful. Anne
Attachment:
pgpweotfdFi85.pgp
Description: PGP signature
