On Fri, Jan 09, 2004 at 03:36:58PM +0100, Sven Luther wrote: > > We'd need somebody good enough at statistics to measure the error popcon > > introduces. I don't think it's in the order of the S/N ratio, though. > > Yep, but the package i maintain have 0 entries in popcon, while i > know this is not the real case. This means infinite error ratio, no ?
Uhm, the error would be over the whole archive. Just because your packages don't seem to be used by anybody havin popcon enabled don't invalidate the results. They may *for you*, but we're not discussing thing here merely for the sake of your packages, but for the whole non-free part of the archive. Michael
