On 29-07-2019 14:04, Brett Schenker via mailop wrote: > Correct however that same system gives the average if the entire user > network and in this case Gabbard is well above the rest of ReturnPath's > network. That pretty much dismisses the complaint that they're somehow > being burdened or targeted.
I was not commenting on the specific case. As far as I know, there isn't anywhere legislation that states that email delivery is a public service, as snail mail is, in most of the world. Thus any public scrutiny of the operations of email providers is futile. What needs to change is the public perception of those players in the public information landscape. And this, i think, is off-topic on this mailing list. To go back to the technical issue: bias would be indeed manifest itself as a different statistic for a particular sample when compared to the whole. Unfortunately, without knowing the exact methodology, it is not possible to exclude it or quantify its effect on a particular measurement. Cheers, Dan > > On Mon, Jul 29, 2019, 4:01 PM Daniele Nicolodi <dani...@grinta.net > <mailto:dani...@grinta.net>> wrote: > > On 26-07-2019 14:38, Brett Schenker via mailop wrote: > > They've always been a bit cagey when I asked but from what I've been > > told it's opted-in individuals and data comes directly from email > > providers they have agreements with. While it's not perfect, it does > > seem to do a decent job of telling trends. > > Investigating one opaque system with data from another equivalently > opaque system does not give much confidence on the obtained results. > Your statement above seems to suggest that some statistical sampling may > be involved. If the details of the sampling are not known, bias could > affect the "measurement". This may be good enough to optimize > deliverability for some source, but I would not use it to derive > conclusions about the system under investigation. > > Cheers, > Dan > > > _______________________________________________ > mailop mailing list > mailop@mailop.org > https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop > _______________________________________________ mailop mailing list mailop@mailop.org https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
