[log in to unmask] quoted and then wrote: >From: [log in to unmask] (Geoff Bull) >Sender: [log in to unmask] (Team Ada: Ada Advocacy Issues (83 & 95)) >Reply-to: [log in to unmask] (Geoff Bull) >To: [log in to unmask] >"Ann S. Brandon" wrote: > >> Yes, that's why it's called a "poll:" the researcher polls a random >> sample of, in this case, embedded systems programmers. > >But the response to an internet poll is nothing like a random sample. > > >> Embedded Systems Programming should've >> safeguarded against multiple voting. > >They were pretty naive. Although the term "multiple voting" has been used by many people, I believe the _real_ action which raises resentment is "Ada fans telling their friends". The only way to avoid that is a random sample, with adequate "margin of error" calculations based on the percentage of "no response". An internet poll with safeguards against multiple votes may not have changed much. Do we have any indication that ESP did not protect against multiple votes ?