John Levine <johnl@xxxxxxxxx> Sun, 06 September 2020 10:00 UTC wrote: > Messages from the ietf list for the week ending Sun Sep 6 > 06:00:00 2020 > > 3 | 31875 | Khaled Omar <eng.khaled.omar@xxxxxxxxxxx> > 2 | 6730 | Fernando Gont <fgont@xxxxxxxxxxxxxxx> > 1 | 7065 | JORDI PALET MARTINEZ > <jordi.palet@xxxxxxxxxxxxxx> > 1 | 6770 | Eliot Lear <lear@xxxxxxxxx> > 1 | 6184 | John C Klensin <john-ietf@xxxxxxx> > 1 | 5797 | Ole Jacobsen <olejacobsen@xxxxxx> > 1 | 5460 | Florian Weimer <fw@xxxxxxxxxxxxx> >... John, Given a series of recent changes, I question the utility of this measurement in the form you are making and presenting it and wonder what others think. What question are we trying to answer? Remember that, when Thomas Narten started his list, it was rare to have a discussion list in what we now know as the General Area other that this list, his original measurements would be best approximated by combining the IETF list, the Last Call list, Manycouches, Gendispatch, Mtgvenue, and perhaps some others. So, to the extent that we want to provide the same information now that the original Narten reports provided, probably all of those lists should be aggregated or reported in other ways. A second issue is what we are really trying to find out. If it to help people notice and remember when they are going on a bit too much or two often, then either a week where the maximum number of posting by one person is three is abnormal or the action is simply elsewhere than on the IETF list. Unless even more people have dropped off in the last few months than I assume, the number of people on the IETF list reaches into four digits. When only 13 people (or roles) have posted, only two people have posted more than one message, and two of the single-message postings are from role accounts and duplicates of postings to IETF-announce, if that were a regular pattern, it would be reasonable to infer either that there is nothing of interest going on in the IETF, that no one cares any more, or that there is no discussion going on, just statements. Finally, we've known this all along but need to remember it: when many messages are in response to others, raw counts of message sizes mix actual comments from the person who posted with greater or lesser quantities of quotation from earlier messages. Different of us do that differently; sometimes even one of us does it differently for different messages or threads. However, someone who wants of minimize their counts can easily do so by not quoting from prior messages at all, even to establish minimal contact. Not clear we would want to encourage that. So: What question are we trying to answer? Given the tendency toward pushing traffic onto specialized lists, what do we need to measure to get that answer? And, given those things, is doing this well worth the effort and, if not, is doing it in a way that might not be meaningful helpful or just a source of more traffic on the list? thanks, john
