Hi Jean-Jacques. At Fri, 31 Jan 2014 16:12:02 +0800 (CST), Subrenat, Jean-Jacques wrote:
Hello Thomas,
it's been some time that I've been wondering, so let me ask:
- as experience shows, items outside the duties of At-Large sometimes make up a large proportion of email traffic, e.g. sending greetings on religious holidays, or commenting hotel facilities, or get-well messages. Without any mechanism to evaluate the relevance of a message, what is the purpose of such statistics?
The statistics are easy to generate, because the metrics they show do not attempt to consider content, which requires (presumably) human judgement. I (and others) find these statistics a useful data point to see how busy a list is, and whether the postings are disproportionate to a small number of people. There is also some hope that those that find themselves at the top of the list (in weeks where there is a lot of traffic) will step back and consider whether they could perhaps post less frequently, but with more conciseness. Not everyone thinks these sorts of metrics are useful. YMMV. And there are others who think they would be even more useful if they showed additional statistics (e.g., per thread messages, per topic, etc.). While others are free to further pursue such ideas, I personally have little interest in doing so.
- Just as a matter of interest, can such statistics be generated by any ALAC or At-Large member from outside the US, or who does not work in one of the larger IT industries in the US and/or the UK?
Anyone can generate them. All you need is an archive of mail messages to run them on. The scripts are use are available at http://www.cs.duke.edu/~narten/ietf/weekly-mail-stats.tgz
- Are these statistics shared with, and analyzed, by any individual/entity outside of our ALAC / At-Large community? If so, who/what?
They are public, as are all messages posted to the list. So anyone has access to them. That said, I am not aware of anyone else other than list members using it. Oh, the same stats are generated/sent to the ietf and ppml (ARIN) mailing lists. Thomas