On Sun, Mar 20, 2016 at 10:53:52AM -0400, John C Klensin wrote: > For whatever it is worth, I share Lloyd's concern. The new > setup is fine is one is trying to skim recent discussions on a > mailing list and reasonably so to find a particular subject > thread but I question its utility for references to particular > messages, even references in later messages that one might want > to trace back. For those purposes, it would be much better to > have a URL that explicitly reflects the list name (WG lists seem > to do that, but the IETF list doesn't) and a date or date > range... The archiving built into Mailman does all this -- I can't think of any reason why *all* IETF list archives shouldn't be using it. (Note that the same archiving mechanism is capable of ingesting archives that were created with other mailing list managers and integrating them into a seamless whole.) > By the way, from a small semi-random sample, a large fraction of > the links to WG mailing lists identified at > https://datatracker.ietf.org/list/wg/ are dead. Is any effort > being made to capture those discussions for historical purposes? > Would it be worth explicitly identifying the useless links? There should be complete archives of all lists maintained "forever", whatever that works out to be. It's quite easy to set this up even for discontinued lists, again using Mailman's built-in archiving mechanism. If the goal here is "make sure that all archives of all IETF mailing lists that have ever existed and that exist now are all stored in an accessible, unified, useful format", then (a) I'm on board with that and (b) I'm willing to do the legwork. ---rsk