> Looking back in hindsight, what would help is to have some means for the > IETF to provide a maintenance vehicle for it's products. I think there is some truth to this. The reality has at times been that some WGs get a bit out of control after they've been around a while, and getting them to deliver final product is a challenge. Rather then finish a MIB or do the boring OAM stuff that a real protocol needs, folk are much more interested in develeping all sorts of extensions and other features. As such, ADs often want to shut down the WG and make everyone go away. The challenge of creating maintenance WGs is that unless handled well, they become magnets for all the leftover and unfinshed (and often uneeded) work and random individual drafts from the original WG. And all the same players with all the same agendas are tempted to show up at the new WG. Maintenance WGs are definitely needed. And they need to be timed correctly. Wait too long, and you miss the window of opportunity during which folk are starting to implement/deploy, and you can still fix the spec. Successful maintenance WGs also need strong chairs and carefuly crafted charters that keep them from becoming nuisance magnets. The challenge in doing that I think keeps ADs from chartering such WGs in some cases when a WG really is needed. Even today, there are maintenance WGs that have issues. They can be identified by the ratio of mail/drafts/discussion to the actual importance of the problem purported to being solved. Thomas