I have been following the R6RS and the R7RS discussion processes since shortly after the beginning of the former. It is educational, if nothing else, and I do enjoy watching the debates, though I have seldom posted to the group. As with virtually all engineering, most decisions are less matters of things that are strictly correct or strictly incorrect (read: wrong) than they are a discussion of tradeoffs. I have little doubt that there would be a lot less heat in these debates if more issues where strictly right or strictly wrong. It would then become less a question of design and more a question of solving the problem in the straight-out method used to solve problems in mathematics. All of those posting are extremely intelligent. This is not surprising. Given the state of the industry, few blub programmers ever make it so far as to hear about Scheme, let alone care about the next standard issued under that name. Most of these people have PhDs and are doing this as part of their research.
So, I would summarize the R6+RS mailing list as being a lot of smart people arguing heatedly over design tradeoffs. At least, it keeps things from being boring. I find it interesting how dedicated these people are to sitting down and proving to the whole group that their way is obviously the best way. It may very well be, but if the majority of those standardizing Scheme do not want it, why worry? Why not take the R 4, 5, or 6 RS and draft your own spec, publishing it under your own name? Just take it and create another Lisp dialect. Show us all that it is better than Scheme or Common Lisp. It is almost like the languages world has decided that we shall have precisely two Lisps: Scheme and Common Lisp. Most of the Lisp-esque languages out there are starting from Scheme or Common Lisp and making some minimum number of tweaks (often, like Clojure, to make it run on some other platform) rather than designing a new language from the ground up. It seems to me that there is plenty of room for interesting experimentation. In fact, it seems to me that the standardization process would be a lot more fruitful if we could see a lot more Lisps out in the wild. We could take the good and avoid the bad and have a real, living model to look at instead of some airy discussions.