Based on brief discussion in #tcl on IRC (27 Nov, 2011 (TZ is PST)): ===none 19:41 [stu]: colin: You can't get rid of snit because too many things use it. 19:41 [stu]: Don't be snit snotty! :p 19:42 [stu]: And some things use stooop. Wow, haven't heard that name in a while. 19:45 [CMcC%|%colin]: I can get rid of snit by rewriting them all. 19:45 [CMcC%|%colin]: Give me a week and a kg of coffee 19:46 [stu]: heh 19:46 [stu]: The plaid LANs of lice and hens. (there may be a joke in there somewhere) 19:47 [stu]: "The best ..." 19:49 [CMcC%|%colin]: what's more of a hassle is the tcllib stuff which invents its own OO system, more or less copypasta'd from other tcllib stuff. 19:49 [CMcC%|%colin]: Yes, I'm looking at you, http.tcl 19:50 [kbk]: http.tcl 's object system is ancient - it was the source of a lot of that copypasta 19:51 [kbk]: mmm, copypasta with a tasty source - give me a nice chianti with it, and an Italian salad and some crusty bread... 19:51 [stu]: The best glazed hams, with rice and peas. 19:51 [yukonbob]: there's next years' (and the year thereafter, and thereafter, and thereafter) GSoC projects. 19:52 [yukonbob]: convert ad hoc OO to New Hotness. === [APN] Any such change would mean new tcllib releases of those modules would not support older Tcl versions, certainly not 8.4- and not 8.5 either since that would require a binary extension. [DKF]: They'd probably be best off being done as new major versions given the substantial change in dependencies. <>GSoC