Path: utzoo!mnetor!tmsoft!torsqnt!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!bonnie.concordia.ca!clyde.concordia.ca!nstn.ns.ca!news.cs.indiana.edu!samsung!think.com!paperboy!hsdndev!husc6!ngo From: ngo@tammy.harvard.edu (Tom Ngo) Newsgroups: comp.std.c++ Subject: SUMMARY: Responses to ~const pre-proposal, Rev 1.6 Message-ID: Date: 19 Feb 91 16:24:57 GMT Sender: news@husc6.harvard.edu Distribution: comp Organization: Harvard Chemistry Department Lines: 34 On December 14, 1990 I posted a message in preparation for submission to the ANSI committee, proposing the addition of a type-specifier "~const". The primary purpose of ~const would be to permit caching member variables. For your convenience, the original posting is included as the last part of this message. In the next few messages I summarize the reactions. This message should be treated as a sort of index to the next few. I hope people prefer many short messages, with visible headers, to one long one! I feel OK about doing this in a relatively low-traffic newsgroup. I would appreciate any further comments from those interested. If I get no comments back, I will write up the final proposal (expounding on issues that X3J16 would like to see addressed in proposals), post it to this group, and also mail it to the committee. * General reactions: positive. * Alternative ways to accomplish what ~const does: dual classes, cast-from-const. * Fate of cast-from-const: keep it but deprecate it. * Generalizations of ~const-like syntax: ~volatile, ~register, ~virtual, ~inline. * Name: maybe make it !const. * Compiler complexity: not addressed in responses. * Experience: not addressed in responses. Many thanks for the reactions, and apologies for not following up earlier. These past few weeks have been very busy! -- Tom Ngo ngo@harvard.harvard.edu 617/495-1768 lab number, leave message