home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: sparky!uunet!mole-end!mat
- From: mat@uunet.uu.net!mole-end
- Subject: Re: Renew?
- Message-ID: <1992Jul27.072223.11801@uunet.uu.net!mole-end>
- Summary: Proposals to x3j16/WG21
- Organization: :
- References: <1992Jul5.002414.390@frumious.uucp> <1992Jul25.151227.8156@hemlock.cray.com>
- Date: Mon, 27 Jul 1992 07:22:23 GMT
- Lines: 51
-
- In article <1992Jul25.151227.8156@hemlock.cray.com>, de@cray.com (Duane Eitzen) writes:
-
- > On the topic of submitting a proposal: how does one do this? How
- > much time should one expect to spend doing it? What percentage of
- > external proposals are successful?
-
- Several of the big C++/OO magazines published an article last year by
- Bjarne Stroustrup on just this topic. The spirit was `Please don't, but
- if you insist on doing it, here's what you must do to have a ghost of a
- chance.'
-
- There's a lot of work involved, including analysis of how the proposal
- would interact with other stuff. C++ is sufficiently rich (like Ben&Jerry's
- ice cream!) that this is a non-trivial task and should be undertaken by
- several people, preferably with experience in compilers and language
- semantics. (Actually, more like a `whole calamari' flavor of B&J's, if
- you get my drift.)
-
- What percentage are `successful'? If you mean `what percentage are
- accepted more or less in the form in which they were submitted?' maybe
- one in twelve. If you mean `What percentage influence the language?'
- the number might be a little higher. (Of course, the season isn't over
- yet, so the percentages aren't complete.)
-
- Proposals by people who've worked in the guts of C++ are more likely to
- address the issues squarely, and probably have a much better chance than
- a proposal written by J. Random User. (I've seen some simple ones that
- I thought worthy, but that never saw the light of day.) Proposals made
- by representatives of large user groups (e.g. an SQL standards body)
- have a better chance of commanding attention, but not necessarily a
- better chance of acceptance.
-
- My opinion on this one is that it is a non-starter. You CAN'T just go
- moving stuff around in C++; it really is tied to the Object model of
- programming and the referential attributes that represent relationships
- have to be maintained (i.e. you can't tolerate dangling pointers). To
- make this work, you have to think through not only C++ and its semantics,
- but the nature of OO programming and how OO maps into C++. It's job
- somewhere between Huge and Fu'Cryin'OutLoud Huge. You are trying to
- shoehorn a brontosaurus into a Habitrail(tm, probably) and it just won't
- go.
-
- realloc() as it stands in C is nice because you MAY be able to expand
- an array in place. But C++ tends to work free store harder, and it's
- less likely, given N. Random Storage Manager, that you'll reap the
- benefit.
- --
- (This man's opinions are his own.)
- From mole-end Mark Terribile
-
- uunet!mole-end!mat, Somewhere in Matawan, NJ
-