[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Issue: COMMON-TYPE (version 1)
- To: CL-Cleanup@sail.stanford.edu
- Subject: Issue: COMMON-TYPE (version 1)
- From: David A. Moon <Moon@STONY-BROOK.SCRC.Symbolics.COM>
- Date: Mon, 20 Mar 89 11:24 EST
- Cc: chapman%aitg.DEC@decwrl.dec.com
This issue came up while reviewing section 2.2 of the draft standard.
Does anyone object if I mail this to X3J13 and bring it up at the
March meeting? I couldn't find any sign that it has already been addressed.
References: CLtL p.35, p.76
Edit history: Version 1, 20-Mar-89, by Moon
The type COMMON is defined in a very peculiar way and does not seem to
be useful for anything. It can be extended by users using DEFSTRUCT,
but not DEFTYPE nor DEFCLASS, but it cannot be extended by implementations.
Whether certain types such as NUMBER and ARRAY are subtypes of COMMON
is implementation-dependent. The goal of having the COMMON type was
probably to improve portability, but it is unclear how it could actually
be used in that way.
Remove COMMON and COMMONP from the language.
Keeping the definition of COMMON accurate in the new specification, in
the face of changes elsewhere in the language such as the introduction
of CLOS and the possible introduction of character registries, is
difficult when no one is sure what COMMON is for. If no one uses COMMON,
it would be less work to just get rid of it.
Every implementation probably implements COMMON. Moon has never seen it
used except in a program to test whether its implementation matched CLtL.
Cost to Implementors:
Cost to Users:
None unless they are actually using COMMON.
Cost of non-adoption:
Implementors would have to maintain COMMON. Users would have to try to
understand it, or figure out that they didn't care about it.