home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.misc
- Path: sparky!uunet!walter!att-out!pacbell.com!ames!sun-barr!cs.utexas.edu!hellgate.utah.edu!lanl!cochiti.lanl.gov!jlg
- From: jlg@cochiti.lanl.gov (J. Giles)
- Subject: Re: Pointers
- Message-ID: <1992Nov12.203014.11596@newshost.lanl.gov>
- Sender: news@newshost.lanl.gov
- Organization: Los Alamos National Laboratory
- References: <BxJzzv.4H7@mentor.cc.purdue.edu> <721539019@sheol.UUCP>
- Date: Thu, 12 Nov 1992 20:30:14 GMT
- Lines: 14
-
- If the only things you can do with a pointer are to dereference it,
- pass it as a procedure argument, or assign it from another of the same
- type, then you never needed the pointer to begin with. What you needed
- was for the data type to which it points to be a first-class type.
-
- Dr. Rubin's use of pointers to functions is exactly of that form.
- He can do nothing with them except dereference, pass, or assign them.
- He would actually get *greater* capability if the language he used
- allowed `function' to be a first-class data type with assignment,
- partial application (currying), and composition as it's defined
- operations.
-
- --
- J. Giles
-