Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!think.com!hsdndev!cmcl2!kramden.acf.nyu.edu!brnstnd
From: brnstnd@kramden.acf.nyu.edu (Dan Bernstein)
Newsgroups: comp.lang.misc
Subject: Re: On whether C has first-class composable functions
Message-ID: <24547:Jan822:05:4191@kramden.acf.nyu.edu>
Date: 8 Jan 91 22:05:41 GMT
References: <442@data.UUCP> <4408:Jan421:44:3391@kramden.acf.nyu.edu> <20021@yunexus.YorkU.CA>
Organization: IR
Lines: 52

In article <20021@yunexus.YorkU.CA> oz@yunexus.yorku.ca (Ozan Yigit) writes:
> In article <4408:Jan421:44:3391@kramden.acf.nyu.edu> 
> brnstnd@kramden.acf.nyu.edu (Dan Bernstein) writes:
> >In article <442@data.UUCP> kend@data.UUCP (Ken Dickey) writes:
> >> There is an important property of "first-class" objects missing here.
> >> A first-class object does not have to be named.
> >Do you have a reference? As always, I'm doing my best to use standard
> >terminology; I just haven't seen any references that demand a syntactic
> >restriction like that.
> There may be many references, though I am not sure if that will that make
> any difference whatsoever. It appears that you would prefer to argue than
> to look them up.

On the contrary. Every reference I've found either (1) doesn't define
first-class; (2) makes ``first-class'' so restrictive that nothing in C
or Ada can possibly be first-class; or (3) defines ``first-class'' as
``can be passed as an argument and [when there are variables] assigned
to variables.''

Note that Dave Gudeman posted a definition equivalent to #3.

> Here is one: William Clinger, Semantics of Scheme, February. 1988, BYTE,
>              pp. 221-227

You have this habit of posting long definitions without verifying that
they have anything to do with your point.

I asked for a reference that *defines* first-class in any way other than
#3. The only statement in what you posted that could possibly be a
definition is this one:

>    All scheme objects are first-class citizens of the language.

But the text also gives scheme objects the following ``inalienable
right'':

>    o Objects never die.

Question 1: When you posted that definition, did you believe that it
defined ``first-class''? (If not, you are not contributing to the
discussion.)

Question 2: Do you realize that if the properties you posted are taken
to define ``first-class,'' then nothing in Ada is ``first-class''?

I conjecture that Scheme advertisers have tried to corrupt the meaning
of ``first-class,'' by adding properties that make Scheme objects into
``first-class'' objects while making objects in other languages into
non-``first-class'' objects. Ozan is drawing upon the Scheme literature
alone, but ``first-class'' wasn't invented with Scheme.

---Dan