Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!gatech!akgua!akguc!mtune!mtunf!mtx5c!mtx5d!mtx5a!mat
From: mat@mtx5a.UUCP (m.terribile)
Newsgroups: net.lang.c++
Subject: Re: Re: enhancement: have C++ announce release number when it runs
Message-ID: <1439@mtx5a.UUCP>
Date: Wed, 23-Jul-86 01:10:49 EDT
Article-I.D.: mtx5a.1439
Posted: Wed Jul 23 01:10:49 1986
Date-Received: Thu, 24-Jul-86 05:08:16 EDT
References: <379@mntgfx.mntgfx.UUCP> <471@opus.nbires.UUCP>
Organization: AT&T Information Systems, Middletown, NJ 07748-4801.
Lines: 43

> > In an environment such as we have with C++, where the compiler will occasion-
> > ally be undergoing revs or intermediate patches, I would find it useful if the
> > compiler would announce a summary at its completion...
> 
> In an environment such as many of us have, I think it would be appreciated
> if the compiler would be <very> quiet unless it has something of substance to
> say.

Indeed, the UNIX philosophy, and the pleasantness of the environment, require
that noise pollution be kept to a minimum.  On the other hand, the system
ought to tell you what it's doing.

Many of the PCC and PCCII based compilers have a ``+V'' option that does
nothing but announce the release number.  This may be the cannonical
solution.  Also, cfront puts its build date into the ``..c'' that it builds,
as a comment.  Perhaps it could write the release as well?

> In UNIX, this is almost a sacred precept (not that there aren't
> blasphemers:-).  I wish it were so in other systems.  That is, if you want
> the compiler to tell you about itself (instead of doing its normal job,
> which is to compile a program), why not ask it that question?
> I'm not just being peevish.  (I AM being peevish, but not JUST so.)  It can
> be quite annoying to filter the output of a commonly-used program to get
> rid of the stuff that doesn't need to be there.  It is more reasonable to
> expect a program to generate just the output that is needed, than to expect
> it to do extra work to generate anything that MIGHT be needed and then let
> its users do extra work to get rid of the noise they don't need.

Here, here.  I wish that programs were more often seen and not heard.
That doesn't mean that there isn't room for ``verbose'' flags or ``statistics''
flags; just that these should not be presumed to be the outspit of interest
to the user when what he needs and wants is the object file.

Program designers should not push their notions of the right way to do it
onto a community that has bought into another way.
-- 

	from Mole End			Mark Terribile
		(scrape .. dig )	mtx5b!mat
					(Please mail to mtx5b!mat, NOT mtx5a!
						mat, or to mtx5a!mtx5b!mat)
					(mtx5b!mole-end!mat will also reach me)
    ,..      .,,       ,,,   ..,***_*.