Re: Bad coding standards

[Robert Dewar <robert_dewar@my-deja.com>]

In article <3A3E5E7E.67817482@acm.org>,
  Marin David Condic <mcondic.nospam@acm.org> wrote:
> In any case (pun intended :-) this seems to be basically a
non-issue. If you
> like all upper case, you're free to use it. If you like C++
coding
> conventions, you're free to do that too. Consistency within a
project is
> probably more important than specifically what style you
choose.


Yes, that is of course true, nevertheless it is an advantage
if everyone, or at least most folks, agree on a consistent
style, as it makes discussion easier (it has for example
been a consistent annoyance in C that there are several
conventions for the use of {}

The fact that indentation is strongly suggested by the RM
has meant that pretty much everyone uses the same indentation
style, which is an advantage.

You certainly get used to anything you use a lot. At first I
found the mixed case style of identifiers annoying (mainly
because they are not well enough distinguished in comments),
but now I am completely used to it, and find anything else
annoying, e.g. I really dislike Mike Feldman's style in
his book, and find this a disadvantage of the book, he uses

   FOR VariableRef in Typ LOOP

....

which to me has the keywords yelling at me :-)

So my recommendation is follow the RM style for keyword and
identifier capitalization unless there is good reason not to.

Robert Dewar

P.S. A neat feature of GNAT is that it learns your style for
error message purposes, so it will tell me:

  "end Try;" expected

and it will tell Mike

  "END Try;" expected

and it may tell someone else

  "end TRY;" expected






Sent via Deja.com
http://www.deja.com/