TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy

TEAM-ADA@LISTSERV.ACM.ORG

Options: Use Classic View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
"Robert I. Eachus" <[log in to unmask]>
Fri, 9 Oct 1998 11:01:37 -0400
text/plain (49 lines)
At 09:20 PM 10/8/98 -0700, Gene Ouye wrote:

>So I'd say the "con" is: Assignment that violates the type system requires
>more typing.

  Agreed. To avoid the possibility of confusion, I would state it as "the
code required to violate strong typing is very verbose.  (Of course, this
is intentional.)
>Isn't this "con" really: Brittle architectures with high degrees of
>inter-module coupling can't be modified without excessive recompilation?
>
>Which then becomes the "pro": Excessive recompilation is a potential early
>indicator of architectural problems?

    Amen!  In my job, one of the strongest advantages of Ada.  Estimating
the time that any C project will spend in integration is a crap shoot.  But
on Ada projects, those problems have usually been fixed (sometimes with a
few months of schedule slip) before the start of integration.

>Yeah, I know, somebody can always come up with examples of necessary global
>types.  But most of the time I see somebody complaining about the high
>recompilation necessitated by Ada, the problem is bad design.  And if it
>really is necessary, they could always use an incremental compiler...   ;-)

     Our experience has always been that it isn't the global types that are
the main problem, it is the global constants that aren't.  I remember one
programmer heatedly explaining that a certain system parameter was a
constant not a variable, even after I pointed out that it had changed four
times in the last three weeks.

>The only "con" I can see with this is: Hackers are uncomfortable working in
>software development shops that place a high value on programming languages
>which support software engineering.

      I phrase it as, "Some people are very uncomfortable with a language
which complains about inadequacies in their design."   I threw together a
quick simulation yesterday.  It did what I wanted, but, because I wrote it
in Ada, it was overengineered for the purpose.  But then again, in C, I
could have thrown something together in under a day that took me an hour
and a half in Ada.  ;-)
(That's not really a knock on C, Ada is a better suited for discrete event
simulation.)

                                        Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

ATOM RSS1 RSS2