Re: Ada95 Pretty-Printers/Coding styles

["Chris Sparks (Mr. Ada)" <sparks@AISF.COM>]

Robert responded with:

> Thankyou Chris for giving a nice example of the attitude that I find
> unacceptable, and which we would not tolerate for a moment at ACT.
> It does not work at all to rely on pretty printing to solve the
> problem of divergent styles, because you still have people writing
> in their own style, and not being happy working on other people's
> code, so you get bad *code ownership* phenomena.

Well I can relate to this statement.  If a programmer isn't as
persnickedy (sp?)
as I am, then looking at their code can be a chore in itself!  The most
offensive Ada code I have seen have come from "C/C++" programmers who
are
forced to use Ada.  Yuch!

> Furthermore, style is much more than a set of mechanical rules wwhich can
> be enforced by an automaton, so you will never get realy consistent style
> (such as is achieved in the GNAT sources) by this approach.

Well you may have a point here.  I have seen your "sources" and found
them
easy to follow.  Your organization is as persnickedy as I am.  On no!
:-)

> To get a really cooperative environment, in which everyone looks at every
> one else's code and there is as little sense of code ownership as possible,
> it is essential that everyone buy into a common style. It is definitely
> possible to come close to the ideal of 100% working in unison, and it is
> a desirable goal.

We can abstract this concept a little further in that most programmers
will
code to a established well-defined (personal) style.  I don't believe
that
they would change their style continually in their software.  Once a
look-and-
feel has be attained, then most programmers stick with it for years.
Look
at the Ada 83 LRM's style.  I programmed in upper case for a long time.
Now
I can't even look at uppercased code.  I say to myself "Why is this
person
shouting?!" :-)  Anyway, your point is valid.  Maybe what I am so
uptight about
coding standards may not lie in the standards itself, but in the process
itself.

> I certainly understand Chris's attitude here, since I have run into it often
> before, and have seen situations in which companies tolerate this kind of
> insistance on personal style. Yes, you may regard my attitude's as extreme,
> so you probably would not like to come to work for ACT, but we have certainly
> found that lots of people get over their initial irritation at an unfamiliar
> style, and end up buying into it with enthusiasm (and the ACT engineers are
> all in this category!)

I am used to abiding to a company's coding standard.  I may like my own,
however,
I am paid to do what they want!  With regards to "coming to work for
ACT", I would
LOVE the opportunity to work for your company, coding styles and all.
:-)  I am a firm
believer that once we get past what is expected (code look and feel)
than we can get
down to some serious programming.  I always say that what is most
important is
the code.  If it isn't right, no amount of style is going to help it!

Later...

Chris Sparks