Re: Ichbiah's Letter

[Jacob Sparre Andersen <jacob@jacob-sparre.dk>]

> https://duckduckgo.com/l/?kh=-1&uddg=http%3A%2F%2Fweb.elastic.org%2F~fche%2Fmirrors%2Fold-usenet%2Fada-with-null

Reading the letter, I must say that I disagree with Ichbiah on some of
the Ada 9X features which he wanted dropped:

+ Aliased objects:

  I have more than one program I don't know how I could have implemented
  sensibly without aliased objects.

+ Protected types:

  I've heard of the term "passive tasks", but I simply don't know how I
  should have written some of my more complex tasking applications
  without using protected objects.

- Decimal types:

  Okay. - One place where I must admit that I only know of example
  sources using this.

+ Unsigned types:

  Something I use quite a lot and find it hard to imagine managing
  without.  Also, how would one interface with C et al. without unsigned
  types?

- Access parameters:

  I suspect that allowing "in out" parameters for functions is a better
  idea than having access parameters. - One more where Ichbiah may well
  be right.

- Accessibility (checks):

  It seems like this definitely is a part of the language, which has
  ended up being too complicated for anyone to understand, but I am not
  in a position to say anything sensible about how to avoid them in the
  language.

+ Tagged types and dispatching:

  Well.  Although Ichbiah puts tagged types on his list of complicated
  features, he still argues for them further down in the letter.  I
  suppose that we agree.

+ Barriers:

  One of those features I can't see how one can manage without.

+ Self-proclaimed children:

  What a wonderful term.  And a feature I have used to append
  compatiblity interfaces to existing libraries, without having to go in
  and modify the existing, stable library.

+ Separate compilation:

  It may not be needed for compilation speed, but it is an excellent way
  of managing multiple implementations for the same interface.

I am happy to have seen this letter, but overall, I think Ichbiah was
not ambitious enough in his view of what Ada could become.

I agree with Ichbiah's belief that the availability of good GUI
libraries should increase the use of Ada, but somehow it doesn't
happen. - Maybe because the GUI libraries aren't good enough.


Ichbiah (and many others) argue strongly for upward compatibility from
one version of Ada to the next.  While I can see the point, I still feel
that we as a community should be prepared to learn, and to remove the
worst mistakes we have introduced into the language.  I know that
breaking compatibility will split the development of pre- and
post-change compilers, but are there changes we can make, which might
both reduce the development cost of post-change compilers and increase
the quality of post-change software development?  Interfaces and
anonymous access types are the candidates I can remember being
critisised most, but can we remove them and still create equally good
software?


Greetings,

Jacob
-- 
"Lots of information, strong cast, but a bit weak on the narrative."
              -- Pratchet, Stewart & Cohen reviews a phone directory