Re: Untyped Ada?

["Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>]

On Fri, 04 Apr 2008 20:52:38 +0200, Georg Bauhaus wrote:

> On Fri, 2008-04-04 at 19:38 +0200, Dmitry A. Kazakov wrote:
> 
> [...]
> The dominant question is rather whether or not the software works profitably. 

Yes, but that is not a technical problem.

> The theory of trust as a function value is not yet fully
> established, let alone established with precision. How could
> it? 

You need a model of uncertainty for that, plus a notion of process in order
to capture temporal aspect. And an elaborated types system of Ada shines
here. Observe that otherwise, the whole argument could be reduced to the
following:

Our model is wrong. Numbers have no defined meaning. What to do? Be cool,
we will still use them with an additional advantage not to care about
accuracy of +,-,*,/ because the input data is rubbish anyway...

> Scientific, even computable, rationalization of the
> trading business? Would you want to pay for being replaced
> with a computer program? So the financial market has a
> more extensive set of rules about number handling than machine
> control I suppose.

It is an interesting issue. One of the reasons of collapse of so-called
"planned economy" was an inability to balance it, just technically,
computers were too slow that time. If somebody tried it now (without an
ideological agenda, of course), could it be possible?

>> What do you mean? In fact Ada's fixed strings handling is the best I know.
>> You should never need Unbounded_String except for rare cases, when you
>> wanted to return two strings out of one function.
> 
> Replacements of varying length are ssllllloooooooowww.

I don't remember single case where I used such replacement. For that matter
"&" has exactly same complexity. I think that the problem is how people
think about strings processing. There is a whole culture of "tokenizing"
style, which is just a wrong paradigm.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de