Thank you for your comments. I was feeling more alone than a Martian
bacterium in this universe. :)
> I'm an aerospace engineer and I would also not like to bother with
>writing my own ASCII text file I/O module, but unfortunately I see it
>as a necessary evil which I've been doing for years (ever since I
>switched from Pascal to Modula-2).
In my opinion, oberon2 should have a standard ASCII text file I/O
module. However, knowing the oberon2 history, I am sure that this goal
is not achievable anymore. I would be happy if OMI realizes that it
should provide an ASCII text file I/O module with Oberon/F.
> In fact, in order to protect myself from all the various "flavors"
>of Modula-2 (and now Oberon-2) libraries, I created my own library of
>ASCII I/O, math, string handling, etc. calls. This way, when I port
>from one compiler to another, (theoretically) most of the effort is
>spent porting the library, then everything else remains more or less
>the same.
I see your point; but if everyone decides to code his own ASCII text
file I/O module we will be protected against the possibility of
exchanging simple pieces of code (not an entirely module) that
eventually uses ASCII files. Moreover, we will multiply efforts losing
what could be loosely called as exchangeability.
Exchangeability is also lost with the current array definition.
Consider this simple example: a piece of code to calculate the norm of
a vector. You could choose to start your vectors with 0, while I could
choose to start my vectors with 1 (wasting the 0 position). Why did
Oberon2 have to imitate C concerning array definition? :(
> And although I may not like having to implement something that
>probably should have been done by the compiler writers themselves, I
>think it's probably better than having to switch to C++... :-)
It is much better than having to switch to C++, but it may be even
better to switch to Ada95. Have you ever though about it? I am
conscious that Ada is a big language, but the Oberon/F framework is also
big. Possibly, with Ada95 I would be able to concentrate on engineering
problems and not on computer science problems.
Regards,
Vinicius