Programming in Oberon - M. Reiser and N. Wirth now in PDF form

[Chris Burrows]

A scanned copy of Programming in Oberon - M. Reiser and N. Wirth is now
available as a PDF download:

http://www.oberon.ethz.ch/books.html

--
Chris Burrows
CFB Software
http://www.cfbsoftware.com/gpcp

[akarl]

Post by Chris Burrows
A scanned copy of Programming in Oberon - M. Reiser and N. Wirth is now
http://www.oberon.ethz.ch/books.html

Great! So far I've only made it to p. 19. I like the sentence: "Pedantry
in programming is not a luxury, but a necessity."

August

[akarl]

Post by Chris Burrows
A scanned copy of Programming in Oberon - M. Reiser and N. Wirth is now
http://www.oberon.ethz.ch/books.html

From section 3.2.2 about the types REAL and LONGREAL:

"The real types approximate the real numbers."

To be precise, the real types are (finite) subsets of the rational
numbers. In fact, the real types cannot store a single irrational
number. In this respect I think the name `float' used in C makes more sense.


August

[Frank Copeland]

Post by akarl

Post by Chris Burrows
A scanned copy of Programming in Oberon - M. Reiser and N. Wirth is now
http://www.oberon.ethz.ch/books.html

"The real types approximate the real numbers."
To be precise, the real types are (finite) subsets of the rational
numbers. In fact, the real types cannot store a single irrational
number. In this respect I think the name `float' used in C makes more sense.
"Pedantry in programming is not a luxury, but a necessity."

However, sometimes it is just pedantry.

[chl]

Post by akarl
"The real types approximate the real numbers."
To be precise, the real types are (finite) subsets of the rational numbers. In fact, the real types cannot store a single
irrational number. In this respect I think the name `float' used in C makes more sense.

I disagree. The fact that in Oberon, real types cannot store irrational numbers excactly, is
only an implementation detail (you could, for example, count multiples of pi). On the
other hand, 'float' comes from floating point. This is a break of abstraction, because it is
not strictly required that an implementation uses floating point arithmetic to approximate the
real numers.

C's double is even better. double what?

Christian

[August Karlstrom]

Post by chl

Post by akarl
"The real types approximate the real numbers."
To be precise, the real types are (finite) subsets of the rational numbers. In fact, the real types cannot store a single
irrational number. In this respect I think the name `float' used in C makes more sense.

I disagree. The fact that in Oberon, real types cannot store irrational numbers excactly, is
only an implementation detail (you could, for example, count multiples of pi). On the
other hand, 'float' comes from floating point. This is a break of abstraction, because it is
not strictly required that an implementation uses floating point arithmetic to approximate the
real numers.

OK, that makes sense. Thanks.

August