Error Testing

[Dennis Lee Bieber]

Sorry, the word "create" was poorly chosen. Would "deploy" be better?
I'm aware there's a family; what I'm trying to say is that, in Pascal
(I prefer to use a better-known language for discussion, where
possible), there's a new operator instead of =.

The point is that Algol-60 already used ":=" for that operation. It was
not new with Pascal. (Actually, Algol-58 [IAL] introduced the := )(And if
one trusts Wikipedia, the \ was added to ASCII to support some Algol
operators).

 

[David Robinow]

...

[Description of idle vs. Command Prompt]

Not quite what I was trying to ask.
I'm well aware that Idle has features that the Python interpreter on
Windows doesn't.
Idle also has features that the unix variants doesn't.
I don't use my Ubuntu much so I'm somewhat ignorant, but I wasn't
aware that the interactive interpreter on Linux had features that the
Windows version didn't. I'm curious what those features might be.

 

[Tim Chase]

I wasn't aware that the interactive interpreter on Linux had
features that the Windows version didn't. I'm curious what those
features might be.

It's mostly the benefits that come from being built with the readline
library, meaning you get

- command history (Win32 offers this, the rest not AFAIK)

- command-history searchability (control+R)

- the ability to pull down things from previous lines (alt+period in
particular)

- the ability to comment out the currently typed command without
executing it (alt+octothorpe)

- the ability to prefix text/commands with a count
(alt+number_of_times followed by the character/command)

- the ability to insert matching filenames (alt+asterisk after typing
path relative to the $CWD)

- clearing to the start/end of line (control+U/control+K)

- the ability to paste content (control+Y) previously-cut by ^U/^K

- the ability to transpose adjacent words (alt+T)

- the ability to jump forward/backward to a specified character
(control+] and control+alt+] followed by the target char) like f/F
in vi/vim

Those are just a subset of the power offered by readline when built
into Python's interpreter, none of which work (other than that first
one) on Win32's cmd.exe (or, I suppose command.com).

-tkc

 

[Albert van der Horst]

'=' stands for equality.

Pascal tried to create a new operator, := to be read "becomes", to
deal with the whole equality-vs-assignment issue. Did it really help
anything? I don't think so. Just syntactic salt. Even the comparison
isn't really mathematical - in maths, "x = y" is a statement of truth,
whereas in programming, it's a question ("is x equal to y").

This suggests that Pascal went against established practice.
This is false. FORTRAN used = and that was a mistake caused by the
language being hacked together haphazardly. Langages that where
designed (ALGOL60 SIMULA ALGOL68 Pascal ADA) all use :=.
C C++ Java C# use = for assignment because of the inertia caused
by FORTRAN. Pascal was created in a culture where it using = would
be unexpected.

Knuth uses k<-n exactly because k=n would cause confusion.

The equivalent of the dreaded
if ( x=getch())
{
}

is possible in ALGOL68 too. It is not likely to be misunderstood
because of the use of :=.

By the way, it is about the only thing that I think is wrong in
Python.

ChrisA

Groetjes Albert

 

[Chris Angelico]

This suggests that Pascal went against established practice.
This is false.

As acknowledged earlier in the thread, my "create" was a poor choice
of word. I didn't mean that Pascal was the first language to do it,
just that it's a language that did that and is likely to be reasonably
knowable. I'd rather pick examples people have a chance of recognizing
than ones they won't, if the former are available

ChrisA

 

[Neil Cerutti]

This suggests that Pascal went against established practice.
This is false. FORTRAN used = and that was a mistake caused by
the language being hacked together haphazardly.

Respectfully, the designers of FORTRAN deserve more respect than
that characterization accords.

 

[rusi]

wrote:
Respectfully, the designers of FORTRAN deserve more respect than
that characterization accords.

???

If I say: "My uncle -- a pilot -- knows more about flying planes than
the Wright brothers" am I disrespecting the Wright brothers??

The state of art shifts with time. Fortran was more pioneering than
most languages that followed -- does not mean it got everything right.

From Backus Turing award speech:

---------------------
Although I refer to conventional languages as "von Neumann languages"
to take note of their origin and style, I do not, of course, blame the
great mathematician for their complexity. In fact, some might say that
I bear some responsibility for that problem.

http://www.thocp.net/biographies/papers/backus_turingaward_lecture.pdf

 

[Neil Cerutti]

???

If I say: "My uncle -- a pilot -- knows more about flying
planes than the Wright brothers" am I disrespecting the Wright
brothers??

No, of course not.

The state of art shifts with time. Fortran was more pioneering
than most languages that followed -- does not mean it got
everything right.

They obviously couldn't have gotten everything right; they had to
work largely in a vaccuum. but in no sense were they haphazardly
throwing syntax together. They designed it as well as anyone at
the time knew how. It's stood the test of time, too. There's
probably a lot more FORTRAN in use and maintained today than
Wright Brothers airplane parts.

From Backus Turing award speech:

Although I refer to conventional languages as "von Neumann
languages" to take note of their origin and style, I do not, of
course, blame the great mathematician for their complexity. In
fact, some might say that I bear some responsibility for that
problem.

http://www.thocp.net/biographies/papers/backus_turingaward_lecture.pdf

Thanks for that reference.

 

[Denis McMahon]

If I say: "My uncle knows more about flying planes than
the Wright brothers" am I disrespecting the Wright brothers??

No, but that's not what you said.

What you said was that "the [FORTRAN] language [was] hacked together
haphazardly."

That's like me saying that when your uncle flies, he just pulls and
pushes levers and knobs randomly hoping he doesn't crash - and that would
be disrespecting your uncle.

 

[rusi]

No, but that's not what you said.

What you said was that "the [FORTRAN] language [was] hacked together
haphazardly."

I did?!

 

[Chris Angelico]

No, but that's not what you said.

What you said was that "the [FORTRAN] language [was] hacked together
haphazardly."

I did?!

No, it wasn't you said it (it was Albert, I think). But that _is_ the
line that was said, it's just not yours.

ChrisA

 

[Denis McMahon]

If I say: "My uncle knows more about flying planes than the Wright
brothers" am I disrespecting the Wright brothers??

No, but that's not what you said.
What you said was that "the [FORTRAN] language [was] hacked together
haphazardly."

I did?!

My mistake, that was what Albert said, you were simply standing up for
him.

Please s/you/he/ in the lines of my previous post quoted above, and
accept my apologies for my mistake.

 

[Mark Lawrence]

FORTRAN used = and that was a mistake caused by the
language being hacked together haphazardly.

Groetjes Albert

Yes, just imagine the massive amount of research that they had to go on,
given that Alan Turing had written *THE* paper some 20 years earlier and
it was 10 years since Tommy Flowers delivered Colussus. Mind you, maybe
there's just a tiny little bit of hindsight behind your remark

 

[rusi]

Somebody must have tough teeth, though thinking about it I recall people
eating bicycles

You just have to be sufficiently non-vegetarian
http://en.wikipedia.org/wiki/DNA_computing

 

[rusi]

My mistake, that was what Albert said, you were simply standing up for
him.

Please s/you/he/ in the lines of my previous post quoted above, and
accept my apologies for my mistake.

Heh! Chill! Yeah there is some hamming distance between the strings
"Albert van der Horst" and "Rusi Mody". But you were objecting not to the state-er but to the statement...

 

[Grant Edwards]

Actually, FORTRAN is probably responsible for more CPU cycles being
executed even today than most other languages. If you think about
the fact that most large scientific simulation codes (weather
forecasting, combustion modeling, finite-element modeling and so on),
are still FORTRAN based, and that those are the codes that occupy
multi-hundred-thousand-core petaflop class supper computers for weeks
on end, you have to respect its longevity.

IIRC, some of the scientific computing libraries I used to use daily
in Python data-crunching programs are still written in FORTRAN.

 

[Dennis Lee Bieber]

Somebody must have tough teeth, though thinking about it I recall people
eating bicycles

Was visualizing them used as plate warmers...

 

[Mark Lawrence]

Was visualizing them used as plate warmers...

Hum, water cooled mainframes, tell that to the kids...