Announcing new scripting/prototyping language

[Sidney Cadot]

I find C harder to read than Java, and easier to read than C++. I
don't know about Ada, as I've never seen it. But C and C++ are a
breeze compared to the likes of Perl, Python and Lisp.

<ot>

You really need to take another look at Python code. I cannot believe
anyone would put this together with Perl regarding difficult syntax.
I've seen quite a number of languages, and Python beats all of them for
clarity. Just my 5 cents.

Best regards,

Sidney

 

[Martin Krischik]

Nothing inherently wrong with it, except that few on the groups you've
posted this to will care. Try being topical next time.


When I need a language that's easy to use and readable, I turn
directly to C. Don't let me stop you, though.

C is the most difficult Language to read and use. And yes, I have 10 years+
experience in C and C++ so I know what I am talking about. After a few
years of C and C++ it became clear to me that all those great K&R invetions
(array = pointer, type convertions left right and center - to name only
two) are crap.

In C and C++ I spend 10 times as much time inside a debugger as for example
in Ada. And I have only 1 year experience in Ada which shows something
about the quality of C or C++.

As for the OP: Scripting in a C style language: No thank you.

With Regards

Martin.

 

[Martin Krischik]

'C' is a right-regular language, with good orthogonality of
operators. That literally means "easy to read" - or at least
having the capacity to be writrten in a fashion that is easy
to read.

I have not yet seen a C programmer who does actualy does that.

If you've ever seen any deep Perl code, you'd understand

If you have ever seen Ada you know what we mean. C on the "easy to read" as
far away from Ada as Peal is from C.

With Regards

Martin

 

[Dan Pop]

I find Perl particularly obtuse. Perl reminds me of those comics where
obscene language is replaced by a string of random puctuation characters.

It is perfectly possible to write Perl code that is as readable as
well written C code. And it is perfectly possible to write C code that
is as unreadable as badly written Perl code.

It's not the tool, it's the way it's used. Unfortunately, many Perl
tutorials encourage conciseness at the expense of readability. I started
to like Perl only when I realised that I don't have to follow the
tutorial's examples in my own coding.

Frankly, I don't understand how Perl has caught on when there are more
powerful, and more readable, string processing languages available.

Probably because Perl is a lot more than a string processing language and
has an excellent interface to the OS primitives. The C-derived syntax
makes it easy to learn to the C programmer, which is another big point,
exploited by many other languages which wanted to benefit from C's
popularity.

Dan

 

[Frank J. Lhota]

I find C particularly obtuse. C reminds me of those comics where
obscene language is replaced by a string of random puctuation characters.
Frankly, I don't understand how C has caught on when there are more
powerful, and more readable, processing languages available.

You may be trying to be funny, but many would agree with this assessment!

 

[Jeff Schwab]

You may be trying to be funny, but many would agree with this assessment!

The first line you clipped specifically says "Not trying to be funny."

 

[Frank J. Lhota]

It is perfectly possible to write Perl code that is as readable as
well written C code. And it is perfectly possible to write C code that
is as unreadable as badly written Perl code.

It's not the tool, it's the way it's used. Unfortunately, many Perl
tutorials encourage conciseness at the expense of readability. I started
to like Perl only when I realised that I don't have to follow the
tutorial's examples in my own coding.

Good point. Even APL can be written in a readable fashion. (After all, APL
started as a documentation language!) A lot of the readability problems are
due more to the culture around programming languages, rather than the
limitations of the languages itself. One wonders how much obfuscated C might
have been avoided if introductory C texts did not present the following as
an acceptable way to copy a null-terminated string:

while( *s++ = *t++ )
;

 

[Martijn Lievaart]

The first line you clipped specifically says "Not trying to be funny."

Obviously there was a certain amount of tongue in cheek, but the point is
real enough.

I find that different languages fit different needs. The choice of
language is often dictated by how much experience one/the team has with
languages. I would never choose ada, simply because I don't know it at all.

Language comparisons can be valid, but there are surprisingly few people
who can say anything solid about that. It quickly degenerates in, well not
necessarely a flame war, but some religion is there always.

That said, seeing an interesting cross of groups, let me make a bold
statement to see if I can sparkle some real discussion, at the risk of
starting another hijad: C++ is an accademic experiment which has thought
us much about computer languages and has grealy influenced how we think
about programming languages today. Unfortunately the experiment has
failed, because the language it produced is so full of subteleties one has
to be an expert to use it effectively. On the other hand, its one of the
best languages around because its multi-paradigm basics make it often a
very good solution for the problem at hand. In particular its support for
generic programming has advanced the state of programming significantly.

To clarify that last remark, there may be languages that do generics
better (I don't know them, but they are bound to exist), but those are not
mainstream.

M4

 

[Martin Krischik]

I find that different languages fit different needs. The choice of
language is often dictated by how much experience one/the team has with
languages. I would never choose ada, simply because I don't know it at
all.

Try to learn it. After 10 years of C and C++ I wanted someting new. I
learned Ada and hey wow I wish I had done earlier.

To clarify that last remark, there may be languages that do generics
better (I don't know them, but they are bound to exist),

Well then let me enlighten you: Ada generics are better. I could explain it
to you but it would be a rather large Article and I am unshure if anybody
is truly interested.

but those are not
mainstream.

Well Boing, Airbus, NASA - your are right, not Mainstream.

With Regards

Martin

 

[Y2KYZFR1]

Richard Heathfield <[email protected]> scribbled the following




I find C harder to read than Java, and easier to read than C++. I
don't know about Ada, as I've never seen it. But C and C++ are a
breeze compared to the likes of Perl, Python and Lisp. They look like
C looked while I was still a BASIC programmer: a random jumble of
punctuation marks.

you have obviously never even seen a Python program, since it does not
use any punction marks. It ( and Lua ) are about as close to
executable pseudo-code as you can get. Python ( and Lua ) is
"comprehendable" by anyone that can read english for the most part. C
and C++ are not.

Python enforces rules about formatting even, so everyones Python code
HAS to conform to the same formatting rules, thus contributing to even
more "readibility"

 

[Joona I Palaste]

Y2KYZFR1 <[email protected]> scribbled the following

you have obviously never even seen a Python program, since it does not
use any punction marks. It ( and Lua ) are about as close to
executable pseudo-code as you can get. Python ( and Lua ) is
"comprehendable" by anyone that can read english for the most part. C
and C++ are not.

Python enforces rules about formatting even, so everyones Python code
HAS to conform to the same formatting rules, thus contributing to even
more "readibility"

You might be right about that. I know neither Perl nor Python, but I
have looked at both - and while Perl looks like a random jumble of
punctuation marks, Python is actually more readable. Not that I could
write a Python program to do anything beyond "Hello world", though -
but I think I could if I took the trouble to learn it.

 

[Les Cargill]

you have obviously never even seen a Python program, since it does not
use any punction marks. It ( and Lua ) are about as close to
executable pseudo-code as you can get. Python ( and Lua ) is
"comprehendable" by anyone that can read english for the most part. C
and C++ are not.

Python enforces rules about formatting even, so everyones Python code
HAS to conform to the same formatting rules, thus contributing to even
more "readibility"

This brings in question the meaning of "readability". I'd say
readability is for those practiced in the language, not "civilians".

 

[Warren W. Gay VE3WWG]

....
you have obviously never even seen a Python program, since it does not
use any punction marks. It ( and Lua ) are about as close to
executable pseudo-code as you can get. Python ( and Lua ) is
"comprehendable" by anyone that can read english for the most part. C
and C++ are not.

Python enforces rules about formatting even, so everyones Python code
HAS to conform to the same formatting rules, thus contributing to even
more "readibility"

IMHO, Python made a mistake there. Rather than dictate
the input format, they should have accepted a variety
of freeform input, but instead provided a pretty print
capability to produce "conforming format". This way,
the person/editor combination is not punished for doing
things a bit different. I do believe that a uniform
format is nice: but take freeform input, and produce
the rigid format as output. Ie. let the machines do
the work ;-)

In the past, _some_ BASIC interpreters worked this way.
It was forgiving about the format upon entry, but when
you listed it, the format was converted from tokens back
to ASCII in a uniform and consistent manner.

Just my 0.02 worth in this now academic post ;-)

 

[Richard Riehle]

Possible... "readable", then "easy to read", then "possible
to read". Some of Chinese ideographic content is easy to remember if
you learn to distinguish the pictures. Doesn't mean that Chinese
writing is easy to understand, e.g. because of context dependence.
Now what does "read" mean?

Actually, there is a system to Chinese characters. While it is true that
the meaning of some characters is lost even to many educated Chinese,
that is no different than the use of some obscure word
on some English readers. Consider the word, omphaloskepsis. A native
speaker of Greek will get a sense of the meaning of the word, but may
not know how it is used in English.

Chinese characters are made up of a small set of elementary [brush]
strokes. These are combined into a comprehensible set (about 300)
of radicals. The radicals, in turn, are used to compose the more
complicated characters. In some ways, written Chinese is more
object-oriented languages built over a discrete alphabet.

To read a newspaper in Chinese, one needs to know about 1800 to
2000 characters. This is well within the range of the vocabulary of
people using languages built over an alphabet. An educated Chinese
will know a lot more than 2000 characters.

As to readability of C versus other languages, my experience, in the
classroom, is that students shown a long program in C and the same
program in Ada, will nearly always find the Ada program easier to
read when they have never seen either language before.

Richard Riehle

 

[Richard Riehle]

Larry Hazel wrote: ..

'C' is a right-regular language, with good orthogonality of
operators. That literally means "easy to read" - or at least
having the capacity to be writrten in a fashion that is easy
to read.

Easy to read may not be the same as easy to understand. My
criteria is focused more on understandability than simple readability.
I am often confronted with mathematical formulations that use
a set of symbols, or a combination of symbols, that I can easily
read, but which require study and hard work to actually understand.

Small C programs are relatively easy to understand. My problem
with the language is that understandability does not scale well as
program grow. There seems to be a rapid loss of easily accessible
meaning in C programs of any serious size.

I don't have this same problem with Ada. That is one of the reasons
I prefer Ada over C, even over C++, for larger programs. As an
Ada program grows, its understandability does not suffer as much
as does a corresponding program in C.

Dr. Robert Dewar, of NYU, has often made the point that Ada is
more readable than writeable. It has so often been my experience
that C programmers find it annoying to write := instead of = for
assignment. And C programmers miss compound assignment
operators. While these features of the language, along with many
more, make it easier for some programmers to lay down code
quickly, they do not make that code easier to read, even by other
C programmers.

From my perspective, the entire C family of languages, with the exception
of C#, has suffered, in terms of understandability, from the persistence
of original C syntax, and the failure to tidy up understandability has they
evolve from the mother tongue.

When I am in the classroom, for students who have never seen much of C
or Ada, programs written in Ada are almost always easier for them to
understand, on first reading, than programs written in any of the C family.

Richard Riehle

 

[Warren W. Gay VE3WWG]

Easy to read may not be the same as easy to understand. My
criteria is focused more on understandability than simple readability.
I am often confronted with mathematical formulations that use
a set of symbols, or a combination of symbols, that I can easily
read, but which require study and hard work to actually understand. ....
Richard Riehle

The frequent use of C macros doesn't help either. They constantly
hide what is really being compiled, so you are frequently
inspecting #include files or CPP output to see what is really
going on.