[OT] Programming Languages

[Simon Brooke]

in message <BE7AC9A6.DABC%[email protected]>, David Martin
('[email protected]') wrote:

> You can get real databases such as PostgreSQL for free on Windows (or
> Linux, or Mac). languages: Perl, Python, Java, PHP all available for
> free on Windows. All will do CGI (web) stuff and database bits.

Another endorsement of Postgres and Java. Those two, taken together,
have been paying my mortgage (and buying my bikes and other toys) since
1998 (Postgres for longer than that), and both just keep getting
better. If you want to play with something new, though, for
non-commercial reasons, Ruby looks interesting.

> I will admit to not really having done much with these on Windows,

Wise man. However, with Java it doesn't really matter.

OH, and the AI .sig generator hits the nail on the head again...

--
[email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/
Windows 95:
You, you, you! You make a grown man cry...
M. Jagger/K. Richards

 

[Andrew Templeman]

Not Responding <[email protected]> wrote:

> > If you want a full on IDE for C, C++ or Java then consider buying a Mac.
> > You can buy a complete Mac for the price Microsoft charges just for it's
> > compiler environment and Apple gives it's tools away for free and
> > doesn't charge for membership of it's developer network unless you want
> > pre-release OSs - http://www.apple.com/xcode/
>
> Eh? Run that by me again. I can buy a Mac and it comes with what
> exactly? An IDE, compilers? What languages?
>
> Sounds promising, tell me more.

IDE is called XCode - look at the above link. Compilers for C, C++,
objective-C, Java and assembler.

plus all the tools you might get on Linux like perl, python, ruby

 

[Tim Woodall]

On Wed, 06 Apr 2005 22:33:25 GMT,
garryb59 <> wrote:
>
>>My question is, where do I start?
>
> Depends what you want to do I guess. For me, I like number crunching,
> always have done, in various absurd ways, and for me, nothing is
> better than Fortran....espeically now in the hands of Intel. Love it,
> it really kicks numerical ass. But there again, what do I know, just a
> poxy manual worker, no, place here dude
>
Once C99 becomes common and compilers actually use restrict[1] for
optimizing, Fortran might start to be replaced in the serious number
crunching space.

OTOH, because there are so many Fortran libraries that have been
optimized to extreme, it's just as likely that it will still be being
used in 50 years time.

The language that would replace Fortran (and C) in the number
crunching space would be a really good language for expressing
parallelizable algorithms. But it doesn't exist yet.

Tim.

[1] most people still don't really use const properly despite
the fact that C++ in particular really depends on const correct code.
restrict is probably even harder to retrofit to existing code than
const and will open up a new world of optimization level dependent bugs.

--
God said, "div D = rho, div B = 0, curl E = - @B/@t, curl H = J + @D/@t,"
and there was light.

http://tjw.hn.org/ http://www.locofungus.btinternet.co.uk/

 

[Tim Woodall]

On Thu, 07 Apr 2005 08:03:18 +0100,
Keith Willoughby <[email protected]> wrote:
> Not Responding <[email protected]> writes:
>
>> My question is, where do I start? I'm so out of date I can't even ask
>> the right questions. I used to write in Pascal, C, C++ and, once, Visual
>> Basic. That was all mainly under DOS (or for embedded systems). What do
>> I want to do? I'd like to be able to tinker with databases, use the
>> Windows GUI with ease and pull stuff off the internet to manipulate.
>>
>> Any ideas?
>
> Python, Perl, or (my choice, but it's not exactly mainstream) Ruby.
>
> Ruby is, IMO, the most elegant of the three, and Perl the ugliest, but
> Perl has a better range of libraries.
>
Nobody has mentioned Lisp yet. It's on my must learn list but i'm short
of round tuits ATM.

Tim.

--
God said, "div D = rho, div B = 0, curl E = - @B/@t, curl H = J + @D/@t,"
and there was light.

http://tjw.hn.org/ http://www.locofungus.btinternet.co.uk/

 

[Not Responding]

Not Responding wrote:

> Any ideas?

Thanks for all the pointers; I'll let you know how I get on.

 

[Jon Senior]

Tim Woodall wrote:
> Nobody has mentioned Lisp yet. It's on my must learn list but i'm short
> of round tuits ATM.

After you've got the hang of it, learn Prolog. Then try rubbing raw
chilli straight into your eyes. You'll find the effect to be remarkably
similar.

Jon

 

[Just zis Guy, you know?]

On Thu, 7 Apr 2005 11:41:52 +0100, "Dave Larrington"
<[email protected]> wrote in message
<[email protected]>:

>Doesn't really matter. /Real/ programmers can, as I'm sure we're all aware,
>write FORTRAN programs in any language...

Tee-hee


Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken

 

[Simon Brooke]

in message <[email protected]>, Tim
Woodall ('[email protected]') wrote:

> On Thu, 07 Apr 2005 08:03:18 +0100,
> Keith Willoughby <[email protected]> wrote:
>> Not Responding <[email protected]> writes:
>>
>>> My question is, where do I start? I'm so out of date I can't even
>>> ask the right questions. I used to write in Pascal, C, C++ and,
>>> once, Visual Basic. That was all mainly under DOS (or for embedded
>>> systems). What do I want to do? I'd like to be able to tinker with
>>> databases, use the Windows GUI with ease and pull stuff off the
>>> internet to manipulate.
>>>
>>> Any ideas?
>>
>> Python, Perl, or (my choice, but it's not exactly mainstream) Ruby.
>>
>> Ruby is, IMO, the most elegant of the three, and Perl the ugliest,
>> but Perl has a better range of libraries.
>>
> Nobody has mentioned Lisp yet. It's on my must learn list but i'm
> short of round tuits ATM.

Ah, LISP.

What can I say? It was my first language, and I loved it. I still think
it's the best and most powerful programming language. I served on the
BSI committee on the standardisation of LISP as we fought the rearguard
action against Common LISP. Common LISP is a horrible abortion -
FORTRAN with brackets - and has really killed the whole LISP thing.

If you want to play with LISP, play with Scheme. No, Scheme isn't really
LISP; lexical closures may be elegant and efficient but they don't have
the expressive power of deep binding. But Common LISP has separate
namespaces for data and functions, and is very badly adapted for
in-core develoment (the basic unit of compilation is the /file/, for
heaven's sake. What place has /file/ got in a LISP world?) so it isn't
really LISP at all. It's just a sad, ugly, sorry, mess.

This isn't to say that earlier LISPs don't have problems. They do. The
chief of these are the issue of binding and its consequences for the
semantics of interpreted and compiled code, and the issue, in big
systems, of namespace collisions. But all the hardware issues that made
big LISP systems problematic have largely been solved by cheaper
hardware, and to have (for example) InterLISP-D on modern hardware
would be an awesome development environment. OK, yes, InterLISP is also
a LISP2, but it is optimised for in-core development and the structure
editor (DEDIT) was just so much nicer than any other code editor I've
ever used.

--
[email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/
"The result is a language that... not even its mother could
love. Like the camel, Common Lisp is a horse designed by
committee. Camels do have their uses."
;; Scott Fahlman, 7 March 1995

 

[Dave Larrington]

Just zis Guy, you know? wrote:
> On Thu, 7 Apr 2005 11:41:52 +0100, "Dave Larrington"
> <[email protected]> wrote in message
> <[email protected]>:
>
>> Doesn't really matter. /Real/ programmers can, as I'm sure we're
>> all aware, write FORTRAN programs in any language...
>
> Tee-hee

You're surely not telling us you've not read "Real Programmers Don't Use
Pascal", Guy?

<URL: http://www.pbm.com/~lindahl/real.programmers.html>

--

Dave Larrington - http://www.legslarry.beerdrinkers.co.uk/
World Domination?
Just find a world that's into that kind of thing, then chain to the
floor and walk up and down on it in high heels. (Mr. Sunshine)

 

[Nick Kew]

Jon Senior wrote:
> Tim Woodall wrote:
>
>> Nobody has mentioned Lisp yet. It's on my must learn list but i'm short
>> of round tuits ATM.

Good fun, but flawed IMO. It starts with a philosophy (functional
programming with no side-effects), then fills up with exceptions
to its own rule. And would you want to use a language one of whose
most important functions is "car"?

Also it's one of those languages - like BASIC or SQL - having lots
of mututally-incompatible variants.

> After you've got the hang of it, learn Prolog.

So you can get a feel for 1980s "Expert Systems"? IMO those were a
waste of time then, and any improvement since has come from things
like Bayesian decision-making, that take you out of prolog-land.

BTW, everyone knows you can write FORTRAN in any language. There
are some converses, like, you can write any language in C.
For example, look up "comma list" in the C documentation,
and you can start writing lisp.

--
Nick Kew

 

[bugbear]

Just zis Guy, you know? wrote:
> On Thu, 7 Apr 2005 11:41:52 +0100, "Dave Larrington"
> <[email protected]> wrote in message
> <[email protected]>:
>
>
>>Doesn't really matter. /Real/ programmers can, as I'm sure we're all aware,
>>write FORTRAN programs in any language...
>
>
> Tee-hee

And bad programmers *do* write FORTRAN in any language,
even when they don't mean to.

BugBear (who's seen spaghetti code in Java, which doesn't
even *have* a goto)

 

[Simon Brooke]

in message <[email protected]>, Nick Kew
('[email protected]') wrote:

> Jon Senior wrote:
>> Tim Woodall wrote:
>>
>>> Nobody has mentioned Lisp yet. It's on my must learn list but i'm
>>> short of round tuits ATM.
>
> Good fun, but flawed IMO. It starts with a philosophy (functional
> programming with no side-effects), then fills up with exceptions
> to its own rule. And would you want to use a language one of whose
> most important functions is "car"?

Why on earth not?

--
[email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/

my other car is #<Subr-Car: #5d480>
;; This joke is not funny in emacs.

 

[Daniel Barlow]

Simon Brooke <[email protected]> writes:

> What can I say? It was my first language, and I loved it. I still think
> it's the best and most powerful programming language. I served on the
> BSI committee on the standardisation of LISP as we fought the rearguard
> action against Common LISP. Common LISP is a horrible abortion -
> FORTRAN with brackets - and has really killed the whole LISP thing.

Strange that this discussion should be happening now. From another
newsgroup, earlier today:

Thanks, I imagine, in large part to the great word of mouth
c.l.l'ers and the Yaboos of #lisp have given it, _Practical Common
Lisp_ is selling like hotcakes. At the moment (they change every
hour) its Amazon Sales Rank is 2,499 which makes it not only the #1
Lisp best seller but also #12 in the Languages & Tools category
(the next level up), and #57 in Programming, and within striking
distance of cracking the Top 100 Computer & Internet books.

http://groups-beta.google.com/group/comp.lang.lisp/msg/2a9ac0a562f9742a

<[email protected]>

http://www.gigamonkeys.com/book/ # full text also available here

I freely admit I'm biased, but it's really not as dead a language as
it looked a few years ago


-dan

 

[Tom Anderson]

On Thu, 7 Apr 2005, David Martin wrote:

> On 7/4/05 12:25 am, in article [email protected], "Pete Bentley"
> <[email protected]> wrote:
>
> > In article <[email protected]>,
> >> My question is, where do I start?
> >
> > FORTRAN 77. You know it makes sense.
>
> ITYM
>
> FORTRAN 77. YOU KNOW IT MAKES SENSE.

Actually, ITYM:

FORTRAN 77. YOU KNOW IT MAKES SENSE.

Column 7, man, column 7!

tom

--
Got a revolution behind my eyes - We got to get up and organise

 

[Tom Anderson]

On Thu, 7 Apr 2005, Tim Woodall wrote:

> On Wed, 06 Apr 2005 22:33:25 GMT,
> garryb59 <> wrote:
> >
> >>My question is, where do I start?
> >
> > Depends what you want to do I guess. For me, I like number crunching,
> > always have done, in various absurd ways, and for me, nothing is
> > better than Fortran....espeically now in the hands of Intel. Love it,
> > it really kicks numerical ass. But there again, what do I know, just a
> > poxy manual worker, no, place here dude
>
> Once C99 becomes common and compilers actually use restrict[1] for
> optimizing, Fortran might start to be replaced in the serious number
> crunching space.
>
> OTOH, because there are so many Fortran libraries that have been
> optimized to extreme, it's just as likely that it will still be being
> used in 50 years time.

Seems horribly likely!

> The language that would replace Fortran (and C) in the number crunching
> space would be a really good language for expressing parallelizable
> algorithms. But it doesn't exist yet.

High Performance Fortran?

tom

--
Got a revolution behind my eyes - We got to get up and organise

 

[Ian]

On Wed, 06 Apr 2005 19:39:32 +0100, Not Responding
<[email protected]> wrote:

>Any ideas?
If you were comfortable with c/c++, then I would recommend Java.
You'll pick it up very quickly, some very nice touches, oh, and the
odd GOTCHA! - but no more than any other prog. language.

Big plus is that it will cost you nothing (but your time and effort)
to develop with it. There are more free compilers/IDEs/Libraries, etc,
than you can shake a big stick at.

Having said that a good book as a primer is always handy. I recommend
Beginning Java 2 (Ivor Horton). I believe the latest edition covers
the latest Java flavour.

 

[Paul Selwood]

Tom Anderson wrote:

> On Thu, 7 Apr 2005, Tim Woodall wrote:
>> OTOH, because there are so many Fortran libraries that have been
>> optimized to extreme, it's just as likely that it will still be being
>> used in 50 years time.
>
> Seems horribly likely!

Not just libraries though. There are lots of O(1E6) line codes out there
that are written in Fortran and are optimised to the hilt. Re-writing in C
C++ would be a lot of work for very little gain.

>> The language that would replace Fortran (and C) in the number crunching
>> space would be a really good language for expressing parallelizable
>> algorithms. But it doesn't exist yet.
>
> High Performance Fortran?

Hardly. HPF is only useful for a very small set of codes (e.g. regular data
patterns). It has fairly weak compiler support and rarely beats a Fortran
MPI implementation of the same code.

Various US govt agencies and Cray are working on a language called Chapel
for high performance parallel coding. Sounds interesting, but I'm not sure
how much uptake it'll have.

Paul
--
Paul Selwood

 

[dwb]

Shaun Murray wrote:
>
> If you want a full on IDE for C, C++ or Java then consider buying a
> Mac. You can buy a complete Mac for the price Microsoft charges just
> for it's compiler environment

http://lab.msdn.microsoft.com/express/

You can get a Mac for free now? Where from?


and Apple gives it's tools away for
> free and doesn't charge for membership of it's developer network
> unless you want pre-release OSs - http://www.apple.com/xcode/

It doesn't cost anything to access all of MSDN either.

 

[dwb]

Jon Senior wrote:
> Not Responding wrote:
>> My question is, where do I start? I'm so out of date I can't even ask
>> the right questions. I used to write in Pascal, C, C++ and, once,
>> Visual Basic. That was all mainly under DOS (or for embedded
>> systems). What do I want to do? I'd like to be able to tinker with
>> databases, use the Windows GUI with ease and pull stuff off the
>> internet to manipulate.
>>
>> Any ideas?
>
> Drop VB. C#

You know that Vb.net is virtually identical in terms of (the most commonly
used functionality) to C# don't you?

 

[David Martin]

On 9/4/05 7:56 pm, in article [email protected], "dwb"
<[email protected]> wrote:

> It doesn't cost anything to access all of MSDN either.

The only times I have had to work with MS tools the rest of the lab were
begging me to stop becasue it was too painful for all concerned. It is just
fantastic when you have only one development environment that is a) buggy
and b) badly documented. VB on Macs is a really bad combination.

...d