I am an absolute novice just getting in to programming. I'm fairly capable with computers in general, but the closest I've come to actual programming is messing around in DrScheme, some pseudocode in math class and webpage markup.
Basically, I chose Lisp as my first programming language, because I like the idea of a uniform syntax, and people seem to be so passionate about what a good language it is, so it seems a good place to start.
So really, my noob questions are
-Is there any advantage between using Common Lisp or Scheme, and are they that different. -Should I use an IDE like PLT Scheme/racket or just use emacs and a compiler
But MOST importantly
-Can anyone recomment a good guide, whether it's a book (I don't mind buying something) or free online guide, that will get me going with the basic operations of Lisp and some example programs, and maybe even start going into more intermediate to advanced stuff? I'm not looking for something that has to be really easy, I don't mind working hard to understand it, I just need to be pointed in the right direction on where to start...
Thanks :):):)
--- Posted via news://freenews.netfront.net/ - Complaints to n...@netfront.net ---
On Fri, 20 Jul 2012 02:11:02, Alex <muzikman...@dontwantspam.com> wrote:
> -Is there any advantage between using Common Lisp or Scheme, and are > they that different. -Should I use an IDE like PLT Scheme/racket or just > use emacs and a compiler
> -Can anyone recomment a good guide, whether it's a book (I don't mind > buying something) or free online guide, that will get me going with the > basic operations of Lisp and some example programs, and maybe even start > going into more intermediate to advanced stuff? I'm not looking for > something that has to be really easy, I don't mind working hard to > understand it, I just need to be pointed in the right > direction on where to start...
For an absolute beginner, your choice of language isn't very important.
You can always learn other languages later. The key thing for a new
programmer is to learn how to think about programming.
I would recommend the Structure and Interpretation of Computer Programs
(available free online from MIT, or you could buy a hard copy) as an
introduction to thinking about programming using Scheme. There are also
videos of accompanying lectures available free online, and I would
especially recommend the 1986 lectures by Abelson and Sussman (the
authors of the book), which MIT has online.
SICP is somewhat math-heavy, which may or may not turn you off to it.
You might like to read The Little Schemer first, which is lighter and
would prepare you for some of the concepts presented earlier on in SICP.
I've also heard a lot of good things about the book How To Design
Programs (which uses Racket, a sort of Scheme, in particular), but I
haven't read it.
Your development environment is entirely personal choice. You can easily
set up Emacs to run a Scheme or Common Lisp interpreter that you can
interact with while you're editing a file, which is what most people do.
While editing a Scheme file, it works perfectly fine to split the window
in two and run "M-x run-scheme" in one (assuming you have a Scheme
implementation installed). Then C-x-e will use Scheme to evaluate the
last expression before your cursor, and you can switch to the Scheme
window to interact with the REPL normally.. That should be enough for a
beginner to get comfortable, but Google around for further functionality
you can have.
Alex <muzikman...@dontwantspam.com> writes:
> -Is there any advantage between using Common Lisp or Scheme, and are
> they that different.
This is usually a loaded question, but in the current context, the
general recommendation would be to use CL. However, you should be
careful when you talk about "Scheme" -- if you want to talk about the
standard language, then there's very little that you can do with it.
But that's not the case with an actual implementation like Racket
(which is how PLT Scheme is called now).
> -Should I use an IDE like PLT Scheme/racket or just use emacs and a
> compiler
DrRacket is intended to be a Racket (or another sexpr-based language)
friendly editor, but it doesn't try to replace Emacs. Learning Emacs
is useful in itself (and independent of any language) but it'll take
some significant effort to get used to it.
> But MOST importantly
> -Can anyone recomment a good guide, whether it's a book (I don't
> mind buying something) or free online guide, that will get me going
> with the basic operations of Lisp and some example programs, and
> maybe even start going into more intermediate to advanced stuff? I'm
> not looking for something that has to be really easy, I don't mind
> working hard to understand it, I just need to be pointed in the
> right direction on where to start...
On Jul 19, 10:11 pm, Alex <muzikman...@dontwantspam.com> wrote:
> -Can anyone recomment a good guide, whether it's a book (I don't mind
> buying something) or free online guide, that will get me going with the
> basic operations of Lisp and some example programs, and maybe even start
> going into more intermediate to advanced stuff? I'm not looking for
> something that has to be really easy, I don't mind working hard to
> understand it, I just need to be pointed in the right
> direction on where to start...
There are a number of online resources. For a start, look at:
Peter Seibel 'Practical Common Lisp'
David Touretsky 'Gentle Introduction to Symbolic Computation'
Pual Graham 'On Lisp' (advanced)
Nick Levine's unfinished book on Lisp for O'Reilly
David Cooper 'Basic Lisp Techniques'
The Common Lisp HyperSpec (can be downloaded)
Common Lisp Quick Reference (clqr)
For dead tree books, purchase:
Conrad Barski 'Land of Lisp'
Paul Graham 'ANSI Common Lisp'
I have found using Common Lisp extraordinarily difficult, in part for
the following reasons:
1. There a number of implementations for Common Lisp, and they are all
different. Learning CL means learning a particular implementation,
which means that the things you read may or may not work in your
implementation, but no on3e tells you that.
2. There's not one way of dealing with external libraries. The
information you read on packages and systems isn't consistent, and
again, no one tells you that.
3. With specific tasks (e.g., building a GUI), different librarys
exist which are incompatible, with different features and different
documentation, and again, no one explains the different libraries.
4. Most languages have a set of tasks for which they are well suited
and they have concise tutorials on using the languages for the task.
Not so in CL. In fact, it's very hard to find any particular thing CL
excels at, so you can learn the language for that purpose and have a
base for further learning.
Still, CL is worth learning. The one thing I would emphasize is that
you have a need for learning the language, so you can write real code
and test it against real requirements. Otherwise, CL is just another
dead language that you might develop an intellectual knowledge but
never an experience base.
ccc31807 <carte...@gmail.com> writes:
> I have found using Common Lisp extraordinarily difficult, in part for
> the following reasons:
> 1. There a number of implementations for Common Lisp, and they are all
> different. Learning CL means learning a particular implementation,
> which means that the things you read may or may not work in your
> implementation, but no on3e tells you that.
This is wrong. All the CL implementation implement the ANSI Common Lisp
language, which is nicely specified in the Hyperspec.
(Almost) all my CL programs are conforming: they run on all the
conforming CL implementations.
> 2. There's not one way of dealing with external libraries. The
> information you read on packages and systems isn't consistent, and
> again, no one tells you that.
> 3. With specific tasks (e.g., building a GUI), different librarys
> exist which are incompatible, with different features and different
> documentation, and again, no one explains the different libraries.
Which is exactly the same with developing GUIs in other programming
languages: will use use libX11? Gtk? Qt? Cocoa?
> 4. Most languages have a set of tasks for which they are well suited
> and they have concise tutorials on using the languages for the task.
> Not so in CL. In fact, it's very hard to find any particular thing CL
> excels at, so you can learn the language for that purpose and have a
> base for further learning.
Indeed, CL is good at everything.
> Still, CL is worth learning. The one thing I would emphasize is that
> you have a need for learning the language, so you can write real code
> and test it against real requirements. Otherwise, CL is just another
> dead language that you might develop an intellectual knowledge but
> never an experience base.
> I am an absolute novice just getting in to programming. I'm fairly > capable with computers in general, but the closest I've come to actual > programming is messing around in DrScheme, some pseudocode in math class > and webpage markup.
> Basically, I chose Lisp as my first programming language, because I like > the idea of a uniform syntax, and people seem to be so passionate about > what a good language it is, so it seems a good place to start.
> So really, my noob questions are
> -Is there any advantage between using Common Lisp or Scheme, and are > they that different. -Should I use an IDE like PLT Scheme/racket or just > use emacs and a compiler
I think the most basic difference between Common Lisp and Scheme is in
the area of personality.
Do you like to collect tools, and work on complex devices? Do you enjoy
rebuilding and refining things? Then you might be happier with Common
Lisp.
Do you like new and elegant things? Do you like to periodically throw
out your old stuff and start over with new things? Then you might
be happier with Scheme.
If you think in terms of just the end program, and don't care how your
code looks, then maybe you would be better off with some other language
family.
On Fri, Jul 20 2012, Pascal J. Bourguignon wrote:
> ccc31807 <carte...@gmail.com> writes:
>> I have found using Common Lisp extraordinarily difficult, in part for
>> the following reasons:
>> 1. There a number of implementations for Common Lisp, and they are
>> all different. Learning CL means learning a particular
>> implementation, which means that the things you read may or may not
>> work in your implementation, but no one tells you that.
> This is wrong. All the CL implementation implement the ANSI Common
> Lisp language, which is nicely specified in the Hyperspec.
> (Almost) all my CL programs are conforming: they run on all the
> conforming CL implementations.
No, it is absolutely right. You have enough experience dealing with the
vagaries of the various Common Lisp implementations that the differences
are a mere annoyance to you.
However, to someone relatively new to Lisp (like myself), the various
implementations of Common Lisp present a sizable barrier to entry.
Heck, /Practical Common Lisp/ has an entire chapter devoted to writing a
library that allows you to smooth over the various warts that these
"conforming" implementations have in reading files and directories.
If you want to do something interesting with Common Lisp (say write a
web application) then the implementation that you use *absolutely*
matters. clisp is not going to run Weblocks, for example. If you would
like to create a GUI then you have a different set of hurdles, and you
might well need a different implementation. The list goes on.
Ten years ago that might have been acceptable in a programming language,
but these days the competition has stepped up their game.
That being the case, there does appear to be at least one *good*
implementation for most platforms. I recommend sbcl for Linux, it seems
to mostly work.
On Jul 20, 10:31 am, "Pascal J. Bourguignon" <p...@informatimago.com>
wrote:
> Indeed, CL is good at everything.
Everything ... and nothing.
For example, using Perl, if you want to use Perl for web applications,
you can get a book on Perl/CGI. If you want to use Perl for database
stuff, you can get a book on Perl/DBI. If you want to use Perl for
building a GUI, you can get a book on Perl/TK.
If you want to use CL for web applications, where's the book on CL/
HTTP? or for database stuff, CL/DB? or for GUIs, CL/GUI?
The closest CL comes to any of this is AI. You want to use CL for AI,
there's Norvig, or Taminoto (sp?), or Winston, and others.
Lispers tend to build things from scratch rather than search for pre-
built libraries. This frustrated me for a while, until I figured out
that, if I wanted some functionality, I had to build it myself, and
how easy it was to build in CL. I am currently developing some
proficiency in R, and there's such a wealth in R that it's not only in
another category than CL -- it's in another universe. Installing
packages in R is dead simple. All you have to do is call
install.packages("somepackage") and, presto chango, you can use it
immediately.
I'm not complaining or griping or maligning CL. I'm just describing
how it is. And yeah, it's difficult for a newby trying to develop
expertise just by reading a book and experimenting.
ccc31807 <carte...@gmail.com> writes:
> On Jul 20, 10:31 am, "Pascal J. Bourguignon" <p...@informatimago.com>
> wrote:
>> Indeed, CL is good at everything.
> Everything ... and nothing.
> For example, using Perl, if you want to use Perl for web applications,
> you can get a book on Perl/CGI. If you want to use Perl for database
> stuff, you can get a book on Perl/DBI. If you want to use Perl for
> building a GUI, you can get a book on Perl/TK.
Why would you need a book? You can't program by yourself???
> Installing
> packages in R is dead simple. All you have to do is call
> install.packages("somepackage") and, presto chango, you can use it
> immediately.
Installing systems in Common Lisp is even deadlier simplier. All you
have to do is to call:
(ql:quickload :some-system)
> I'm not complaining or griping or maligning CL. I'm just describing
> how it is.
No. You're describing how it was ten or twenty years ago.
On Jul 20, 4:56 pm, "Pascal J. Bourguignon" <p...@informatimago.com>
wrote:
> No. You're describing how it was ten or twenty years ago.
And most of the written material is that old. AFAIK, there are only
two Lisp books written in the past decade or so: Seibel and Barski.
I probably live 120 miles from the nearest Lisp programmer, but I run
into Perlistas on a monthly basis. I probably know a dozen or so guys
that write Java, or .NET, or C++ in my wider circle of friends. Lisp?
I don't know a soul.
The CL community is small, fractured, and inarticulate. This isn't
good for the health of the language.
Alex <muzikman...@dontwantspam.com> writes:
> I am an absolute novice just getting in to programming. I'm fairly > capable with computers in general, but the closest I've come to actual > programming is messing around in DrScheme, some pseudocode in math class > and webpage markup.
> Basically, I chose Lisp as my first programming language, because I like > the idea of a uniform syntax, and people seem to be so passionate about > what a good language it is, so it seems a good place to start.
> So really, my noob questions are
> -Is there any advantage between using Common Lisp or Scheme, and are > they that different. -Should I use an IDE like PLT Scheme/racket or just > use emacs and a compiler
I found this webpage comparing Common Lisp, Scheme, Clojure, and Emacs
Lisp very helpful when I was trying to figure this out a few years ago.
> -Can anyone recomment a good guide, whether it's a book (I don't mind > buying something) or free online guide, that will get me going with the > basic operations of Lisp and some example programs, and maybe even start > going into more intermediate to advanced stuff? I'm not looking for > something that has to be really easy, I don't mind working hard to > understand it, I just need to be pointed in the right > direction on where to start...
You can try working any of the 99 Lisp problems that interest you.
ccc31807 <carte...@gmail.com> writes:
> On Jul 20, 4:56 pm, "Pascal J. Bourguignon" <p...@informatimago.com>
> wrote:
>> No. You're describing how it was ten or twenty years ago.
> And most of the written material is that old. AFAIK, there are only
> two Lisp books written in the past decade or so: Seibel and Barski.
> I probably live 120 miles from the nearest Lisp programmer, but I run
> into Perlistas on a monthly basis. I probably know a dozen or so guys
> that write Java, or .NET, or C++ in my wider circle of friends. Lisp?
> I don't know a soul.
And you're probably at least a few meters away from the nearest human
being, but there are billions of microbes on and in your very own body.
Perhaps you should forget being multi-cellular organism, and consider reverting to the microbial life instead. (And similar analogies
mentionning flies and McDonald's).
ccc31807 wrote:
> On Jul 20, 10:31 am, "Pascal J. Bourguignon" <p...@informatimago.com>
> wrote:
> > Indeed, CL is good at everything.
> Everything ... and nothing.
What Pascal is saying, although he doesn't realize it, is that
DARPA CL is a sort of Turing machine: theoretically you can
do anything with it, but it is extremely clunky and awkward---
not really practical.
>> -Is there any advantage between using Common Lisp or Scheme, and are
>> they that different.
> This is usually a loaded question, but in the current context, the
> general recommendation would be to use CL. However, you should be
> careful when you talk about "Scheme" -- if you want to talk about the
> standard language, then there's very little that you can do with it.
> But that's not the case with an actual implementation like Racket
> (which is how PLT Scheme is called now).
I'm a CL user myself, and I still think Racket has a lot of good things going for it.
Depending on how new you are as a programmer, the following might be a useful guide.
ccc31807 <carte...@gmail.com> writes:
> On Jul 20, 4:56 pm, "Pascal J. Bourguignon" <p...@informatimago.com>
> wrote:
>> No. You're describing how it was ten or twenty years ago.
> And most of the written material is that old. AFAIK, there are only
> two Lisp books written in the past decade or so: Seibel and Barski.
Until the updated 4th edition came out a few months ago, I was perfectly
happy with my copy of 'Programming Perl'. If you kept up with the
changes (documented in the perldelta perldocs), its description of the
nuts and bolts of Perl was still relevant, even though it was ten years
out of date.
The written material for Common Lisp has changed little because the
language has changed little. It may be old, but it's still relevant.
Heck, if you start looking at Ada (my #3 favourite language), most docs
are based on the '95 standard, and Ada-2005 is only a minor update on
that.
Languages reach a point of maturity, and after that don't change
much. Lots of change in the language definition and standards are a sign
of an immature language (or just a plain bad one, see PHP).
> I probably live 120 miles from the nearest Lisp programmer, but I run
> into Perlistas on a monthly basis. I probably know a dozen or so guys
> that write Java, or .NET, or C++ in my wider circle of friends. Lisp?
> I don't know a soul.
Neither do I. Still doesn't stop me from redoing my hobby projects in
Lisp, while still doing Perl at work.
Incidentally, Perl has a lot of little idioms that translate fairly well
to lisp. The use of 'for' as a statement modifier for example, is nearly
identical to constructs like 'mapcar'; giving credence to the motto that
all languages eventually implement Lisp as they mature.
Mart
-- "We will need a longer wall when the revolution comes."
--- AJS, quoting an uncertain source.
