caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Anil Madhavapeddy <anil@recoil.org>
To: Simon Cruanes <simon.cruanes.2007@m4x.org>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] Use of OCaml in universities and engineering schools
Date: Mon, 18 Mar 2013 10:47:17 -0500	[thread overview]
Message-ID: <E4ABAA5A-715F-4C90-8C58-AEF1426CF26C@recoil.org> (raw)
In-Reply-To: <51471FA2.6050602@m4x.org>

OPAM is just one component of the OCaml Platform we're working on
at the moment.  The Platform design and implementation currently underway
is looking at the build/test/documentation tool situation, as well as
making it easier to utilise third party standard libraries.

This is still a work-in-progress, but if you want to participate in
thrashing out the gory details, do feel free to join the mailing list at:
http://lists.ocaml.org/listinfo/platform

A very useful aspect of the current OCaml toolchain is how little
change is needed to build all of this.  OCaml-4+ exports the compiler
libraries, exports the typed AST via bin-annot, and OPAM generally makes
it easier to rebuild the toolchain in cases where patches to the core are
required.

-anil

On 18 Mar 2013, at 09:07, Simon Cruanes <simon.cruanes.2007@m4x.org> wrote:

> Wouldn't promoting the use of an extended standard library, like
> Batteries or Core, in addition to a friendly toplevel such as utop help
> making the language nicer to beginners? Opam makes it easier to use
> them; the biggest source of complexity from there is imho how to build
> programs (which build system to chose, how to have it use the libraries
> with findlib, etc.).
> 
> All this is very easy and natural in python: the standard library is
> big, you can write programs with networking, http, serialization,
> databases very easily, there is no build system (nor 'annoying' typing
> that prevents from writing `1 + 2.0 < 3L`). Ipython can also be used for
> the 'nice toplevel' part. This is also partly true for universities that
> use Java, since IDEs like eclipse provide autocompletion and hide the
> build system away... Even `ghc --make` and ghci are much easier than
> ocamlbuild and ocaml's toplevel (which doesn't even have readline).
> 
> So I think this explains the high threshold for starting with OCaml.
> 
> Simon
> 
> On 03/18/2013 03:00 PM, Yaron Minsky wrote:
>> I know of few places in the US that teach Haskell in the intro
>> sequence.  The main concern, I think, is that teaching imperative
>> programming in Haskell requires too much sophistication.
>> 
>> That said, I think there are a decent number of places in Europe that
>> do teach Haskell in the intro sequence, so YMMV.
>> 
>> My sense is that in the US, ML has quite a good spot relative to other
>> languages when it comes to University teaching.  The primary languages
>> people teach with in elite US institutions are, I think:
>> 
>> - Java
>> - Python
>> - C
>> - OCaml/SML
>> - Scheme
>> 
>> With Java and Python having the lion's share.  C and OCaml/SML are
>> most often taught as part of the "advanced" intro class.
>> 
>> I think this is a real opportunity for OCaml.  If we can make OCaml
>> much easier to use for newbies who want to do something outside of
>> class, I think it's a real chance to reach a wider audience.  OPAM
>> gets us a chunk of the way, but there's more work to do beyond that.
>> 
>> y
>> 
>> On Sun, Mar 17, 2013 at 5:29 PM, Lukasz Stafiniak <lukstafi@gmail.com> wrote:
>>> Evolution isn't about how much you achieve but about how well you compete.
>>> How do we stand in relation to Haskell in education? Or is your worry solely
>>> about giving ground to Python?
>>> 
>> 
> 
> 
> -- 
> Caml-list mailing list.  Subscription management and archives:
> https://sympa.inria.fr/sympa/arc/caml-list
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
> 


  reply	other threads:[~2013-03-18 15:47 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-15 16:49 Nicolas Barnier
2013-03-15 17:56 ` Yaron Minsky
2013-03-15 18:04   ` Roberto Di Cosmo
2013-03-15 18:24     ` Bertrand Bonnefoy-Claudet
2013-03-15 18:35       ` Roberto Di Cosmo
2013-03-15 20:12     ` Ashish Agarwal
2013-03-16  1:02       ` Philippe Wang
2013-03-16  1:15         ` Lukasz Stafiniak
2013-03-16  5:20           ` Bertrand Bonnefoy-Claudet
2013-03-15 20:53 ` Marek Kubica
2013-03-16  5:26 ` Jason Yeo
2013-03-16  5:34   ` Valentin ROBERT
2013-03-16  8:37     ` Dagnat Fabien
2013-03-16 15:26     ` Milan Stanojević
2013-03-17 15:58 ` Alan Schmitt
2013-03-17 21:07 ` Kristopher Micinski
2013-03-17 21:29   ` Lukasz Stafiniak
2013-03-18 14:00     ` Yaron Minsky
2013-03-18 14:07       ` Simon Cruanes
2013-03-18 15:47         ` Anil Madhavapeddy [this message]
2013-03-19  5:43       ` D. LoBraico
2013-03-18  8:20   ` Marc Pantel
2013-03-18  9:52     ` Christophe Garion
2013-03-18 10:16       ` Sebastien Ferre
2013-03-26 10:21         ` Nicolas Braud-Santoni
2013-03-19  7:35 ` Simão Sousa
2013-03-22 14:52 ` Luca Saiu
2013-03-17 16:22 Mark Raymond

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=E4ABAA5A-715F-4C90-8C58-AEF1426CF26C@recoil.org \
    --to=anil@recoil.org \
    --cc=caml-list@inria.fr \
    --cc=simon.cruanes.2007@m4x.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).