[Caml-list] A student feedback on OCaml

[Caml-list] A student feedback on OCaml

[Christophe Raffalli]

[-- Attachment #1: Type: text/plain, Size: 124 bytes --]


Hello,

Here is a student feedback:

https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png

Cheers,
Christophe

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [Caml-list] A student feedback on OCaml

[Viet Le]

[-- Attachment #1: Type: text/plain, Size: 390 bytes --]

Nice drawing! Care to give the Little Prince's version of the python eating
the camel?

What was your use case and what caused dissapointment?

On Tue, 25 Jul 2017 at 21:36, Christophe Raffalli <christophe@raffalli.eu>
wrote:

>
> Hello,
>
> Here is a student feedback:
>
> https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png
>
> Cheers,
> Christophe
>
-- 
Kind regards,
Viet

[-- Attachment #2: Type: text/html, Size: 941 bytes --]

Re: [Caml-list] A student feedback on OCaml

[Oliver Bandel]

IEEE Spectrum said,  Python is the number one programming language...


Zitat von Viet Le <vietlq85@gmail.com> (Tue, 25 Jul 2017 20:49:13 +0000)

> Nice drawing! Care to give the Little Prince's version of the python eating
> the camel?
>
> What was your use case and what caused dissapointment?
>
> On Tue, 25 Jul 2017 at 21:36, Christophe Raffalli <christophe@raffalli.eu>
> wrote:
>
>>
>> Hello,
>>
>> Here is a student feedback:
>>
>> https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png
>>
>> Cheers,
>> Christophe
>>
> --
> Kind regards,
> Viet
>
> --
> 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

[Caml-list] Probabilistic Functional Programming

[Van Chan Ngo]

[-- Attachment #1: Type: text/plain, Size: 458 bytes --]

Hi all,

The following link gives a simple example of “probabilistic functional programming”. 
http://channgo2203.github.io/articles/2017-02/probfind <http://channgo2203.github.io/articles/2017-02/probfind>

How do you think about the importance of the expected runtime for this class of functional programs?
Is an automatic analysis valuable in practice? And any idea about the applications of probabilistic functional programming.

Best,
-Chan

[-- Attachment #2: Type: text/html, Size: 915 bytes --]

Re: [Caml-list] Probabilistic Functional Programming

[Francois BERENGER]

On 07/26/2017 06:12 AM, Van Chan Ngo wrote:
> Hi all,
> 
> The following link gives a simple example of “probabilistic functional 
> programming”.
> http://channgo2203.github.io/articles/2017-02/probfind

If you are interested in the topic:

http://okmij.org/ftp/kakuritu/

and

http://okmij.org/ftp/kakuritu/Hansei.html

might be interesting pages.

Disclaimer: I never used any of those

PS: kakuritu in the URLs should probably be kakuritsu

> How do you think about the importance of the expected runtime for this 
> class of functional programs?
> Is an automatic analysis valuable in practice? And any idea about the 
> applications of probabilistic functional programming.
> 
> Best,
> -Chan

Re: [Caml-list] Probabilistic Functional Programming

[Van Chan Ngo]

[-- Attachment #1: Type: text/plain, Size: 1399 bytes --]

Thanks Francois. I heard about them. There exists also a wiki page of probabilistic programming
http://probabilistic-programming.org/wiki/Home <http://probabilistic-programming.org/wiki/Home>

In my opinion, it will (and currently) be attractive techniques in machine learning and robotics.

Best,
-Chan

> On Jul 25, 2017, at 8:06 PM, Francois BERENGER <berenger@bioreg.kyushu-u.ac.jp> wrote:
> 
> On 07/26/2017 06:12 AM, Van Chan Ngo wrote:
>> Hi all,
>> The following link gives a simple example of “probabilistic functional programming”.
>> http://channgo2203.github.io/articles/2017-02/probfind
> 
> If you are interested in the topic:
> 
> http://okmij.org/ftp/kakuritu/
> 
> and
> 
> http://okmij.org/ftp/kakuritu/Hansei.html
> 
> might be interesting pages.
> 
> Disclaimer: I never used any of those
> 
> PS: kakuritu in the URLs should probably be kakuritsu
> 
>> How do you think about the importance of the expected runtime for this class of functional programs?
>> Is an automatic analysis valuable in practice? And any idea about the applications of probabilistic functional programming.
>> Best,
>> -Chan
> 
> -- 
> 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


[-- Attachment #2: Type: text/html, Size: 2521 bytes --]

Re: [Caml-list] Probabilistic Functional Programming

[Oleg]

> How do you think about the importance of the expected runtime for this
> class of functional programs?
> Is an automatic analysis valuable in practice? And any idea about the
> applications of probabilistic functional programming.

Speaking of applications and importance, I would like to point out the
DARPA PPAML program that has just finished:
        http://ppaml.galois.com/wiki/

The challenge problems on the above opage gave the indication of the
depth and breadth. 

About the static analysis, etc. I would recommend browsing the materials
of the two Probabilistic Programming Semantics workshops

        http://pps2017.soic.indiana.edu/
        http://pps2016.soic.indiana.edu/

RE: [Caml-list] A student feedback on OCaml

[Soegtrop, Michael]

Dear Christophe,

comparing Python with OCaml is like comparing a hand milling tool (a "Dreml") with a CNC milling machine. For quick tasks the manual tool is better, for complex/precision/production work the CNC machine is better. I use both Python and OCaml for different tasks. There is a certain overlap area where I discuss with colleagues what they would use for this task before I start, but there is a large area where one or the other is clearly better. And I agree that for many tasks students are confronted with, Python is the better choice. But one shouldn't conclude from this, that this is the case for all tasks, and I would expect that teachers make this point.

Best regards,

Michael

Intel Deutschland GmbH
Registered Address: Am Campeon 10-12, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de
Managing Directors: Christin Eisenschmid, Christian Lamprechter
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928

Re: [Caml-list] A student feedback on OCaml

[Michael C Vanier]

It's a cute drawing.  It's hard to know more without any context.  I 
teach both Python and OCaml to undergraduates, but the students have to 
learn Python first (to get a firm grasp on basic programming).  My 
experience with OCaml (which I vastly prefer to Python) is that it's 
polarizing.  Some students absolutely love it, but others struggle with 
the language and have a hard time getting their programs to compile even 
after weeks of effort. Such students are obviously going to prefer other 
languages.

Mike


On 7/25/17 4:35 PM, Christophe Raffalli wrote:
> Hello,
>
> Here is a student feedback:
>
> https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png
>
> Cheers,
> Christophe

Re: [Caml-list] A student feedback on OCaml

[Gary Trakhman]

[-- Attachment #1: Type: text/plain, Size: 2681 bytes --]

Having used OCaml as a professional but not an enthusiast (I have a
clojure-depth), I am not polarized but ambivolent three months in of daily
use.  I enjoy the type system and life inside the bounds of a shared
codebase, but I'm finding a lot of the edge-system-integration (lack of 3rd
party libs) adding up.  Our company (Arena) fills in many of the gaps with
python due to its relevance as a data-science language and our problem
space, but I still think keeping everything in a JVM single process
monolith (like I have in my past) saves a lot of process-churn and
overhead.  Shelling out for _anything_ breaks abstraction boundaries and
requires complexity around build systems, etc., and we find ourselves doing
it for things like making SOAP calls, parsing XLS files, etc.  All these
prevent me from spending more time writing OCaml code.  Things like shared
infrastructure, plentiful libraries, a fast shared GC in something like the
JVM or CLR (or python or JS VM) can affect productivity more than the
language design itself.

Saying a language is 'better' depends on your use-cases, team-scale factors
and tradeoffs.  I myself can be very productive in a dynamic language, but
I'm becoming convinced that a type-system can really matter in larger
codebases, larger teams, or bitrot-prevention in code that is rarely
touched and out of mind.

Anyway, I provide these 2-cents because I think there are few people that
come to work in OCaml (or any functional language) by happen-stance and it
might be valuable feedback to someone.  Being really excited about a
language makes it easier to gloss over shortcomings.

On Thu, Jul 27, 2017 at 12:49 PM Michael C Vanier <mvanier@cms.caltech.edu>
wrote:

> It's a cute drawing.  It's hard to know more without any context.  I
> teach both Python and OCaml to undergraduates, but the students have to
> learn Python first (to get a firm grasp on basic programming).  My
> experience with OCaml (which I vastly prefer to Python) is that it's
> polarizing.  Some students absolutely love it, but others struggle with
> the language and have a hard time getting their programs to compile even
> after weeks of effort. Such students are obviously going to prefer other
> languages.
>
> Mike
>
>
> On 7/25/17 4:35 PM, Christophe Raffalli wrote:
> > Hello,
> >
> > Here is a student feedback:
> >
> > https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png
> >
> > Cheers,
> > Christophe
>
>
> --
> 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
>

[-- Attachment #2: Type: text/html, Size: 3511 bytes --]

Re: [Caml-list] A student feedback on OCaml

[Damien Guichard]

Does anyone teach programming using EasyOCaml ?

http://easyocaml.forge.ocamlcore.org/

And most importantly : does EasyOCaml  have a better adhesion potential 
than plain OCaml ?

Regards,

-- damien guichard


Michael C Vanier a écrit :
> It's a cute drawing.  It's hard to know more without any context.  I 
> teach both Python and OCaml to undergraduates, but the students have 
> to learn Python first (to get a firm grasp on basic programming).  My 
> experience with OCaml (which I vastly prefer to Python) is that it's 
> polarizing.  Some students absolutely love it, but others struggle 
> with the language and have a hard time getting their programs to 
> compile even after weeks of effort. Such students are obviously going 
> to prefer other languages.
>
> Mike
>
>
> On 7/25/17 4:35 PM, Christophe Raffalli wrote:
>> Hello,
>>
>> Here is a student feedback:
>>
>> https://lama.univ-savoie.fr/~raffalli/pics/caml-versus-python.png
>>
>> Cheers,
>> Christophe
>
>