[Caml-list] UFO (United Forces of OCaml)

[Caml-list] UFO (United Forces of OCaml)

[Gour]

Hello,

I'm at the beginning of learning and using OCaml which is very fine
language and I've decided to use it over Ada.

Many new things are happening like development of OPAM (very nice
project), then moving some projects to ocaml.org subdomains, new design
of the site as well...

However, there are certain things which seems like wasting of resources,
so my humble proposal is whether it is possible to make UFO - United
Forces of OCaml in order not to have too many (similar) projects
tackling the same problems.

One area which quickly comes to my mind is standard library. Although
I'm not (yet) able to conclude how much is standard library incomplete,
I see there are two larger projects trying to fill the gap.

Jane Street is calling it "'Jane Street's alternative to the standard
library", while 'batteries' are labelled as: "community-driven effort to
standardize on an consistent, documented, and comprehensive development
platform for the OCaml programming language."

For the uninitiated it's seems as attempts to solve the same problem,
but, unfortunately, it looks they're tackling it in a non-compatible
way.

It's especially sensitive considering that the RWO book -  which might
be used for many noobs to learn the language - is promoting Core, while
'community-driven' project is working on something else.

Another thing which I can think of are build systems and although I'm
aware there is certain overlap in functionality and/or interdependances,
to me it seems that there are too many of them:

a) OASIS

b) ocamlbuild

c) omake

d) ocp-build

e) yenga

f) ...

The list can be, of course, annotated by the need to have actively
developed/maintained bindings for truly multio-platform GUI bindings
(e.g. wx/Qt).

So, my naive proposal is to try to combine forces together and produce
small(er) set of tools libraries and make it somewhat 'standard' within
community. It will attract new people to the language itself by making
it (more) clear what are the standard tools to be used.

The OCaml community is, imho, not big-enough to allow such luxury of
re-inventing the wheels...

Moreover, OCaml is advertised as "general purpose industrial-strength
programming language" and it behooves to have mature fully baked
ecosystem.


Sincerely,
Gour

-- 
The work of a man who is unattached to the modes of material 
nature and who is fully situated in transcendental knowledge 
merges entirely into transcendence.

http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810

Re: [Caml-list] UFO (United Forces of OCaml)

[Török Edwin]

On 09/27/2013 10:11 AM, Gour wrote:
> 
> However, there are certain things which seems like wasting of resources,
> so my humble proposal is whether it is possible to make UFO - United
> Forces of OCaml in order not to have too many (similar) projects
> tackling the same problems.

I think this is worked on under the name "OCaml Platform":
http://www.cl.cam.ac.uk/projects/ocamllabs/tasks/platform.html

> So, my naive proposal is to try to combine forces together and produce
> small(er) set of tools libraries and make it somewhat 'standard' within
> community.

I'd prefer if existing tools are improved, and the best one considered as "standard",
rather than developing a new set of tools to replace all the existing ones.
As you noticed we have too many alternatives already, and if yet another one is created we'll end up with: http://xkcd.com/927/

Best regards,
--Edwin

Re: [Caml-list] UFO (United Forces of OCaml)

[Lukasz Stafiniak]

[-- Attachment #1: Type: text/plain, Size: 318 bytes --]

On Fri, Sep 27, 2013 at 9:11 AM, Gour <gour@atmarama.net> wrote:

>
> It's especially sensitive considering that the RWO book -  which might
> be used for many noobs to learn the language - is promoting Core, while
> 'community-driven' project is working on something else.
>
>
We need an OCaml-Batteries book now :-)

[-- Attachment #2: Type: text/html, Size: 662 bytes --]

Re: [Caml-list] UFO (United Forces of OCaml)

[Yaron Minsky]

On Fri, Sep 27, 2013 at 7:20 AM, Török Edwin <edwin+ml-ocaml@etorok.net> wrote:
> On 09/27/2013 10:11 AM, Gour wrote:
>>
>> However, there are certain things which seems like wasting of resources,
>> so my humble proposal is whether it is possible to make UFO - United
>> Forces of OCaml in order not to have too many (similar) projects
>> tackling the same problems.
>
> I think this is worked on under the name "OCaml Platform":
> http://www.cl.cam.ac.uk/projects/ocamllabs/tasks/platform.html
>
>> So, my naive proposal is to try to combine forces together and produce
>> small(er) set of tools libraries and make it somewhat 'standard' within
>> community.
>
> I'd prefer if existing tools are improved, and the best one considered as "standard",
> rather than developing a new set of tools to replace all the existing ones.
> As you noticed we have too many alternatives already, and if yet another one is created we'll end up with: http://xkcd.com/927/

I very much agree.  The problem of having too many libraries isn't
solved by a vote or by creating Yet Another Library.  It's solved by
one of the alternatives getting good enough that people migrate to it
of their own accord.  I'm hopeful that the work we're doing on Core
and its related libraries will get it there, but whether it does or
not is a question for people to vote on with their feet (or with their
fingers, I suppose.)

y

Re: [Caml-list] UFO (United Forces of OCaml)

[Gour]

On Fri, 27 Sep 2013 14:20:01 +0300
Török Edwin <edwin+ml-ocaml@etorok.net> wrote:

> I'd prefer if existing tools are improved, and the best one
> considered as "standard", rather than developing a new set of tools
> to replace all the existing ones. As you noticed we have too many
> alternatives already, and if yet another one is created we'll end up
> with: http://xkcd.com/927/

I was not thinking about 'new' set of tools, but choosing some onto
which the community effort should be focused.


Sincerely,
Gour

-- 
He is a perfect yogī who, by comparison to his own self, 
sees the true equality of all beings, in both their 
happiness and their distress, O Arjuna!

http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810

Re: [Caml-list] UFO (United Forces of OCaml)

[Paolo Donadeo]

On Fri, Sep 27, 2013 at 2:04 PM, Yaron Minsky <yminsky@janestreet.com> wrote:
> I very much agree.  The problem of having too many libraries isn't
> solved by a vote or by creating Yet Another Library.

I agree too, of course. But, regarding the specific case of the
"extended library", there aren't many libraries, but only two:
Batteries and Core. And I think both of them have very strong points,
and the design is quite different. There is enough room for both.


-- 
Paolo

Re: [Caml-list] UFO (United Forces of OCaml)

[David MENTRE]

Hello,

2013/9/27 Paolo Donadeo <p.donadeo@gmail.com>:
> But, regarding the specific case of the
> "extended library", there aren't many libraries, but only two:
> Batteries and Core. And I think both of them have very strong points,
> and the design is quite different.

Is there a short document somewhere explaining those design differences?

Best regards,
david

Re: [Caml-list] UFO (United Forces of OCaml)

[Siraaj Khandkar]

On 09/27/2013 11:55 AM, David MENTRE wrote:
> Hello,
>
> 2013/9/27 Paolo Donadeo <p.donadeo@gmail.com>:
>> But, regarding the specific case of the
>> "extended library", there aren't many libraries, but only two:
>> Batteries and Core. And I think both of them have very strong points,
>> and the design is quite different.
>
> Is there a short document somewhere explaining those design differences?
>

I haven't seen a document comparing the goals of both directly, but 
individually each project stated differing goals:

- Batteries aims to be a compatible _extension_ of stdlib (so minimal or 
no changes required for code already using stdlib)

- Core is an incompatible _replacement_ for stdlib, aiming to take the 
greatest advantage of the type system (existing code would have to be 
updated to use Core's API)

Re: [Caml-list] UFO (United Forces of OCaml)

[Ashish Agarwal]

[-- Attachment #1: Type: text/plain, Size: 1506 bytes --]

Gour, once the videos from the OCaml Meeting are posted online, you should
watch Anil's talk on OCaml Labs. He nicely explains that a large community
effort is well underway to address your proposals.


On Fri, Sep 27, 2013 at 12:25 PM, Siraaj Khandkar <siraaj@khandkar.net>wrote:

> On 09/27/2013 11:55 AM, David MENTRE wrote:
>
>> Hello,
>>
>> 2013/9/27 Paolo Donadeo <p.donadeo@gmail.com>:
>>
>>> But, regarding the specific case of the
>>> "extended library", there aren't many libraries, but only two:
>>> Batteries and Core. And I think both of them have very strong points,
>>> and the design is quite different.
>>>
>>
>> Is there a short document somewhere explaining those design differences?
>>
>>
> I haven't seen a document comparing the goals of both directly, but
> individually each project stated differing goals:
>
> - Batteries aims to be a compatible _extension_ of stdlib (so minimal or
> no changes required for code already using stdlib)
>
> - Core is an incompatible _replacement_ for stdlib, aiming to take the
> greatest advantage of the type system (existing code would have to be
> updated to use Core's API)
>
>
> --
> Caml-list mailing list.  Subscription management and archives:
> https://sympa.inria.fr/sympa/**arc/caml-list<https://sympa.inria.fr/sympa/arc/caml-list>
> Beginner's list: http://groups.yahoo.com/group/**ocaml_beginners<http://groups.yahoo.com/group/ocaml_beginners>
> Bug reports: http://caml.inria.fr/bin/caml-**bugs<http://caml.inria.fr/bin/caml-bugs>
>

[-- Attachment #2: Type: text/html, Size: 2356 bytes --]

Re: [Caml-list] UFO (United Forces of OCaml)

[Francois Berenger]

On 09/27/2013 04:11 PM, Gour wrote:
> [...]
>
> Another thing which I can think of are build systems and although I'm
> aware there is certain overlap in functionality and/or interdependances,
> to me it seems that there are too many of them:
>
> a) OASIS
>
> b) ocamlbuild
>
> c) omake
>
> d) ocp-build
>
> e) yenga
>
> f) ...

You forgot a very interesting one: obuild

Cf.
https://github.com/vincenthz/obuild

Re: [Caml-list] UFO (United Forces of OCaml)

[Gour]

On Mon, 30 Sep 2013 10:40:44 +0900
Francois Berenger <berenger@riken.jp> wrote:

> You forgot a very interesting one: obuild
> 
> Cf.
> https://github.com/vincenthz/obuild

That one looks interesting (being based on Cabal), although I notice
there is no 'uninstall' target. :-/


Sincerely,
Gour

-- 
One who sees inaction in action, and action in inaction, 
is intelligent among men, and he is in the transcendental position, 
although engaged in all sorts of activities.

http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810

Re: [Caml-list] UFO (United Forces of OCaml)

[Francois Berenger]

On 09/30/2013 03:01 PM, Gour wrote:
> On Mon, 30 Sep 2013 10:40:44 +0900
> Francois Berenger <berenger@riken.jp> wrote:
>
>> You forgot a very interesting one: obuild
>>
>> Cf.
>> https://github.com/vincenthz/obuild
>
> That one looks interesting (being based on Cabal), although I notice
> there is no 'uninstall' target. :-/

This is a known issue, since there is no install target currently.

I think there is a proposition in the bug tracker to generate
.install files for OPAM, though I cannot find the issue back.

-- 
Best regards,
Francois Berenger.

Re: [Caml-list] UFO (United Forces of OCaml)

[Sylvain Le Gall]

On 30-09-2013, Francois Berenger <berenger@riken.jp> wrote:
> On 09/27/2013 04:11 PM, Gour wrote:
>> [...]
>>
>> Another thing which I can think of are build systems and although I'm
>> aware there is certain overlap in functionality and/or interdependances,
>> to me it seems that there are too many of them:
>>
>> a) OASIS

OASIS is not a build system it is more like autoconf/automake. It relies
on OCamlbuild or any other build system.

Cheers,
Sylvain Le Gall
--
Website:     http://sylvain.le-gall.net/
OCaml forge: http://forge.ocamlcore.org
OCaml blogs: http://planet.ocaml.org

Re: [Caml-list] UFO (United Forces of OCaml)

[Gour]

[-- Attachment #1: Type: text/plain, Size: 2346 bytes --]

On Fri, 27 Sep 2013 19:08:28 +0200
"Damien Guichard" <alphablock@orange.fr> wrote:

Hiya Damien,

> OCaml is multi-paradigm in nature.

Great.

> There are multiple ways (multiple trade-offs) a thing can be done.

Nice.

> Same about books, documentation, tutorials, IDEs ...
> In contrast with Caml-Light, OCaml has no 'official' book, tutorial
> or IDE. However i expect many to come, some to become almost
> forgotten, some to be cited as a reference point (Ocaml evolves
> continually). Actually there is an 'official' OCaml 2.2 book, but
> somewhat obsolete now.

I plan to use RWO book.

> Just my humble opinion of course.

Of course.

> I understand that, as a beginner, you may consider diversity as a
> danger. It seems your software is more fragile because some basic
> block could be discontinued.

Right.

> I would answer : the software fragility lies in the programmer,
> not in the language ecosystem being too weak.  

Heh, "batteries included" still helps.

> Actually GTK 2.24 looks very great when a good theme is applied.

On Mac?

> You can have this metal brush look. Or this white water-button look.
> Or whatever look you can imagine (or design). 

I'm more worried about the state of GTK's support on Win/Mac than about
LablGTK...

> Plus LablGtk is an excellent binding that makes GTK programming far
> much easier.

...which I actually consider very good bindings.

> Otherwise i would warmly welcome a Qt binding.

Me too...similar with wxOCaml if having them actively maintained.

> However, i dont wait for it, i go ahead.

Sure, I'll start learning OCaml and working on non-GUI things.

> Even if i switch to Qt some day i don't loose my time,
> because knowing GTK, whatever language, is an asset anyway.   

That's true, but I'd prefer saving some of my free time meant to be
spent for working on this open-source project, if possible.

> I hope i am not too much rude with you,
> I apologize if i am,

Not at all. Thank you.


Sincerely,
Gour

-- 
A person is considered still further advanced when he regards honest 
well-wishers, affectionate benefactors, the neutral, mediators, the
envious, friends and enemies, the pious and the sinners all with an
equal mind.

http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [Caml-list] UFO (United Forces of OCaml)

[Damien Guichard]

Hello Gour,


OCaml is multi-paradigm in nature.
There are multiple ways (multiple trade-offs) a thing can be done.

Let's adopt the Open Source way:
- At first there is nothing (i remember the days when OCaml had zero MySQL
bindings)
- Then they are multiple things (the number of MySQL bindings is 3 or 4)
- Then some become unmaintained, some others gain attraction
- Then (in a few years) you end with 2 or 3 good / excellent ways to do it

Same about books, documentation, tutorials, IDEs ...
In contrast with Caml-Light, OCaml has no 'official' book, tutorial or IDE.
However i expect many to come, some to become almost forgotten,
some to be cited as a reference point (Ocaml evolves continually).  
Actually there is an 'official' OCaml 2.2 book, but somewhat obsolete now.

Just my humble opinion of course.

I understand that, as a beginner, you may consider diversity as a danger.
It seems your software is more fragile because some basic block could be
discontinued.
I would answer : the software fragility lies in the programmer,
not in the language ecosystem being too weak.  

Actually GTK 2.24 looks very great when a good theme is applied.
You can have this metal brush look. Or this white water-button look.
Or whatever look you can imagine (or design). 
Plus LablGtk is an excellent binding that makes GTK programming far much
easier.

Otherwise i would warmly welcome a Qt binding.
However, i dont wait for it, i go ahead.
Even if i switch to Qt some day i don't loose my time,
because knowing GTK, whatever language, is an asset anyway.   

I hope i am not too much rude with you,
I apologize if i am,

- Damien

 

--
Mail created using EssentialPIM Free - www.essentialpim.com

Le 27/09/2013 à 09:11, Gour <gour@atmarama.net> à écrit :
>Hello,
>
>I'm at the beginning of learning and using OCaml which is very fine
>language and I've decided to use it over Ada.
>
>Many new things are happening like development of OPAM (very nice
>project), then moving some projects to ocaml.org subdomains, new design
>of the site as well...
>
>However, there are certain things which seems like wasting of resources,
>so my humble proposal is whether it is possible to make UFO - United
>Forces of OCaml in order not to have too many (similar) projects
>tackling the same problems.
>
>One area which quickly comes to my mind is standard library. Although
>I'm not (yet) able to conclude how much is standard library incomplete,
>I see there are two larger projects trying to fill the gap.
>
>Jane Street is calling it "'Jane Street's alternative to the standard
>library", while 'batteries' are labelled as: "community-driven effort to
>standardize on an consistent, documented, and comprehensive development
>platform for the OCaml programming language."
>
>For the uninitiated it's seems as attempts to solve the same problem,
>but, unfortunately, it looks they're tackling it in a non-compatible
>way.
>
>It's especially sensitive considering that the RWO book -  which might
>be used for many noobs to learn the language - is promoting Core, while
>'community-driven' project is working on something else.
>
>Another thing which I can think of are build systems and although I'm
>aware there is certain overlap in functionality and/or interdependances,
>to me it seems that there are too many of them:
>
>a) OASIS
>
>b) ocamlbuild
>
>c) omake
>
>d) ocp-build
>
>e) yenga
>
>f) ...
>
>The list can be, of course, annotated by the need to have actively
>developed/maintained bindings for truly multio-platform GUI bindings
>(e.g. wx/Qt).
>
>So, my naive proposal is to try to combine forces together and produce
>small(er) set of tools libraries and make it somewhat 'standard' within
>community. It will attract new people to the language itself by making
>it (more) clear what are the standard tools to be used.
>
>The OCaml community is, imho, not big-enough to allow such luxury of
>re-inventing the wheels...
>
>Moreover, OCaml is advertised as "general purpose industrial-strength
>programming language" and it behooves to have mature fully baked
>ecosystem.
>
>
>Sincerely,
>Gour
>
>-- 
>The work of a man who is unattached to the modes of material 
>nature and who is fully situated in transcendental knowledge 
>merges entirely into transcendence.
>
>http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810
>
>
>
>-- 
>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