[Caml-list] Whither the Caml Consortium?

[Caml-list] Whither the Caml Consortium?

[Brian Rogoff]

Hi,
    What's become of the Caml Consortium? I put a lot of pressure on my
manager to join and we've hardly heard anything since doing so. That was
several months ago.

-- Brian


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Patrick M Doane]

While on this subject, it would be very useful if a future roadmap for
Caml development (including any Consortium efforts) were available to look
at publicly.  This was discussed a little in the past and it seemed like
INRIA wanted to keep this information private. 

I can understand motivations to keep quiet about plans but I think it
would really help to open up this information. It may so happen that some
unexpected volunteers are willing to help achieve those goals. 

For example, I am currently working with Jerome to build a test suite for
the libre code. I'm not a big user of regular expressions myself but I
think this is a critical area where Caml could be improved. If I knew of
specific efforts that were being directed by the Consortium, I'd be
inclined to help in a similar fashion with them. Testing software is a
great way for me to learn how to use it. 

Patrick

P.S: Thanks also to Brian for suggesting I put together FORT in the first
place. It seems to be working pretty well in practice. 

On Mon, 15 Oct 2001, Brian Rogoff wrote:

> Hi,
>     What's become of the Caml Consortium? I put a lot of pressure on my
> manager to join and we've hardly heard anything since doing so. That was
> several months ago.
> 
> -- Brian
> 
> 
> -------------------
> Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
> To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr
> 

-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Michel Mauny]

Brian>     What's become of the Caml Consortium?

Patrick> I can understand motivations to keep quiet about plans but I think it
Patrick> would really help to open up this information. 

Well, it is not that the Consortium is willing to keep its plans
secret, but more that it is starting slowly. We have 3 members so far,
and we'd like to have a few more before the end of this year.

Anyway, the Consortium activities will start at the beginning of 2002,
and we should have our first formal meeting around December
2001/January 2002.

The Consortium Web pages

    http://caml.inria.fr/consortium/

have a new look (to be improved, of course, but better that before),
and all information will be available from there (members, actions,
and so on). Unless members disagree with me, the Consortium's
activities shall be public.

-- Michel
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Brian Rogoff]

On Sat, 20 Oct 2001, Michel Mauny wrote:
> Brian>     What's become of the Caml Consortium?
>
> Patrick> I can understand motivations to keep quiet about plans but I think it
> Patrick> would really help to open up this information.
>
> Well, it is not that the Consortium is willing to keep its plans
> secret, but more that it is starting slowly. We have 3 members so far,
> and we'd like to have a few more before the end of this year.

That number is embarassingly small. I've seen a number of posts here from
people at large companies. It is amazing that so few of these companies
are willing to join, and that relatively small companies like Artisan,
where I work, and FluxMedia, should be taking a leadership role.

Next time someone whines on the list about "O when can functional
programming languages be accepted in industry", I think my answer will be
something like, "When functional programmers quit whining and start
actually trying to make their favorite language successful". Some part of
that task for industrial programmers is joining the Consortium. Surely
there are more than three companies using OCaml?

Thanks for getting the ball rolling Michel, and hopefully some more
lurkers will change their status soon.

-- Brian


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Julian Assange]

> > secret, but more that it is starting slowly. We have 3 members so far,
> > and we'd like to have a few more before the end of this year.

The stalking ground is this mailinglist. You should be sending out
invitations to the list once a month. It is only through repeated
exposure that the idea will sink in.

--
 Julian Assange        |If you want to build a ship, don't drum up people
                       |together to collect wood or assign them tasks and
 proff@iq.org          |work, but rather teach them to long for the endless
 proff@gnu.ai.mit.edu  |immensity of the sea. -- Antoine de Saint Exupery
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

On Fri, 19 Oct 2001, Brian Rogoff wrote:
> That number is embarassingly small. I've seen a number of posts here
> from people at large companies. It is amazing that so few of these
> companies are willing to join, and that relatively small companies like
> Artisan, where I work, and FluxMedia, should be taking a leadership
> role.

Unfortunate as it is, but I am absolutely not surprised by the low number
of members in the Consortium, and I already explained the reasons here
when it began to form.

While three small companies have stepped in (most likely with the minimum
investment), the big rest is simply free riding. One can assume that
most industrial companies share similar needs with respect to OCaml.
Since they do not win opportunities by joining the Consortium (rather
lose them = money), because other members are likely to do the "home work"
for the rest anyway, it's a rational decision to stay outside.

It may even be the case that the mentioned companies only joined
in, because they have so specific needs that it is unlikely that
other companies will solve their problems in the Consortium. So the
realistic assumption is that the current three members haven't joined
the Consortium out of altruism and love for OCaml, but merely out of
very specific selfish reasons, which may not necessarily be for the
benefit of the whole OCaml-community. I am not accusing anybody here,
it's just one possible, rational explanation...

I'd be very, very surprised if the situation improved significantly in
the future. Unless INRIA finds a way to let people benefit from being
members of (= having rights in) the Consortium irrespective of the direct
benefit of "produced goods", in other terms, as long as there is no way to
invest for financial benefit, we won't see any change here anytime soon.

Therefore, I still propose that membership rights, whose amount of
control must be clearly defined, be permanent and tradeable. The last
property may require infrastructure that INRIA isn't allowed to build up
or use out of legal considerations, I don't know. But if it is possible,
you can bet that a significantly higher amount of money can be raised
for the future development of OCaml.

To answer Brian here: it's not the fault of the companies that haven't
yet joined in. It's probably just that the current scheme may not be
the most appropriate one for our goals...

Regards,
Markus Mottl

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Brian Rogoff]

On Sat, 20 Oct 2001, Markus Mottl wrote:
> It may even be the case that the mentioned companies only joined
> in, because they have so specific needs that it is unlikely that
> other companies will solve their problems in the Consortium. So the
> realistic assumption is that the current three members haven't joined
> the Consortium out of altruism and love for OCaml, but merely out of
> very specific selfish reasons, which may not necessarily be for the
> benefit of the whole OCaml-community. I am not accusing anybody here,
> it's just one possible, rational explanation...

I think altruism and love for OCaml is a lot closer than very specific
selfish reasons. That's still not quite right, maybe idealism and civic
virtue come closer?

I can't speak for the other companies, but I think that's not an
inaccurate picture of where I work.

I'm not convinced that my, or anyone else's, behavior is entirely rational,
or at least that the objective function and even the decision variables aren't
somewhat arbitrary. So the model of a corporation as a purely money optimizing
entity is inaccurate. But I'll stop here, this list isn't the place for a
discussion of my world view; if you want that, join the Consortium and
I'll send you a private, copyrighted e-mail :-).

> To answer Brian here: it's not the fault of the companies that haven't
> yet joined in. It's probably just that the current scheme may not be
> the most appropriate one for our goals...

No doubt the process and goals of the Consortium can be tuned.

Thanks for your altruistic work on behalf of OCaml, Markus!

-- Brian


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

On Sat, 20 Oct 2001, Brian Rogoff wrote:
> I think altruism and love for OCaml is a lot closer than very specific
> selfish reasons.

I love OCaml, because it helps me solve my problems much more easily. Does
this make my love selfish? ;)

> That's still not quite right, maybe idealism and civic virtue come
> closer?

The older I get the less I am sure what "idealism" is supposed to mean.
Especially the last weeks have made me much older.

Is it: "Act according to what the majority wants even if this does not
correlate with your desires."?

Hm, I am not sure whether I can identify with this without
restrictions. E.g. when I desperately need better and more ADT-libraries,
why should I finance GUI-building tools only because the majority
wants them? - A goal conflict...

Furthermore, the majority may know what it wants, but it may not know
what it needs. It could well be that it also needs better ADT-libraries
rather than GUI-building tools, but due to lack of intelligence they
choose unwisely. - Bounded rationality...

Even if the goals coincided for some miraculous reasons, there could
be a lot of dispute concerning the concrete way to reach them. A means
conflict...

As you see, there could be plenty of causes why a Consortium from which
one cannot exit without a complete loss of investment may be a rather
unfavourable choice. Which might, again, explain its current state.

(Sorry, I am in an illusion-smashing mood today ;)

> I'm not convinced that my, or anyone else's, behavior is entirely
> rational, or at least that the objective function and even the decision
> variables aren't somewhat arbitrary.

People act so as to maximize their utility function, whatever this
may be: in economics this notion is so general that it can explain
any kind of behaviour, which gives it little significance in practice.
But restricting its applicability is impossible without making value
judgements.

> So the model of a corporation as a purely money optimizing entity
> is inaccurate.

If people were maximizing their monetary assets only, they'd all be
starving. Since not all are doing this, only some are. The two groups
do not necessarily overlap.

> But I'll stop here, this list isn't the place for a discussion of my
> world view; if you want that, join the Consortium and I'll send you
> a private, copyrighted e-mail :-).

No spam, please! ;)

> No doubt the process and goals of the Consortium can be tuned.

This will have to happen in any case if they want to be more successful.
I was actually surprised that there was no invitation for discussion
before its foundation. E.g., it seems to me that the fees were set
quite arbitrarily. Some initial "market analysis" as to how much how
many people would be prepared to donate might have turned out useful to
maximize the income of the Consortium.

> Thanks for your altruistic work on behalf of OCaml, Markus!

This is only a misconception: it's out of purely selfish reasons, sold
under the label "altruism"... ;)

Regards,
Markus

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Rolf Wester]

On Mon, 22 Oct 2001, Markus Mottl wrote: 
> On Sat, 20 Oct 2001, Brian Rogoff wrote:
> > I think altruism and love for OCaml is a lot closer than very specific
> > selfish reasons.
> 
> I love OCaml, because it helps me solve my problems much more easily. Does
> this make my love selfish? ;)
> 
I think the lack of willingness to support OCaml financially is correlated with
the lack of perception of the extent to which OCaml makes software development 
easier and as a consequence cheaper. Those who use a programming language in their daily
work are in many cases not those who decide on money. And in order to convince
someone to spend money for the development of a product like OCaml 
(not mainstream, almost no one knows it, there is not even an English text book, 
no commercial support)  one must have very good arguments. I think what could
be very helpful is a detailed list of OCaml's strength (and weeknesses if any) 
compared to languages like C++, Java and also compared to other ML-implementations,
Lisp, Haskell, Clean etc.? This list should also include real world examples (not to
complicated) to demonstrate OCaml's benefits. 

Another point could be that people are more likely to spend money for getting a product
or support than for supporting someone else to develop a programming language. So why not 
taking a fee for commercial use of OCaml or for support (Clean, Python, Lisp)?  

Rolf Wester


-------------------------------------
Rolf Wester
rolf.wester@ilt.fraunhofer.de
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Michel Mauny]

Markus,

Markus Mottl wrote/écrivait (Oct 20 2001, 05:29PM +0200):

> While three small companies have stepped in

I wouldn't say that Dassault-Aviation is such a small company (~ 9000
employees, as far as I know).

> So the realistic assumption is that the current three members
> haven't joined the Consortium out of altruism and love for OCaml,
> but merely out of very specific selfish reasons, which may not
> necessarily be for the benefit of the whole OCaml-community. I am
> not accusing anybody here, it's just one possible, rational
> explanation...

I don't see the point in speculating on why current members joined the
Consortium. Instead, let them explain why they joined, in case they
want to do so. And it's no problem if they don't want to explain.

Furthermore, I'm not sure that such assumptions about the current
members and their "very specific selfish reasons which may not
necessarily be for the benefit of the whole OCaml-community" in this
mailing list, are of great help for attracting new members.

I really believe that we can have a useful and successful consortium
even with a small number of companies at the beginning. Of course, 3
are not enough, but I think we can attract a few more, and start
something that will be useful for the whole community. Not only the
developments and promotion of OCaml are of general interest for the
community, but the existence of the group itself could be a rather
strong argument when a decision of ``choosing OCaml or not'' has to be
made. Especially when the manager is the only one remaining to be
convinced.

I should probably at that point remind the list with the URL of the
membership request :-)

  http://caml.inria.fr/consortium/documents/form.shtml

and invite all of you to convince their management to join the
Consortium. We'd like to plan a first formal meeting for either
mid-december or january.

For non-European users, I understand that the membership process
(payment, in particular) can be a bit painful (contract signed by both
INRIA and the Member, then invoice sent by INRIA, and then payment by
the Member). I can try to do my best to alleviate it, but I'm afraid
the French rules applying to state-funded institutes such as ours are
rather unflexible, unfortunately.

Best regards,

-- 
Michel
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Ken Rose]

Michel Mauny wrote:
> I should probably at that point remind the list with the URL of the
> membership request :-)
> 
>   http://caml.inria.fr/consortium/documents/form.shtml
> 
> and invite all of you to convince their management to join the
> Consortium. We'd like to plan a first formal meeting for either
> mid-december or january.

What about a reasonably priced individual membership?  Something in the
range of 50-100 dollars or euros might be reasonable.  I'm certainly not
willing to personally cough up 2000 euros.  It also seems likely to
attract the people who are really using ocaml, which seems like mainly
individuals doing stealthy things inside of bigger organizations.

 - ken
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

Michel,

On Mon, 22 Oct 2001, Michel Mauny wrote:
> Markus Mottl wrote/écrivait (Oct 20 2001, 05:29PM +0200):
> I wouldn't say that Dassault-Aviation is such a small company (~ 9000
> employees, as far as I know).

Right, fair enough...

> I don't see the point in speculating on why current members joined
> the Consortium.

But I do indeed, because it is important to know about the chances that
more members are going to join and whether one can improve these chances.
Just to make sure there is no misunderstanding: I do think that some
kind of industrial (or even private) financial support is very important
for the future of OCaml and that a Consortium would be a good idea. The
question is only how to make the Consortium attractive to people.

> Instead, let them explain why they joined, in case they want to do
> so. And it's no problem if they don't want to explain.

Most likely, because some tireless heroes managed to convince them
(btw.: my congratulations for their great lobbying work!). After all,
the fees are not so high that companies couldn't afford them. The problem
is rather opportunity costs: why join this Consortium and not another?
Big companies usually have many alternative opportunities so we better
make sure that they stay with us and not go elsewhere...

> Furthermore, I'm not sure that such assumptions about the current
> members and their "very specific selfish reasons which may not
> necessarily be for the benefit of the whole OCaml-community" in this
> mailing list, are of great help for attracting new members.

I had expected that one might misunderstand my argument here. The word
"selfish" has a negative touch in most people's eyes: to me it basically
means "they think they will benefit from it". This does absolutely not
mean that they see a benefit in other members not having one. In fact,
they surely know that the benefit of the whole OCaml-community is
to some extent correlated to theirs. The question is how strong this
correlation is.

There is definitely a point where interest conflicts can arise, and then
a donation scheme may not be able to keep members. Nobody will join any
kind of interest group without having an interest in it. If INRIA were
a charity, a scheme that builds on donations would be fine, but this is
not the case here.

Also, I don't fear that any of my ramblings will prevent anybody who
is decided from joining the Consortium (I wouldn't write this much if I
didn't take this issue very serious). But some change (if it were legally
possible) to the current statutes of the Consortium might attract many
more members. This was my primary interest in this discussion.

> I really believe that we can have a useful and successful consortium
> even with a small number of companies at the beginning. Of course,
> 3 are not enough, but I think we can attract a few more, and start
> something that will be useful for the whole community.

The Consortium is definitely better than no consortium at all - no
objection against it as such! If I take a look around, consortia in
this field (languages, compilers) are usually formed to define industry
standards. Not being able to influence the latter to their favour can
be extremely costly for companies, which is a strong incentive for
joining. Unfortunately, OCaml doesn't seem to be widespread enough to
justify this. (At least at the moment ;)

> Not only the developments and promotion of OCaml are of general interest
> for the community, but the existence of the group itself could be a
> rather strong argument when a decision of ``choosing OCaml or not''
> has to be made. Especially when the manager is the only one remaining
> to be convinced.

Sure! This, however, requires that the Consortium consists of a
significant number of influential members. It's great for the popularity
of OCaml that Dassault-Aviation has joined, and I hope that they stay! It
certainly won't hurt to give them some more arguments why continuous
financial support is a good idea (= will benefit them, too). Moral
support alone will probably not be enough to make OCaml really popular...

> For non-European users, I understand that the membership process
> (payment, in particular) can be a bit painful (contract signed by both
> INRIA and the Member, then invoice sent by INRIA, and then payment by
> the Member). I can try to do my best to alleviate it, but I'm afraid
> the French rules applying to state-funded institutes such as ours are
> rather unflexible, unfortunately.

I feared that this would be a major obstacle. The private and public
sectors are usually strictly separated (same here in Austria), which makes
it very difficult to combine their advantages (efficiency and sustained
long term investment). But maybe such an attempt would just foster their
combined disadvantages and create a myopic monster of inefficiency,
who knows? ;)

Anyway, sometimes I really wish that my analyses are wrong, and if not
in the case that concerns the OCaml-Consortium, where else?

Best regards,
Markus

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Xavier Leroy]

> > Well, it is not that the Consortium is willing to keep its plans
> > secret, but more that it is starting slowly. We have 3 members so far,
> > and we'd like to have a few more before the end of this year.
> 
> That number is embarassingly small. 

It is certainly lower than we initially expected, but not all is
lost.  From our preliminary meetings, there are other companies (small
and big) that are interested, but not ready to sign in yet.  Remember
that the economic situation in the IT industry isn't so good right
now: many start-ups are literally strapped for cash, and large
companies tend to just "wait and see if it gets better"...  

> Thanks for getting the ball rolling Michel, and hopefully some more
> lurkers will change their status soon.

We sure hope so!

- Xavier Leroy
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

On Mon, 22 Oct 2001, Ken Rose wrote:
> What about a reasonably priced individual membership?  Something in the
> range of 50-100 dollars or euros might be reasonable.  I'm certainly
> not willing to personally cough up 2000 euros.  It also seems likely
> to attract the people who are really using ocaml, which seems like
> mainly individuals doing stealthy things inside of bigger organizations.

If a donation scheme is the only option, I'd strongly encourage the
Consortium to consider lower minimum fees. I think that OCaml is used by
already numerous individuals, but not by too many industrial companies.
Many small contributions could outweigh few large ones!

A simple scheme that introduces a reasonably low minimum subscription fee
and weights votes in (online?) ballots depending on the donation sum could
be really an option. It may also be more democratic than just allowing
"full members" to vote. Separate treatment of donations of industrial
members and individuals could still be possible.

Would a minimum of 100 Euros for individuals be reasonable? This is
lower than the subscription rate of some province newspapers, but here
one gets the right to vote for OCaml - wouldn't this be a bargain? ;)

Regards,
Markus Mottl

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Daniel de Rauglaudre]

Hi,

A reason to help the Consortium to work is to think of what would
happen if we stop OCaml, here, at INRIA, intentionnaly or not. Not
only a question of loving OCaml.

-- 
Daniel de RAUGLAUDRE
daniel.de_rauglaudre@inria.fr
http://cristal.inria.fr/~ddr/
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

Rolf Wester schrieb am Montag, den 22. Oktober 2001:
> I think the lack of willingness to support OCaml financially is
> correlated with the lack of perception of the extent to which OCaml
> makes software development easier and as a consequence cheaper.

That's not quite what I think, but it certainly also explains to some
extent why it is so difficult to gain more members for OCaml in general
and for the Consortium in particular.

> Those who use a programming language in their daily work are in many
> cases not those who decide on money.

Which is, I fear, most often a good idea... ;)

> And in order to convince someone to spend money for the development
> of a product like OCaml (not mainstream, almost no one knows it,
> there is not even an English text book, no commercial support) one
> must have very good arguments.

And it would be unwise to believe that it's only technical arguments
that are considered here. There simply must be economic incentives to
convince companies. To say it clearly: a manager who doesn't consider
the latter is a bad manager. We should really try to avoid our natural
mindset of enthusiastic technicians or scientists and put ourselves into
the role of a manager who is responsible for his investment decisions.

> I think what could be very helpful is a detailed list of OCaml's
> strength (and weeknesses if any) compared to languages like C++,
> Java and also compared to other ML-implementations, Lisp, Haskell,
> Clean etc.? This list should also include real world examples (not to
> complicated) to demonstrate OCaml's benefits.

This does not work. I am sure that most of us have already tried such
strategies, but they do not convince, because all competing languages
use buzzwords and lists of "advantages". If you want to convince people,
write a killer-app in their respective field of interest, otherwise they
won't even listen.

> Another point could be that people are more likely to spend money
> for getting a product or support than for supporting someone else to
> develop a programming language. So why not taking a fee for commercial
> use of OCaml or for support (Clean, Python, Lisp)?

Taking a look at companies that base their business on open source,
I wouldn't say that their strategy "sell services" was particularly
successful up to now. Mostly, because they overlooked that services have
a rather strong impact on costs and therefore on profit. Don't forget,
as soon as you demand fees for your products, you are liable for them
and no kind of licence will get you around this (at least not in the
countries I know).

Regards,
Markus Mottl

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

Hi,

On Tue, 23 Oct 2001, Daniel de Rauglaudre wrote:
> A reason to help the Consortium to work is to think of what would
> happen if we stop OCaml, here, at INRIA, intentionnaly or not. Not
> only a question of loving OCaml.

If I were the only one (or one of three) to join the Consortium, paying
a not insignificant fee, and if INRIA decided to shut down OCaml shortly
after due to lack of financial support, I'd have lost both my money and
support for OCaml.  Sometimes I hate game theory... ;)

But a realistic scenario would probably be:

  Xavier and Daniel (to INRIA):

    We know you have shut down project Cristal, but we'd like to continue
    work on OCaml and CamlP4. What do you think?

  INRIA:   Hm, is five Euro per hour ok?

  Daniel:  That's just fine!

  Xavier:  Same for me!

  Xavier (whispering to Daniel):

    Phew, we are lucky! They could have asked for twice as much!

I hope I haven't inspired the management of INRIA to seek very alternative
forms of financing its business ;)

Regards,
Markus

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Rolf Wester]

> Rolf Wester schrieb am Montag, den 22. Oktober 2001:
> > I think the lack of willingness to support OCaml financially is
> > correlated with the lack of perception of the extent to which OCaml
> > makes software development easier and as a consequence cheaper.
> 
> That's not quite what I think, but it certainly also explains to some
> extent why it is so difficult to gain more members for OCaml in general
> and for the Consortium in particular.
> 
> > Those who use a programming language in their daily work are in many
> > cases not those who decide on money.
> 
> Which is, I fear, most often a good idea... ;)
> 
> > And in order to convince someone to spend money for the development
> > of a product like OCaml (not mainstream, almost no one knows it,
> > there is not even an English text book, no commercial support) one
> > must have very good arguments.
> 
> And it would be unwise to believe that it's only technical arguments
> that are considered here. There simply must be economic incentives to
> convince companies. To say it clearly: a manager who doesn't consider
> the latter is a bad manager. We should really try to avoid our natural
> mindset of enthusiastic technicians or scientists and put ourselves into
> the role of a manager who is responsible for his investment decisions.
> 
The good reasons to use OCaml cannot be "I love it" or "it's fun to program 
in OCaml" but that it makes me more productive (at least when I'm paid
for what I'm doing).

> > I think what could be very helpful is a detailed list of OCaml's
> > strength (and weeknesses if any) compared to languages like C++,
> > Java and also compared to other ML-implementations, Lisp, Haskell,
> > Clean etc.? This list should also include real world examples (not to
> > complicated) to demonstrate OCaml's benefits.
> 
> This does not work. I am sure that most of us have already tried such
> strategies, but they do not convince, because all competing languages
> use buzzwords and lists of "advantages". If you want to convince people,
> write a killer-app in their respective field of interest, otherwise they
> won't even listen.
> 
But even if you have a killer-app written in OCaml you will still have to explain 
to your manager (and even more your colleagues) why you would not have 
been able to write this app in C++ or Java (or why it would have been much
more effort to do it in another language). I think that because competing 
languages are advertised with buzzwords and their list of "advantages" 
OCaml should be advertised too. OCaml's features should be compared to 
other languages and statements made concerning other languages should 
objectively be analyzed and criticized. And if for a certain kind of application
another language is more suitable this should also be clearly stated. I think
this could help those who are looking for an alternative to the main stream
languages and those who have to argue in favour of using OCaml.  
 
> > Another point could be that people are more likely to spend money
> > for getting a product or support than for supporting someone else to
> > develop a programming language. So why not taking a fee for commercial
> > use of OCaml or for support (Clean, Python, Lisp)?
> 
> Taking a look at companies that base their business on open source,
> I wouldn't say that their strategy "sell services" was particularly
> successful up to now. Mostly, because they overlooked that services have
> a rather strong impact on costs and therefore on profit. Don't forget,
> as soon as you demand fees for your products, you are liable for them
> and no kind of licence will get you around this (at least not in the
> countries I know).
> 
Good point.

Regards

Rolf Wester

-------------------------------------
Rolf Wester
rolf.wester@ilt.fraunhofer.de
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

RE: [Caml-list] Whither the Caml Consortium?

[Mattias Waldau]

It is very difficult to get enough money to be able to support, maintain and
sell a programming language. Just see how difficult it is for ML companies
(2 failed attempts) and Lisp. (Lisp companies typically charge $5000 and
upwards per developer).

(SICStus Prolog actually lives on licenses and it works.)

That means that when Inria stops to support Ocaml, the language will most
likely die unless the open source community takes over. The code the Ocaml
is reasonable documentet which makes it possible to maintain it.

Regarding a killer-application in Ocaml to show that Ocaml is best, we have
to find a program that MANY people need. Just look how difficult it is to
create a good email program for Linux (Evolution is at last getting ready).
Such an application could be built using Ocaml instead. It would work, no
buffer overuns would reduce the number of security glitches. The only
problem is that almost all killer application has graphical user interfaces,
and I still haven't seen a good looking Ocaml-program (any pointers?).

Unison could be a killer application, but it lacks all user friendliness.
However, not so many people need syncronization.

/mattias

P.s. As a CTO for a programming company, I would say that $500-1000 per
developer and year would be a resonable licensing fee.

-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Mike Leary]

On Tue, Oct 23, 2001 at 12:22:10PM +0200, Markus Mottl wrote:
>If you want to convince people,
> write a killer-app in their respective field of interest, otherwise they
> won't even listen.

It might be a good idea to put in place a mechanism for collecting "use
cases", where a company has used OCaml and derived some happy (perhaps
unexpected) benefit in the real world.  Nothing succeeds like success and
all...

-- 
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

Rolf Wester schrieb am Dienstag, den 23. Oktober 2001:
> The good reasons to use OCaml cannot be "I love it" or "it's fun to
> program in OCaml" but that it makes me more productive (at least when
> I'm paid for what I'm doing).

Everybody claims that their language is the most productive one,
but not everbody can prove it. To be honest: it is much easier to
"prove" this for Java than for OCaml, because there are legions of more
Java-programmers with zillions of projects.  So Mr. J. can say: "Look
at the many cool things that have been done in Java!". Whether these
things were produced in a short time or how much effort was necessary
is usually not observable anyway.

The pure quantitative lack of significant OCaml-projects (on a
comparative scale) makes it difficult to argue, which places us into
the chicken-and-egg problem. So we better write code rather than lament
about the lack thereof...

> But even if you have a killer-app written in OCaml you will still
> have to explain to your manager (and even more your colleagues) why
> you would not have been able to write this app in C++ or Java (or why
> it would have been much more effort to do it in another language).

Sure! But having a "constructive" proof of your claim is more convincing
than the claim alone. Especially for managers, who have a tough time
estimating the validity of your theoretical claims in fields they are
not experts in.

> OCaml's features should be compared to other languages and statements
> made concerning other languages should objectively be analyzed and
> criticized.

And some Java-guru would then "objectively" analyze things from his
point of view...

> And if for a certain kind of application another language is more
> suitable this should also be clearly stated.

No, never say anything bad about your product. Never! We all know that
this is dishonest, but that's the way Java, Windows, VB, etc. have
conquered the market. There is good reason why I have switched to a
technical field from business... :(

If you want to do marketing, then do marketing, not science. You'll
have to play by the rules of psychology then rather than use technical
measures.

Regards,
Markus Mottl

P.S.:  Even though it is much more effective, I don't want to do
       "marketing" for OCaml: I will still continue trying to convince
       by honest arguments.

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

On Wed, 24 Oct 2001, Mattias Waldau wrote:
[snip stuff on which I generally agree]
> Unison could be a killer application, but it lacks all user
> friendliness.

It depends on your notion of "user friendliness": for the tasks it was
mainly aimed at, it seems to be doing an excellent job, IMHO.

> However, not so many people need syncronization.

I don't think this is true: when I pushed unison at our institute,
it was very quickly taken up by a quite significant number of staff,
both Unix and Windows users. Today it is already available on all of our
machines. It's the only kid in town that can really claim to efficiently,
reliably and securely synchronize both ways across multiple platforms.

Even though this hasn't made anybody switch to OCaml, people have at
least stopped giving me benign smiles when I mention it ;)

Regards,
Markus Mottl

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Cuihtlauac ALVARADO]

> Regarding a killer-application in Ocaml to show that Ocaml is best, we have
> to find a program that MANY people need. Just look how difficult it is to
> create a good email program for Linux (Evolution is at last getting ready).
> Such an application could be built using Ocaml instead. It would work, no
> buffer overuns would reduce the number of security glitches. The only
> problem is that almost all killer application has graphical user interfaces,
> and I still haven't seen a good looking Ocaml-program (any pointers?).

Just a 30 seconds dream : a small, fast and robust web
browser... Wouldn't you like to have a GC in Netscape ? Wouldn't you
like to use powerfull tools able to program a defensive parsing for
all the HTTP/Javascript garbage you can find out there ? Isn'it what
*every* people need ?

I can dream no other definitive killer app for Ocaml.

-- 
Cuihtlauac ALVARADO - France Telecom R & D - DTL/MSV/MFL
2, avenue Pierre Marzin - 22307 Lannion Cedex - France
Tel: +33 2 96 05 32 73 - Mob: +33 6 08 10 80 41 - Fax: +33 2 96 05 39 45
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Xavier Leroy]

> What about a reasonably priced individual membership?  Something in the
> range of 50-100 dollars or euros might be reasonable.  I'm certainly not
> willing to personally cough up 2000 euros.

This issue was discussed on this list a while ago.  To summarize:

The current setup of the Consortium is definitely geared towards
corporate members, not individuals.  Raising money to fund future
developments is one goal of the Consortium, but an equally important
goal is to answer the age-old question "what other corporations are
using it?" that we get from prospective industrial users.  The best
answer to this question is: "well, here are some high-tech companies
that no only use it, but support it financially by being members of
the consortium".  (An even better answer would be: "And by the way,
your main competitor is already a member"; instant adhesion guaranteed :-)

There are other reasons for not soliciting individual memberships.
One is that OCaml users, as individuals, already do a great job of
supporting the language, via code contributions, feedback, advocacy,
volunteer's work (e.g. the collective translation of the OCaml book),
etc.  I'd feel a bit guilty asking them for money on top of that...

The second reason is less idealistic: given the legal structure of the
Consortium, registering a new member requires some paperwork on
INRIA's side, whose effective cost is likely to be in the $50-100
range...

- Xavier Leroy
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Markus Mottl]

On Wed, 24 Oct 2001, Xavier Leroy wrote:
> The best answer to this question is: "well, here are some high-tech
> companies that no only use it, but support it financially by being
> members of the consortium".  (An even better answer would be: "And by
> the way, your main competitor is already a member"; instant adhesion
> guaranteed :-)

This intimitating psychological effect is a good argument ;)

> The second reason is less idealistic: given the legal structure of the
> Consortium, registering a new member requires some paperwork on INRIA's
> side, whose effective cost is likely to be in the $50-100 range...

This surely depends on the number of membership requests. If only one
individual applies, this may be true, but average costs per request are
likely to be much lower if 100 do so. Why don't we run an opinion poll
on this to see whether enough people have interest? Yahoo Groups offers
such poll functionality for free, and it may come handy more than once
when INRIA wants to get mass feedback from OCaml-users...

Regards,
Markus Mottl

-- 
Markus Mottl                                             markus@oefai.at
Austrian Research Institute
for Artificial Intelligence                  http://www.oefai.at/~markus
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Sven]

On Mon, Oct 22, 2001 at 10:36:50AM -0700, Ken Rose wrote:
> Michel Mauny wrote:
> > I should probably at that point remind the list with the URL of the
> > membership request :-)
> > 
> >   http://caml.inria.fr/consortium/documents/form.shtml
> > 
> > and invite all of you to convince their management to join the
> > Consortium. We'd like to plan a first formal meeting for either
> > mid-december or january.
> 
> What about a reasonably priced individual membership?  Something in the
> range of 50-100 dollars or euros might be reasonable.  I'm certainly not
> willing to personally cough up 2000 euros.  It also seems likely to
> attract the people who are really using ocaml, which seems like mainly
> individuals doing stealthy things inside of bigger organizations.

My understanding of this is that this kind of thing would cost more in
paperwork than the real benefit to the consortium, that is why i was
contemplating creating a association in france to group several users
which would, once we are able to sum up the 2000 euros, become one member of
the consortium. I was aiming at a fee of around 50 Euros, this would need at
least 40 members, or for some of them to donate more, or other means of
raising founds.

Sadly, i have not had time to fully devote myself to fullfiling this in the
last month, but now that the subject has surfaced again on this list, it may
be the right time for that.

To create an "association loi 1901" here in strasbourg, we would need to define
the status, and have at least 7 founding members, as well as people willing to
take the administratives jobs like secretary and the financial aspect. It can
happen relatively quickly once we have enough members.

So, unless someone else would like to do the same, it would be nice if
interrested people send me a mail about this, (personal mail at :
luther@dpt-info.u-strasbg.fr), and once a first step of discution is reached,
we will hold a founding meeting, either trough IRC or something such, or
trough a mailing list i will create, the second solution would be maybeeasier
on non european persons.

Friendly,

Sven Luther
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[georges mariano]

On Wed, 24 Oct 2001 17:57:07 +0200  Sven <luther@dpt-info.u-strasbg.fr> wrote:
 
 > contemplating creating a association in france to group several users
 > which would, once we are able to sum up the 2000 euros, become one member of
 > the consortium. I was aiming at a fee of around 50 Euros, this would need at
 > least 40 members, or for some of them to donate more, or other means of
 > raising founds.
 
Hi Sven,
This idea (smaller fee and grouping people) does fit my needs and capabilities
much more better than the one proposed by the Consortium (do not forget
about __small__ companies !! There is place between 50 and 2000 euros ...
150, 200, hmm, so on...;-) 

Question : 
Suppose that such idea will reach the implementation stage,
how will be this Association considered by "THE" Consortium ? 

??
-- 
# mailto:Georges.Mariano@inrets.fr     tel: (33) 03 20 43 84 06   
# INRETS, 20 rue Élisée Reclus         fax: (33) 03 20 43 83 59   
# BP 317 -- 59666 Villeneuve d'Ascq       
# http://www3.inrets.fr/estas/mariano
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Joshua D. Guttman]

Xavier Leroy <xavier.leroy@inria.fr> writes:

>   
>   
>   The current setup of the Consortium is definitely geared towards
>   corporate members, not individuals.  Raising money to fund future
>   developments is one goal of the Consortium, but an equally
>   important goal is to answer the age-old question "what other
>   corporations are using it?" that we get from prospective
>   industrial users.
>   

A corporation is of course not an entity with a single point of view.
The people working on a particular project or in a particular area may
use OCaml, and want to support the Consortium, although their
management may not have much interest.  

For instance, in my own case, I have been using OCaml for several
years, and I have found a few colleagues who now also use it
effectively.  My management tolerates this peculiarity in me, so long
as I am not strident about it.  (:-).  I imagine there are other
people in similar situations.

For people like me, it would be good to have a level of consortium
membership that a single project could justify.  If there was
something in the three figure range (i.e. hundreds but not thousands
of dollars), I would simply make the arrangements and I'd never have
to convince anyone organizationally very distant from me.  But for a
consortium membership that costs thousands, I would have to convince
people I rarely interact with, and they would want to decide whether
OCaml should play some company-wide role, and probably it would be a
dead end.

Would the OCaml Consortium consider something like this?  

Cheers --

        Joshua 



-- 
	Joshua D. Guttman		<guttman@mitre.org> 
	MITRE, Mail Stop S119 
	202 Burlington Rd.		Tel:	+1 781 271 2654
	Bedford, MA 01730-1420 USA	Fax:	+1 781 271 8953
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Sven]

On Wed, Oct 24, 2001 at 04:18:24PM +0200, Cuihtlauac ALVARADO wrote:
> > Regarding a killer-application in Ocaml to show that Ocaml is best, we have
> > to find a program that MANY people need. Just look how difficult it is to
> > create a good email program for Linux (Evolution is at last getting ready).
> > Such an application could be built using Ocaml instead. It would work, no
> > buffer overuns would reduce the number of security glitches. The only
> > problem is that almost all killer application has graphical user interfaces,
> > and I still haven't seen a good looking Ocaml-program (any pointers?).
> 
> Just a 30 seconds dream : a small, fast and robust web
> browser... Wouldn't you like to have a GC in Netscape ? Wouldn't you
> like to use powerfull tools able to program a defensive parsing for
> all the HTTP/Javascript garbage you can find out there ? Isn'it what
> *every* people need ?
> 
> I can dream no other definitive killer app for Ocaml.

And ocam plugin for mozilla would already be a nice thing, and maybe not so
difficult to obtain.

That and a serie of easy to understand examples to show how to use it ...

I know we have a caml written app that will probably rewritten in java or
something else in order to have it able to be accesed via web browsers, for
demonstration purpose. We (not i personnally though) tried with javacaml, but
as expected it was too slow, and don't tell me about the various way to use
html as user interface, we are speaking graphical stuff here.

Friendly,

Sven Luther
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Sven]

On Wed, Oct 24, 2001 at 07:05:44PM +0200, georges mariano wrote:
> On Wed, 24 Oct 2001 17:57:07 +0200  Sven <luther@dpt-info.u-strasbg.fr> wrote:
>  
>  > contemplating creating a association in france to group several users
>  > which would, once we are able to sum up the 2000 euros, become one member of
>  > the consortium. I was aiming at a fee of around 50 Euros, this would need at
>  > least 40 members, or for some of them to donate more, or other means of
>  > raising founds.
>  
> Hi Sven,
> This idea (smaller fee and grouping people) does fit my needs and capabilities
> much more better than the one proposed by the Consortium (do not forget
> about __small__ companies !! There is place between 50 and 2000 euros ...
> 150, 200, hmm, so on...;-) 
> 
> Question : 
> Suppose that such idea will reach the implementation stage,
> how will be this Association considered by "THE" Consortium ? 

As i understood, we will be a full member (if we manage to pat 2000 euros), we will
send a representative to the meetings, and our opinion will be heard as well
asq others, i guess. If we manage to get lot of money, then we could even be
considered as 2 or more members i guess, not sure though.

That said, as we are not enough people yet i don't know, but we would need to
reach a sort of agreement on what we will propose or so on. Probably using a
(using ssh for security or so on) electronic vote or something, well, we could
use debian's vote handling code, or write something of our own, since i am not
really sure all that complicated condorcet stuff is really needed for us.

There will also be some initial paperwork involved, and pgp/gpg key exchange,
maybe some solution needs to be found for people not easily able to transfer
found to france and so on, but this can be discussed in the founding
conference, or whatever this will be called, and we can speak about it then.

I will try to make up a more formal announcement, both in french and english,
this week end.

Anyway, if we have not enough members interrested, it will not work, ...

Friendly,

Sven Luther
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Frank Atanassow]

Rolf Wester wrote (on 23-10-01 15:41 +0200):
> But even if you have a killer-app written in OCaml you will still have to explain 
> to your manager (and even more your colleagues) why you would not have 
> been able to write this app in C++ or Java (or why it would have been much
> more effort to do it in another language). I think that because competing 
> languages are advertised with buzzwords and their list of "advantages" 
> OCaml should be advertised too. OCaml's features should be compared to 
> other languages and statements made concerning other languages should 
> objectively be analyzed and criticized. And if for a certain kind of application
> another language is more suitable this should also be clearly stated. I think
> this could help those who are looking for an alternative to the main stream
> languages and those who have to argue in favour of using OCaml.  

Not that I want to become embroiled in a discussion on the finer points of
propaganda---oops, `advocacy', but you can find such a buzzword list here:

  http://pauillac.inria.fr:80/caml/FAQ/general-eng.html

and here (for SML, but all the points still hold except the last two):

  http://cm.bell-labs.com/cm/cs/what/smlnj/sml.html

Suitable for pointing your colleague to, if (s)he wants to know what (CA)ML is
all about.

-- 
Frank Atanassow, Information & Computing Sciences, Utrecht University
Padualaan 14, PO Box 80.089, 3508 TB Utrecht, Netherlands
Tel +31 (030) 253-3261 Fax +31 (030) 251-379
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Hendrik Tews]

Hi,

Sven writes:
   Date: Wed, 24 Oct 2001 17:57:07 +0200
   Subject: Re: [Caml-list] Whither the Caml Consortium?
   
   i was contemplating creating a association in france to group
   several users which would, once we are able to sum up the 2000
   euros, become one member of the consortium.

Are you sure that your application to become a member of the
consortium will be successful? As I understand it, INRIA decides
on the applications.

My impression from the current thread (and from previous ones on
the consortium) is that there is some confusion about the real
purpose of the consortium (i.e., INRIAS position) and what
members of this list think about the consortium. Some members on
the mailing list (myself included) would like to join the
consortium to influence the development of Ocaml. But for reasons
that I have not been able to grasp, INRIA seems to be not really
keen on seeing us ``small'' ocaml users in the consortium.

I would therefore suggest that we first discuss what we want to
achieve by becomming a consortium member. Then we can see how to
make our interests compatible with INRIAS constraints. If we
reach some consensus here, it is probably easier to get the
programming work done ourselfs, instead of convincing the
consortium to pay somebody to do the job.

I, for instance, would like some improvements for ocaml for which
it is absolutely impossible to get scientific reward, and which
are therefore very difficult to get implemented by the ocaml
developers. Take for instance better error diagnostics from the
ocaml parser or Thierry Bravier's ocamlyacc patch
(http://caml.inria.fr/archives/199712/msg00020.html). One way to
get these things done is to join the consortium ...

Another possibility is to write a patch (like Thierry did) and
get it into the ocaml distribution (what Thierry not achieved).
However, it is not clear (at least to me) what requirements have
to be met, to get such an improvement accepted by the ocaml
developers.

I think, what is needed is that the ocaml developers give some
guidelines on how we ``small'' ocaml users should proceed, if we
want to contribute something to the ocaml kernel. Then there
would be no need for us to join the consortium.


Bye,

Hendrik
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Fabrice Le Fessant]

>  I, for instance, would like some improvements for ocaml for which
>  it is absolutely impossible to get scientific reward, and which
>  are therefore very difficult to get implemented by the ocaml
>  developers. Take for instance better error diagnostics from the
>  ocaml parser or Thierry Bravier's ocamlyacc patch
>  (http://caml.inria.fr/archives/199712/msg00020.html). One way to
>  get these things done is to join the consortium ...
>  
>  Another possibility is to write a patch (like Thierry did) and
>  get it into the ocaml distribution (what Thierry not achieved).
>  However, it is not clear (at least to me) what requirements have
>  to be met, to get such an improvement accepted by the ocaml
>  developers.
>  
>  I think, what is needed is that the ocaml developers give some
>  guidelines on how we ``small'' ocaml users should proceed, if we
>  want to contribute something to the ocaml kernel. Then there
>  would be no need for us to join the consortium.

If you want to contribute to ocaml, one possibility is to put your
libraries and patches in the CDK (Caml Development Kit). Maybe you
don't know, but the CDK is distributed with a patched version of the
ocaml compiler. As a consequence, the ocamlyacc patch for example can
be applied to the CDK ocaml compiler, and distributed in next
releases, so that it is possible to test whether a patch is really
useful or nor before complete integration in the original ocaml
compiler. I think the CDK can be seen as a complement to the
Consortium, since the consortium is used by corporates to contribute
(with money) to ocaml, while the CDK is used by programmers to
contribute (with code) ...

Regards,

-- 
Fabrice
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Michel Mauny]

Dear all,

Let me first inform you that another french company is currently in
the "pipe" of becoming a member. I'll announce it more precisely when
the papers will be signed, but I think that we can reasonably consider
that we currently have 4 members in the Caml Consortium.

We need more, but we are on the right way :-)

About a non-profit organization such as an "Association loi 1901" as
proposed by Sven:

On Wed, Oct 24, 2001 at 07:05:44PM +0200, georges mariano wrote:

> Question : 
> Suppose that such idea will reach the implementation stage,
> how will be this Association considered by "THE" Consortium ? 

Sven replied (Oct 25 2001, 08:13AM +0200):

> As i understood, we will be a full member

I confirm that there would be absolutely no problem, and we would be
very happy to have such an organization as a member.

Regards,

-- Michel
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Michel Mauny]

Hendrik Tews wrote/écrivait (Oct 25 2001, 12:00PM +0200):
> Sven writes:
>    i was contemplating creating a association in france to group
>    several users which would, once we are able to sum up the 2000
>    euros, become one member of the consortium.
> 
> Are you sure that your application to become a member of the
> consortium will be successful? As I understand it, INRIA decides
> on the applications.

I confirm that we want such an application to be successful and I see no
problem at all.

> My impression from the current thread (and from previous ones on
> the consortium) is that there is some confusion about the real
> purpose of the consortium (i.e., INRIAS position)

Well, Xavier's post (http://caml.inria.fr/archives/200110/msg00272.html)
was rather clear, as well as the description of the Consortium given
as appendix of the membership agreement at

  http://caml.inria.fr/consortium/documents/Annex1-eng.shtml

Roughly speaking, the idea is to bring together major users (as many
as possible) in order to make other potential users more confident in
the future of OCaml. The existence of the Consortium should bring
(probably partial) answers to questions such as "Who else is using
it?", "What happens in case INRIA stops supporting it?". Hopefully, at
some point, some companies able to sell services around Caml will be
members of the CC, and that could also help in answering other
important questions such as "What company could maintain, adapt or
further develop this OCaml app that we are planning to use?"

The goal of further developing OCaml for it to match more precisely
the needs of programmers, and getting funds for that, is of course
also important, but I think the primary goal is the one given above.

We all understand that it's a chicken/egg problem, and this is why
it's starting a bit slowly (my opinion). (Well, honestly, one should
probably also add some work overload on my side as an another reason.)

And, again, forming an OCaml users group that would be a member of the
Consortium is definitely a good idea.

-- Michel
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Michel Mauny]

Joshua D. Guttman wrote/écrivait (Oct 24 2001, 01:35PM -0400):

> For people like me, it would be good to have a level of consortium
> membership that a single project could justify.  If there was
> something in the three figure range (i.e. hundreds but not thousands
> of dollars), I would simply make the arrangements and I'd never have
> to convince anyone organizationally very distant from me.  But for a
> consortium membership that costs thousands, I would have to convince
> people I rarely interact with, and they would want to decide whether
> OCaml should play some company-wide role, and probably it would be a
> dead end.
> 
> Would the OCaml Consortium consider something like this?  

This is one of the points that should be discussed at the first
meeting of the Consortium.

If we go that way, I'll have to argue with my hierarchy, because they
already didn't like that much the different contribution levels in the
current agreement, but it doesn't look impossible.

-- 
Michel
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Benjamin C. Pierce]

> However, not so many people need syncronization.

It's a little difficult to tell precisely, but we estimate the current
size of the Unison user community at between 1000 and 3000.  My sense is
that the latent demand is much larger -- we have grown to this size with
very little advertising of any kind.  I don't know if this qualifies as a
killer app for you, but it's clearly moved beyond the ivory tower at this
point.

(One thing that has surprised me is how many windows-only users we seem
to have.  Since there are several pretty good commercial synchronizers
for Windows, I'd have expected a smaller demand from this community.)

> > But even if you have a killer-app written in OCaml you will still have to explain 
> > to your manager (and even more your colleagues) why you would not have 
> > been able to write this app in C++ or Java (or why it would have been much
> > more effort to do it in another language). I think that because competing 
> > languages are advertised with buzzwords and their list of "advantages" 
> > OCaml should be advertised too. 

An early version of Unison was written in Java.  In many ways, this was
fine (Java is a pretty nice language, if you don't mind the fact that
your code gets about 10x longer than an equivalent ML version), but there
were several serious problems with Java:

[Disclaimer -- we switched to OCaml three years ago, and some of these
points have undoubtedly gotten somewhat better in the interim.]

   - poor run-time performance (huge memory footprints, poor GC, etc.,
     etc.) 

   - poor portability (despite all the "write-once-run-anywhere" hype, we 
     had neverending hassles with this -- different Swing bugs on
     different platforms, Linux implementation always a year behind
     Windows and Solaris, etc., etc.)

   - poor system-level APIs (in particular, no support for critical
     filesytem operations like fstat)

   - poor distribution support (RMI is a nice design, but the
     implementation was unbelievably heavy and expensive)
   
When we switched to OCaml, all these hassles vanished.  The one technical
downside was that going to OCaml involved doing a certain amount of work
ourselves that was done for us in Java.  In particular, we had to roll
our own RPC package.  However, once this was done, we had something that
was right for the job, and that we could understand and control.
Moreover, it was fun. :-)

There is also, of course, a social downside of using OCaml rather than
Java: the developer pool is *much* smaller.  However, I don't think this
has really made much of a difference for the success of the project --
although the code is GPLed, most of it is pretty intense, and I suspect
that there are not that many people even in the bigger Java community
that would be capable of picking it up and contributing to the effort
without significant help from us (e.g., spending a summer at Penn, or
whatever).  Conversely, switching to OCaml has made a huge difference for
the core development team, and this has contributed significantly to the
overall success of the project.

All things considered, switching to OCaml was a huge win for Unison.

    -- B
    
P.S...

> Unison could be a killer application, but it lacks all user friendliness.

I'd be interested to hear more about what you meant by this (perhaps
off-list).

-----------------------------------------------------------------------------
BENJAMIN C. PIERCE
Associate Prof., Computer & Information Science        bcpierce@cis.upenn.edu
University of Pennsylvania                                    +1 215 898-2012
200 South 33rd St.                                       Fax: +1 215 898-0587
Philadelphia, PA 19104, USA                http://www.cis.upenn.edu/~bcpierce
-----------------------------------------------------------------------------





    
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Sven]

On Thu, Oct 25, 2001 at 12:00:24PM +0200, Hendrik Tews wrote:
> Hi,
> 
> Sven writes:
>    Date: Wed, 24 Oct 2001 17:57:07 +0200
>    Subject: Re: [Caml-list] Whither the Caml Consortium?
>    
>    i was contemplating creating a association in france to group
>    several users which would, once we are able to sum up the 2000
>    euros, become one member of the consortium.
> 
> Are you sure that your application to become a member of the
> consortium will be successful? As I understand it, INRIA decides
> on the applications.

Yes, i have contacted them already, and they approve on it, as michel surely
did say in a message below. It seems they are only administratives reasons for
why they cannot start such a thing themselves.

> My impression from the current thread (and from previous ones on
> the consortium) is that there is some confusion about the real
> purpose of the consortium (i.e., INRIAS position) and what
> members of this list think about the consortium. Some members on
> the mailing list (myself included) would like to join the
> consortium to influence the development of Ocaml. But for reasons
> that I have not been able to grasp, INRIA seems to be not really
> keen on seeing us ``small'' ocaml users in the consortium.

It seems to be for administrative reasons, the processing fee and so on. The
association is the best way to solve this problem, and when i proposed it
here some time ago, michel told me it was a good idea, nobody else reacted
though, and it will not work if people are not interrested. Like said, i will
try to send a more formal mail about it this weekend, and i need at least 7
founding members to launch the thing. Right now, i have 3 offers, myself,
michel mauny, altough he cannot take a leading role from what i understood,
and georges mariano. That is not enough.

> I would therefore suggest that we first discuss what we want to
> achieve by becomming a consortium member. Then we can see how to
> make our interests compatible with INRIAS constraints. If we
> reach some consensus here, it is probably easier to get the
> programming work done ourselfs, instead of convincing the
> consortium to pay somebody to do the job.

mmm, ok yes, this is another way of putting it, but it may not be compatible
with the consortium or association thingy.

> I, for instance, would like some improvements for ocaml for which
> it is absolutely impossible to get scientific reward, and which
> are therefore very difficult to get implemented by the ocaml
> developers. Take for instance better error diagnostics from the
> ocaml parser or Thierry Bravier's ocamlyacc patch
> (http://caml.inria.fr/archives/199712/msg00020.html). One way to
> get these things done is to join the consortium ...

Yes and no, it may be possible to get them done already if you manage to
convince the ocaml team that these stuff are needed and well proven and good
code. The same will also be true once we join the consortium, since you will
not be the only member there. Sure, you could always fork the code, but this
would not be a good thing.

> Another possibility is to write a patch (like Thierry did) and
> get it into the ocaml distribution (what Thierry not achieved).

Do you know what were the reasons for it ?

> However, it is not clear (at least to me) what requirements have
> to be met, to get such an improvement accepted by the ocaml
> developers.
> I think, what is needed is that the ocaml developers give some
> guidelines on how we ``small'' ocaml users should proceed, if we
> want to contribute something to the ocaml kernel. Then there
> would be no need for us to join the consortium.

I will let them respond themselves here, ...

Friendly,

Sven Luther


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] user contributions to the core OCaml distribution?

[Xavier Leroy]

> I think, what is needed is that the ocaml developers give some
> guidelines on how we ``small'' ocaml users should proceed, if we
> want to contribute something to the ocaml kernel.

Here are some suggestions.

- Post your feature wish (or your patches, if you've already implemented
  the feature) to caml-bugs@inria.fr, not caml-list@inria.fr.
  Messages to caml-bugs are archived and tracked by our bug database,
  which has special categories for "feature wish" and "wish granted".
  In contrast, it's all too easy to overlook a message on caml-list,
  there are so many of them :-)

- Be explicit.  Use a descriptive "Subject" line, not just "feature wish".
  Explain why you need the feature, and what your patch actually does.

- If the patch is big (more than 100 lines), don't put it in the
  e-mail (the bug tracking system will truncate the message);
  put it on a Web or FTP server and include a URL.

- You can check the status of your (and others') wishes on the Web,
    http://caml.inria.fr/bin/caml-bugs/feature%20wish?user=guest

- Don't be disappointed if your wish or patch isn't incorporated in the
  OCaml working sources.  There might be several reasons for this:
  maybe we don't like the feature; maybe we like it but would
  implement it differently; maybe we have longer term plans to address
  the problem; maybe we have higher-priority stuff to deal with; etc.
  We (the core Caml development team) keep control on what gets in,
  Consortium or not.

In summary: we welcome users' input -- and a clear expression of a
need is as valuable to us as a patch -- but still follow a "cathedral"
model.

Hope this answers the question,

- Xavier Leroy
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Dmitry Bely]

Cuihtlauac ALVARADO <cuihtlauac.alvarado@rd.francetelecom.com> writes:

> > Regarding a killer-application in Ocaml to show that Ocaml is best, we have
> > to find a program that MANY people need. Just look how difficult it is to
> > create a good email program for Linux (Evolution is at last getting ready).
> > Such an application could be built using Ocaml instead. It would work, no
> > buffer overuns would reduce the number of security glitches. The only
> > problem is that almost all killer application has graphical user interfaces,
> > and I still haven't seen a good looking Ocaml-program (any pointers?).
> 
> Just a 30 seconds dream : a small, fast and robust web
> browser... Wouldn't you like to have a GC in Netscape ? Wouldn't you
> like to use powerfull tools able to program a defensive parsing for
> all the HTTP/Javascript garbage you can find out there ? Isn'it what
> *every* people need ?
> 
> I can dream no other definitive killer app for Ocaml.

A Web browser written in OCaml already exists, although its development
seems to be frozen:

http://pauillac.inria.fr/mmm/

Unfortunately, no Win32 version is there, so I do not know how small, fast
and robust it is (and compatible with current Internet standards, I would
add myself).

Hope to hear from you soon,
Dmitry


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

RE: [Caml-list] Whither the Caml Consortium?

[Francois Rouaix]

Well, I've stopped working on MMM back in 1997; since then, our friends
Jun and Pierre have essentially maintained the source compatible with
the latest OCaml distribution. But I wouldn't recommend it for general
Web browsing these days. It's only HTTP 1.0/HTML 3.2, with terrible
table support. That was fine in 1996, but not today.

--f
Francois Rouaix


-----Original Message-----
From: owner-caml-list@pauillac.inria.fr
[mailto:owner-caml-list@pauillac.inria.fr] On Behalf Of Dmitry Bely
Sent: Friday, October 26, 2001 7:17 AM
To: caml-list@inria.fr
Subject: Re: [Caml-list] Whither the Caml Consortium?

Cuihtlauac ALVARADO <cuihtlauac.alvarado@rd.francetelecom.com> writes:

> > Regarding a killer-application in Ocaml to show that Ocaml is best,
we have
> > to find a program that MANY people need. Just look how difficult it
is to
> > create a good email program for Linux (Evolution is at last getting
ready).
> > Such an application could be built using Ocaml instead. It would
work, no
> > buffer overuns would reduce the number of security glitches. The
only
> > problem is that almost all killer application has graphical user
interfaces,
> > and I still haven't seen a good looking Ocaml-program (any
pointers?).
> 
> Just a 30 seconds dream : a small, fast and robust web
> browser... Wouldn't you like to have a GC in Netscape ? Wouldn't you
> like to use powerfull tools able to program a defensive parsing for
> all the HTTP/Javascript garbage you can find out there ? Isn'it what
> *every* people need ?
> 
> I can dream no other definitive killer app for Ocaml.

A Web browser written in OCaml already exists, although its development
seems to be frozen:

http://pauillac.inria.fr/mmm/

Unfortunately, no Win32 version is there, so I do not know how small,
fast
and robust it is (and compatible with current Internet standards, I
would
add myself).

Hope to hear from you soon,
Dmitry


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ:
http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives:
http://caml.inria.fr
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] user contributions to the core OCaml distribution?

[Sven]

On Fri, Oct 26, 2001 at 02:13:47PM +0200, Xavier Leroy wrote:
> - Don't be disappointed if your wish or patch isn't incorporated in the
>   OCaml working sources.  There might be several reasons for this:
>   maybe we don't like the feature; maybe we like it but would
>   implement it differently; maybe we have longer term plans to address
>   the problem; maybe we have higher-priority stuff to deal with; etc.
>   We (the core Caml development team) keep control on what gets in,
>   Consortium or not.

mmm, this sound as no particular feedback or reason for rejecting a proposal
will be given.

Maybe, even a few line status or reason for rejection may be appended in the
BTS for this feature wish, or something such ?

Friendly,

Sven Luther
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Hendrik Tews]

Hi,


Fabrice Le Fessant writes:
   Date: Thu, 25 Oct 2001 17:45:03 +0200 (CEST)
   Subject: Re: [Caml-list] Whither the Caml Consortium?
   
   Maybe you don't know, but the CDK is distributed with a
   patched version of the ocaml compiler.

Indeed I was not aware of this. Where are these patches
documented? I only found cdk/sources/ocaml.patches/README, which
describes only 2 of the 17 patches.


Bye,

Hendrik
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Hendrik Tews]

Hi,

Sven writes:
   Date: Fri, 26 Oct 2001 11:56:13 +0200
   Subject: Re: [Caml-list] Whither the Caml Consortium?
   
     [about the ocamlyacc patch]   
   > Another possibility is to write a patch (like Thierry did) and
   > get it into the ocaml distribution (what Thierry not achieved).
   
   Do you know what were the reasons for it ?
   
No, maybe Xavier can answer this?

Bye,

Hendrik
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] Whither the Caml Consortium?

[Fabrice Le Fessant]

> Indeed I was not aware of this. Where are these patches
> documented? I only found cdk/sources/ocaml.patches/README, which
> describes only 2 of the 17 patches.

Most of the patches in the CDK are useful to improve the compilation
and distribution of the CDK (eg binary distribs independent of linux
distributions), and are not useful for other ocaml users. However,
some patches discussed on the mailing-list have been integrated and
can be used by users. For example:

14_yacc.patch is Thierry Bravier's ocamlyacc patch, to allow symbolic $
  variables in mly files.
13_scaml.patch is described on  http://algol.prosalg.no/~malc/scaml .
16_clink.patch is used to allow recursive modules in very simple
  cases (forward declaration of functions).
17_option.patch is a patch to allow optional values in modules 
  (from Alain Frisch).

As said by Xavier, these patches are not included in the standard
distribution for good reasons (either they need to be improved, or
better tested, or a better general solution is coming). They are
mainly in the CDK for beta-testing, and they will be kept until either
a big bug appears or a better replacement is found.

- Fabrice

-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr