Wikipedia

Wikipedia

[Jon Harrop]

Wikipedia is a large, famous, on-line, reader-editable encyclopaedia with a 
page dedicated to OCaml:

  http://en.wikipedia.org/wiki/Ocaml

The page gets a lot of hits and is, most likely, the first port of call by 
many people when trying to learn about OCaml.

Unfortunately, the quality of this page is substantially worse than that of 
the equivalent pages on SML, Haskell and so on. I have tried to improve the 
page myself but most of my links have been removed following complaints to 
admim by an anonymous, German-speaking, OCaml-using physicist with the IRC 
nic "tf" and all of my corrections were removed by Mike Lin. My code examples 
remain though.

So if anyone out there has a little spare time and wants to do something 
productive, please try to improve this page.

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
Objective CAML for Scientists
http://www.ffconsultancy.com/products/ocaml_for_scientists

Re: [Caml-list] Wikipedia

[Gerd Stolpmann]

Am Donnerstag, den 03.11.2005, 17:26 +0000 schrieb Jon Harrop:
> Wikipedia is a large, famous, on-line, reader-editable encyclopaedia with a 
> page dedicated to OCaml:
> 
>   http://en.wikipedia.org/wiki/Ocaml
> 
> The page gets a lot of hits and is, most likely, the first port of call by 
> many people when trying to learn about OCaml.
> 
> Unfortunately, the quality of this page is substantially worse than that of 
> the equivalent pages on SML, Haskell and so on. I have tried to improve the 
> page myself but most of my links have been removed following complaints to 
> admim by an anonymous, German-speaking, OCaml-using physicist with the IRC 
> nic "tf" and all of my corrections were removed by Mike Lin. My code examples 
> remain though.
> 
> So if anyone out there has a little spare time and wants to do something 
> productive, please try to improve this page.

The article focuses on the practical aspects of O'Caml, and this is IMHO
off-topic because it doesn't explain anything (e.g. look under
"Philosophy", you find some random thoughts about typing and speed). A
fanzine may do so, but not an encyclopedia.

In the first place, O'Caml is a research language where scientists
enhance ML in a sound way, and implement it such that it is practically
usable. The article mentions the "strict type system" several times, but
that remains mysterious and somehow ensures that there are no runtime
errors. Actually, there is a lot of theoretical work behind that. The
type system is actually a calculus that has some properties (e.g. there
is always a principal type if you don't use subtyping - which explains
something, in particular that the compiler can always deduce a type if
possible at all). At least, I would expect that such an article lists
the most important properties of type checking and evaluation (the
article does not even mention that O'Caml is an eager language).

If O'Caml has a revolutionary aspect, this can only be that the
developers do not allow that practical benefits destroy theoretical
soundness, and that O'Caml is nevertheless extremely well-suited for
practical problems at the same time. If that was the motto of the
article this would have my consent.

Gerd
-- 
------------------------------------------------------------
Gerd Stolpmann * Viktoriastr. 45 * 64293 Darmstadt * Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
Telefon: 06151/153855                  Telefax: 06151/997714
------------------------------------------------------------

Re: [Caml-list] Wikipedia

[Kip Macy]

[-- Attachment #1: Type: text/plain, Size: 1576 bytes --]

This isn't on-topic for the list, and may even qualify as a troll, but I
think your experiences point to another fundamental deficiency of the
wikipedia. Why do you think that others will have more success in their
updates than you have?


-Kip

On 11/3/05, Jon Harrop <jon@ffconsultancy.com> wrote:
>
>
> Wikipedia is a large, famous, on-line, reader-editable encyclopaedia with
> a
> page dedicated to OCaml:
>
> http://en.wikipedia.org/wiki/Ocaml
>
> The page gets a lot of hits and is, most likely, the first port of call by
> many people when trying to learn about OCaml.
>
> Unfortunately, the quality of this page is substantially worse than that
> of
> the equivalent pages on SML, Haskell and so on. I have tried to improve
> the
> page myself but most of my links have been removed following complaints to
> admim by an anonymous, German-speaking, OCaml-using physicist with the IRC
> nic "tf" and all of my corrections were removed by Mike Lin. My code
> examples
> remain though.
>
> So if anyone out there has a little spare time and wants to do something
> productive, please try to improve this page.
>
> --
> Dr Jon D Harrop, Flying Frog Consultancy Ltd.
> Objective CAML for Scientists
> http://www.ffconsultancy.com/products/ocaml_for_scientists
>
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>

[-- Attachment #2: Type: text/html, Size: 2281 bytes --]

Re: [Caml-list] Wikipedia

[Matt Gushee]

Kip Macy wrote:
> This isn't on-topic for the list, and may even qualify as a troll, but I
> think your experiences point to another fundamental deficiency of the
> wikipedia. Why do you think that others will have more success in their
> updates than you have?

Probably no single individual will have more success, but if several
people submit changes along similar lines, I would think the maintainers
will at least pay attention.

-- 
Matt Gushee
Englewood, CO, USA

Re: [Caml-list] Wikipedia

[Mike Lin]

It was my considered opinion that Jon's revisions reduced the overall
quality of the article, and therefore I reverted those parts I found
detrimental. Others may disagree; this is a natural part of the
anarchistic editorial process on Wikipedia and eventually you must
grow accustomed to having changes reverted without being offended. But
I do affirm and repeat that a community effort to improve the OCaml
article on Wikipedia would be wonderful. End of OT discussion? -Mike

On 11/3/05, Matt Gushee <matt@gushee.net> wrote:
> Kip Macy wrote:
> > This isn't on-topic for the list, and may even qualify as a troll, but I
> > think your experiences point to another fundamental deficiency of the
> > wikipedia. Why do you think that others will have more success in their
> > updates than you have?
>
> Probably no single individual will have more success, but if several
> people submit changes along similar lines, I would think the maintainers
> will at least pay attention.
>
> --
> Matt Gushee
> Englewood, CO, USA
>
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>

Re: [Caml-list] Wikipedia

[Florian Weimer]

* Jon Harrop:

> Unfortunately, the quality of this page is substantially worse than
> that of the equivalent pages on SML, Haskell and so on. I have tried
> to improve the page myself but most of my links have been removed
> following complaints to admim by an anonymous, German-speaking,
> OCaml-using physicist with the IRC nic "tf" and all of my
> corrections were removed by Mike Lin. My code examples remain
> though.

Mike has offered to reaccept your submission if it's more
encyclopedia-style.  However, I think your claim that you can do
completely without type annotations is a bit misleading.  You need
some (module) types to give structure to your programs, and some types
simply cannot be inferred (polymorphic record fields, for example).

Re: [Caml-list] Wikipedia

[skaller]

On Thu, 2005-11-03 at 20:24 +0100, Gerd Stolpmann wrote:

> If O'Caml has a revolutionary aspect, this can only be that the
> developers do not allow that practical benefits destroy theoretical
> soundness, and that O'Caml is nevertheless extremely well-suited for
> practical problems at the same time. If that was the motto of the
> article this would have my consent.

And if it had a historical importance, I would say it is the
first popular *functional* programming language to prove,
beyond doubt, that functional programming languages, particular
systems using garbage collectors, need not be slow.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

[Caml-list] what is high-level (was: Wikipedia)

[Blue Prawn]

Hello,

Perhaps it would be a good idea to create un page on Wikipedia to explain what 
is a high-level programming language, because its definition in all documents 
related to OCaml definitively do not fit the definition the penguins gave to 
me when I started to go in my lug, which was a language easy to learn and use 
designed for common users, and not only for programers and computer 
scientists. They told me about ABC which led to python which is comonly used 
in softwares as scripting extention for the users.
But OCaml do need some background knowledge to understand the official manual,
which is not true for PHP, Python or Ruby.

-- 
Cheers

Re: [Caml-list] what is high-level (was: Wikipedia)

[Brian Hurt]

On Fri, 4 Nov 2005, Blue Prawn wrote:

> Hello,
>
> Perhaps it would be a good idea to create un page on Wikipedia to 
> explain what is a high-level programming language,

As near as I can figure, the definition is simple.  Assembly language is 
not a high level language.  Everything else is.

http://www.webopedia.com/TERM/C/C.html

>From that link:
> Although it is a high-level language, C is much closer to assembly 
> language than are most other high-level languages.

In other words, all languages are high-level, some are just more 
high-level than others.


> because its definition in all documents related to OCaml definitively do 
> not fit the definition the penguins gave to me when I started to go in 
> my lug, which was a language easy to learn and use designed for common 
> users, and not only for programers and computer scientists. They told me 
> about ABC which led to python which is comonly used in softwares as 
> scripting extention for the users. But OCaml do need some background 
> knowledge to understand the official manual, which is not true for PHP, 
> Python or Ruby.

OK, here's the thing: Ocaml is a different paradigm than Python, Ruby, and 
PHP.  If you know Pascal, C, Fortran, etc., then learning PHP isn't 
difficult, because it too is a procedural language.  If you know C++, 
Java, etc., then learning Python or Ruby isn't that hard, because they're 
Object Oriented languages too.  If you already know SML or Haskell, 
learning Ocaml wouldn't be that hard.  The problem is that most people 
don't know SML or Haskell.

Learning a new paradigm is hard.  As someone who has done it three times 
now (moving from the sphagetti code of Basic to the procedural style of 
Pascal, then moving to Object Oriented, and most recently Functional), 
trust me on this.  Learning a new paradigm makes learning a new language 
10 times as hard AT LEAST as learning a new language in the old paradigm.

Brian

Re: [Caml-list] what is high-level (was: Wikipedia)

[David Teller]

Le vendredi 04 novembre 2005 à 09:13 -0600, Brian Hurt a écrit :

> >From that link:
> > Although it is a high-level language, C is much closer to assembly 
> > language than are most other high-level languages.
> 
> In other words, all languages are high-level, some are just more 
> high-level than others.

Fair enough on that. Still, we might need to define a notion of
higher-level language. Perhaps a language A is of higher-level than a
language B if the mode of thought imposed/encouraged by A are less
related to actual technical issues of language implementation and more
to issues of the target domain of your program ?

> Learning a new paradigm is hard.  As someone who has done it three times 
> now (moving from the sphagetti code of Basic to the procedural style of 
> Pascal, then moving to Object Oriented, and most recently Functional), 
> trust me on this.  Learning a new paradigm makes learning a new language 
> 10 times as hard AT LEAST as learning a new language in the old paradigm.

Same here, plus logical programming somewhere along the way.

Still, in OCaml/Haskell/ML, you do need some understanding of the type
system, which is typically not necessary in other programming languages.

> Brian

Cheers,
 David
-- 
Read, Write, and Publish Standard eBooks
  Free, Open Software, Open Standards and multi-platform
    The OpenBerg project http://www.openberg.org

Re: [Caml-list] what is high-level (was: Wikipedia)

[skaller]

On Fri, 2005-11-04 at 15:28 +0000, David Teller wrote:
> Le vendredi 04 novembre 2005 à 09:13 -0600, Brian Hurt a écrit :
> 
> > >From that link:
> > > Although it is a high-level language, C is much closer to assembly 
> > > language than are most other high-level languages.
> > 
> > In other words, all languages are high-level, some are just more 
> > high-level than others.
> 
> Fair enough on that. 

I think he was being sarcastic (reference to Animal Farm .. :)

> Still, we might need to define a notion of
> higher-level language.

I use the term 'advanced programming language' myself.

There is a list of features of such languages here:

http://www.felix.cybercloud.net/wiki/index.php/Advanced_programming

C, C++, Java, Php, etc clearly do not qualify :)

And quoting:

"Ocaml, on the other hand, supports all these features except 
for dynamic loading"

Note the page was written for C++ programmers.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Re: [Caml-list] what is high-level

[Alan Falloon]

David Teller wrote:

>Still, in OCaml/Haskell/ML, you do need some understanding of the type
>system, which is typically not necessary in other programming languages.
>  
>
To write software you need to have some notion of types. By that I mean 
that you need an understanding of what values a certain operation can 
produce.

In run-time type checked languages (and in untyped lagnuages),  its up 
to the programmer to keep the type system in thier head. This makes the 
type system fairly flexible and lets them use whatever concepts make 
sense to them, but its still a type system. This is useful for novices, 
but tends to scale poorly. It also means that even users who invent a 
sophisticaed type system for thier project are unlikely to share 
concepts with other advanced users.

In compile time typed languages, everyone has to agree on the type 
system. So, in modern languages, you get a push toward more 
sophisticated type systems than your average beginner can cope with, and 
with different concepts than an advanced user from dynamic languages 
would understand.

I think the biggest barrier is the language. It took me forever to 
figure out what a 'row variable' was. As far as I can tell its how you 
do "duck typing" (if it quacks like a duck and walks like a duck...) in 
a static language. Row variable is a good term when you are inventing 
the system or writing a paper on it, but when introducing it to users 
who already know Python saying "duck typing" will make it clear in seconds.

What we really need is a concept map from the popular languages (C, C++, 
Java, Python, Perl) to OCaml. Show common idioms in those languages and 
how they look in OCaml, and if there is a better way in OCaml then show 
that too. It might not make a good Wikipedia article, but it is the sort 
of project well suited to a Wiki. Is there an OCaml Wiki?

--
Alan Falloon

Re: [Caml-list] what is high-level

[William D. Neumann]

On Fri, 4 Nov 2005, Alan Falloon wrote:

> What we really need is a concept map from the popular languages (C, C++, 
> Java, Python, Perl) to OCaml. Show common idioms in those languages and how 
> they look in OCaml, and if there is a better way in OCaml then show that too. 
> It might not make a good Wikipedia article, but it is the sort of project 
> well suited to a Wiki. Is there an OCaml Wiki?

Yep.  <http://wiki.cocan.org/>

William D. Neumann

---

"There's just so many extra children, we could just feed the
children to these tigers.  We don't need them, we're not doing 
anything with them.

Tigers are noble and sleek; children are loud and messy."

         -- Neko Case

Life is unfair.  Kill yourself or get over it.
 	-- Black Box Recorder

Re: [Caml-list] what is high-level

[David Teller]

Le vendredi 04 novembre 2005 à 11:06 -0500, Alan Falloon a écrit : 
> David Teller wrote:
> 
> >Still, in OCaml/Haskell/ML, you do need some understanding of the type
> >system, which is typically not necessary in other programming languages.
> 
> To write software you need to have some notion of types. By that I mean 
> that you need an understanding of what values a certain operation can 
> produce.

Sure. But in Python/Boo, as you mention, you have duck typing. In my
books, that makes it easier to learn Python than OCaml, because that's
one less thing you need to know before starting your first program.

Of course, in OCaml, you have "static duck typing" for objects, but
that's a different issue.


> I think the biggest barrier is the language. It took me forever to 
> figure out what a 'row variable' was. 

Er... what is a row variable ? :)
Is that a polymorphic variant ?

> What we really need is a concept map from the popular languages (C, C++, 
> Java, Python, Perl) to OCaml. Show common idioms in those languages and 
> how they look in OCaml, and if there is a better way in OCaml then show 
> that too. It might not make a good Wikipedia article, but it is the sort 
> of project well suited to a Wiki. Is there an OCaml Wiki?

Iirc, there's an OCaml Wikibook on the Wikipedia.

Cheers,
 David
> 
-- 
Read, Write, and Publish Standard eBooks
  Free, Open Software, Open Standards and multi-platform
    The OpenBerg project http://www.openberg.org

Re: [Caml-list] what is high-level

[Matt Gushee]

Brian Hurt wrote:

> OK, here's the thing: Ocaml is a different paradigm than Python, Ruby,
> and PHP.  If you know Pascal, C, Fortran, etc., then learning PHP isn't
> difficult, because it too is a procedural language.  If you know C++,
> Java, etc., then learning Python or Ruby isn't that hard, because
> they're Object Oriented languages too.  If you already know SML or
> Haskell, learning Ocaml wouldn't be that hard.  The problem is that most
> people don't know SML or Haskell.

I used to believe that too, and have told people more than once: "It's
not inherently harder, it's just different from what you're used to."
But my opinion has changed. Actually, I would argue, making effective
use of functional techniques *is* harder because it requires more
abstract thinking. Procedural programming is full of "Do A to X, then do
B to Y ..."--a series of concrete operations performed on explicit
entities. There are no lambdas, no partial evaluation, no HOFs ... and
the fact that functions are always named and always defined in a
specific, identifiable piece of code is limiting, but also can make it
much easier to understand what a program is doing.

And I would argue that functional programming is a way of "working
smarter, not harder"--which of course requires knowing something. It may
be that the same is true of good OO programming, but as far as I can
tell few OO practitioners have a very deep understanding of the paradigm
(note that my impressions come mostly from exposure to run-of-the-mill
corporate Java programmers).

-- 
Matt Gushee
Englewood, CO, USA

Re: [Caml-list] Wikipedia

[Thomas Fischbacher]

On Thu, 3 Nov 2005, Jon Harrop wrote:

Yawn... the usual corrections:

> the equivalent pages on SML, Haskell and so on. I have tried to improve the 
> page myself but most of my links have been removed following complaints to 

"most of my links" = "most of the links I set to my own pages".
Wikipedia is not for self-promotion. At least, it *does* smell a bit 
strange if there are far more links to ffconsultancy on the Wikipeda 
page than to INRIA pages, don't you agree?

Especially as quite some people did repeatedly question the quality of 
your material.

> admim by an anonymous, German-speaking, OCaml-using physicist with the IRC 
> nic "tf" and all of my corrections were removed by Mike Lin. My code examples 
> remain though.

That person is called Thomas Fischbacher, and could not reply to your 
private email in time and in deail as he right now is on a conference and 
only has limited network access. More - maybe, but I hope not - in a few 
days.

-- 
regards,               tf@cip.physik.uni-muenchen.de              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

Re: [Caml-list] Wikipedia

[Alex Goldman]

On 11/4/05, Thomas Fischbacher
<Thomas.Fischbacher@physik.uni-muenchen.de> wrote:

> "most of my links" = "most of the links I set to my own pages".
> Wikipedia is not for self-promotion. At least, it *does* smell a bit
> strange if there are far more links to ffconsultancy on the Wikipeda
> page than to INRIA pages, don't you agree?

The title of his user page alone will leave you speechless: Dr. Jon D.
Harrop BA MSci PhD. I've never met anyone who used both Dr. and Ph.D.
at the same time, and that includes very accomplished people.

Wikipedia has a policy against self-promotion and link spam. The
purpose of Wikipedia is to disseminate generally useful and objective
encyclopedic knowledge. Instead, Harrop uses it to make
self-aggrandizing claims and increase the Google ranking of and
traffic to his "company" to sell his book.

Leave his spam all over Wikipedia, and soon anyone searching for
"OCaml" on Google will be directed to Harrop's unscientific BS.

Re: [Caml-list] Wikipedia

[Gerd Stolpmann]

Am Freitag, den 04.11.2005, 13:05 -0800 schrieb Alex Goldman:
> On 11/4/05, Thomas Fischbacher
> <Thomas.Fischbacher@physik.uni-muenchen.de> wrote:
> 
> > "most of my links" = "most of the links I set to my own pages".
> > Wikipedia is not for self-promotion. At least, it *does* smell a bit
> > strange if there are far more links to ffconsultancy on the Wikipeda
> > page than to INRIA pages, don't you agree?
> 
> The title of his user page alone will leave you speechless: Dr. Jon D.
> Harrop BA MSci PhD. I've never met anyone who used both Dr. and Ph.D.
> at the same time, and that includes very accomplished people.

Come on, this is not fair.

> Wikipedia has a policy against self-promotion and link spam. The
> purpose of Wikipedia is to disseminate generally useful and objective
> encyclopedic knowledge. Instead, Harrop uses it to make
> self-aggrandizing claims and increase the Google ranking of and
> traffic to his "company" to sell his book.

This is not fair, too. The links to his own pages can nevertheless be
objective. Actually, there are lots of scientific sites where very
well-known authors also sell their books (by including links to Amazon
etc.).

The point is that Wikipedia has no idea what quality is, and falls back
to a half-baked idea of "being objective = I must not advertise myself".
Very poor.

> Leave his spam all over Wikipedia, and soon anyone searching for
> "OCaml" on Google will be directed to Harrop's unscientific BS.

If his site has top-quality I don't have a problem with that.

Gerd
-- 
------------------------------------------------------------
Gerd Stolpmann * Viktoriastr. 45 * 64293 Darmstadt * Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
Telefon: 06151/153855                  Telefax: 06151/997714
------------------------------------------------------------

Re: [Caml-list] Wikipedia

[Alex Goldman]

On 11/4/05, Gerd Stolpmann <info@gerd-stolpmann.de> wrote:
> Am Freitag, den 04.11.2005, 13:05 -0800 schrieb Alex Goldman:
> > On 11/4/05, Thomas Fischbacher
> > <Thomas.Fischbacher@physik.uni-muenchen.de> wrote:
> >
> > > "most of my links" = "most of the links I set to my own pages".
> > > Wikipedia is not for self-promotion. At least, it *does* smell a bit
> > > strange if there are far more links to ffconsultancy on the Wikipeda
> > > page than to INRIA pages, don't you agree?
> >
> > The title of his user page alone will leave you speechless: Dr. Jon D.
> > Harrop BA MSci PhD. I've never met anyone who used both Dr. and Ph.D.
> > at the same time, and that includes very accomplished people.
>
> Come on, this is not fair.
>
> > Wikipedia has a policy against self-promotion and link spam. The
> > purpose of Wikipedia is to disseminate generally useful and objective
> > encyclopedic knowledge. Instead, Harrop uses it to make
> > self-aggrandizing claims and increase the Google ranking of and
> > traffic to his "company" to sell his book.
>
> This is not fair, too. The links to his own pages can nevertheless be
> objective. Actually, there are lots of scientific sites where very
> well-known authors also sell their books (by including links to Amazon
> etc.).

Do the authors of said scientific sites (Harrop's isn't one of them)
spam Wikipedia with links to their sites, while hiding behind
anonymous IPs (until they are discovered that is)?

The point is, if Harrop doesn't have megalomania, he sure acts like he
does. Look at some earlier versions of the Wikipedia article. He
spammed every section with a blatant link to his "ffconsultancy".

"Uses of Ocaml: 1. Something 2. Visualization {{link to
ffconsultancy}}. (No more uses) "
C'mon, Ocaml is a general purpose language.

He also diluted his spamming with "useful" info that incidentally
betrayed his less than stellar knowledge of the subject in the first
place.

The bad benchmark page is just an ad, really, a pretext to link-spam.

Look at his constant USENET spamming and trolling, if you don't see a
very old pattern here.

Re: [Caml-list] what is high-level

[skaller]

On Fri, 2005-11-04 at 11:06 -0500, Alan Falloon wrote:
>  typing" will make it clear in seconds.
> 
> What we really need is a concept map from the popular languages (C, C++, 
> Java, Python, Perl) to OCaml. 

Felix. It isn't a concept, but an ML like language which binds
to C++.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Re: [Caml-list] what is high-level

[Michael Walter]

It is also a rabbit, see e.g.
http://images.google.com/images?q=felix%20hase <0.5 wink>.

Michael

On 11/5/05, skaller <skaller@users.sourceforge.net> wrote:
> On Fri, 2005-11-04 at 11:06 -0500, Alan Falloon wrote:
> >  typing" will make it clear in seconds.
> >
> > What we really need is a concept map from the popular languages (C, C++,
> > Java, Python, Perl) to OCaml.
>
> Felix. It isn't a concept, but an ML like language which binds
> to C++.
>
> --
> John Skaller <skaller at users dot sf dot net>
> Felix, successor to C++: http://felix.sf.net
>
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>

Re: [Caml-list] what is high-level

[skaller]

On Sat, 2005-11-05 at 23:05 +0100, Michael Walter wrote:
[re Felix]

> It is also a rabbit, see e.g.
> http://images.google.com/images?q=felix%20hase <0.5 wink>.

Thanks for that link :)  .. but I'm afraid your German and/or
biology is a bit rusty! Felix Hase .. thats a *hare* not
a rabbit :)

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Re: [Caml-list] Wikipedia

[Thomas Fischbacher]

On Thu, 3 Nov 2005, Jon Harrop wrote:

...and once again, it may be interesting to see a bit more context.

>   http://en.wikipedia.org/wiki/Ocaml
> 
> The page gets a lot of hits and is, most likely, the first port of call by 
> many people when trying to learn about OCaml.
> 
> Unfortunately, the quality of this page is substantially worse than that of 
> the equivalent pages on SML, Haskell and so on. I have tried to improve the 
> page myself but most of my links have been removed following complaints to 
> admim by an anonymous, German-speaking, OCaml-using physicist with the IRC 
> nic "tf" and all of my corrections were removed by Mike Lin. My code examples 
> remain though.

On Oct 26, 19:10, I put a note into the "talk" section of the OCaml 
wikipedia article, pointing out that it contained an excessively large 
amount of links to ffconsultancy.com, which were indeed entered by Jon 
Harrop, by now infamous also on Usenet.

http://en.wikipedia.org/w/index.php?title=Talk:O%27Caml_programming_language&oldid=26550471

To give an excerpt from the article as it was at that time:

http://en.wikipedia.org/w/index.php?title=Ocaml&oldid=26478312

Please note the number of ffconsultancy.com links:

====>
<p>Fun:</p>
<ul>
<li>Several <a 
href="/wiki/International_Conference_on_Functional_Programming_Contest" 
title="International Conference on Functional Programming 
Contest">International Conference on Functional Programming Contest</a> 
winners</li>

<li><a href="http://www.ffconsultancy.com/free/maze/" class='external 
text' title="http://www.ffconsultancy.com/free/maze/">A 2D maze 
generator</a></li>
<li><a href="http://www.ffconsultancy.com/free/ray_tracer/" 
class='external text' title="http://www.ffconsultancy.com/free/ray 
tracer/">A mini ray tracer</a></li>
<li><a 
href="http://www.ffconsultancy.com/products/ocaml_for_scientists/visualisation/" 
class='external text' title="http://www.ffconsultancy.com/products/ocaml 
for scientists/visualisation/">Graphical examples from a book on OCaml for 
scientists</a></li>
<li><a href="http://handhelds.freshmeat.net/projects/planets/" 
class='external text' 
title="http://handhelds.freshmeat.net/projects/planets/">Gravity 
simulator</a></li>
</ul>
<p>Education:</p>
<ul>
<li><a href="http://home.gna.org/geocaml/" class='external text' 
title="http://home.gna.org/geocaml/">Drgeocaml</a>, a dynamic geometry 
software</li>
</ul>
<p>Scientific:</p>

<ul>
<li><a 
href="http://www.ffconsultancy.com/products/ocaml_for_scientists/complete/" 
class='external text' title="http://www.ffconsultancy.com/products/ocaml 
for scientists/complete/">Examples from a book on OCaml for 
scientists</a></li>
</ul>
<p>Commercial:</p>
<ul>
<li><a href="http://www.ffconsultancy.com/products/presenta" 
class='external text' 
title="http://www.ffconsultancy.com/products/presenta">Presenta</a> 
technical presentation software</li>
</ul>
<p>Engineering:</p>
<ul>
<li><a href="http://www.confluent.org" class='external text' 
title="http://www.confluent.org">Confluence</a> is a language for 
synchronous reactive system design. A Confluence program can generate 
digital logic for an FPGA or ASIC platform, or C code for hard real-time 
software.</li>

</ul>
<====

This was the state of the article when I put that comment into the 
discussion section. I did not re-enter that discussion from then on, so 
when Jon says:

> I have tried to improve the 
> page myself but most of my links have been removed following complaints to 
> admim by (...) tf

his "improvements" which were removed following my "complaints" can only 
refer to his act of Wikipedia vandalism (as I would call it), putting 
excessive link spam (see above) into the article.


Let's face it: if one takes Google rank as a a rough first measure of 
relevance (in the sense of "how much do people talk about it") in the 
non-academic world, in a search for "OCaml", none of Jon's material 
(despite all his efforts) is in the top 10 (note: google searches may 
depend on region/country), a link to his book is #17, and not one further 
single ffconsultancy link is among the top 50. 

So, if a single individual makes about half of all the external links in a 
Wikipedia page point to an irrelevant, obscure web site of his own(!), 
that hardly can be considered an "improvement", can it?

But unfortunately, such distorted view and presentation of reality seems 
to be not too uncommon in Jon's writings.


-- 
regards,               tf@cip.physik.uni-muenchen.de              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

RE: [Caml-list] Wikipedia

[Tony Edgin]

> -----Original Message-----
> From: caml-list-bounces@yquem.inria.fr [mailto:caml-list-
> bounces@yquem.inria.fr] On Behalf Of Thomas Fischbacher
> Sent: Monday, 7 November 2005 8:33 a.m.
> To: Jon Harrop
> Cc: caml-list@inria.fr
> Subject: Re: [Caml-list] Wikipedia
> 
> 
> On Thu, 3 Nov 2005, Jon Harrop wrote:
> 
> ...and once again, it may be interesting to see a bit more context.
> 
> >   http://en.wikipedia.org/wiki/Ocaml
> >
> > The page gets a lot of hits and is, most likely, the first port of call
> by
> > many people when trying to learn about OCaml.
> >
> > Unfortunately, the quality of this page is substantially worse than that
> of
> > the equivalent pages on SML, Haskell and so on. I have tried to improve
> the
> > page myself but most of my links have been removed following complaints
> to
> > admim by an anonymous, German-speaking, OCaml-using physicist with the
> IRC
> > nic "tf" and all of my corrections were removed by Mike Lin. My code
> examples
> > remain though.
> 
> On Oct 26, 19:10, I put a note into the "talk" section of the OCaml
> wikipedia article, pointing out that it contained an excessively large
> amount of links to ffconsultancy.com, which were indeed entered by Jon
> Harrop, by now infamous also on Usenet.
> 
> http://en.wikipedia.org/w/index.php?title=Talk:O%27Caml_programming_langua
> ge&oldid=26550471
> 
> To give an excerpt from the article as it was at that time:
> 
> http://en.wikipedia.org/w/index.php?title=Ocaml&oldid=26478312
> 
> Please note the number of ffconsultancy.com links:

I really don't see the issue here.  Yep.  He posted several examples hosted
on his own server and referenced his own book.  It would have been
significantly more work for him to dig up and test links to examples he
didn't produce.  If more people would contribute, the diversity of websites
reference would increase, and the density of FFConsultancy links would
decrease.  Maybe this should be a call for other contributors to aid his
wikipedia work instead of an attack on him attempting to improve the
wikipedia article.

Cheers,
Tony

RE: [Caml-list] Wikipedia

[Thomas Fischbacher]

On Mon, 7 Nov 2005, Tony Edgin wrote:

> I really don't see the issue here.  Yep.  He posted several examples hosted
> on his own server and referenced his own book.  It would have been
> significantly more work for him to dig up and test links to examples he
> didn't produce.  If more people would contribute, the diversity of websites
> reference would increase, and the density of FFConsultancy links would
> decrease. 

Actually, this would mean to turn this into a 100+ KB article on OCaml. 
The question arises whether this is appropriate for a brief overview in 
an encyclopedia.

Maybe it would be much more appropriate to reference just one page that 
shows example applications, which furthermore is supported and maintained 
by INRIA, such as http://caml.inria.fr/about/successes.en.html

-- 
regards,               tf@cip.physik.uni-muenchen.de              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

RE: [Caml-list] Wikipedia

[skaller]

On Mon, 2005-11-07 at 13:23 +0100, Thomas Fischbacher wrote:

> Actually, this would mean to turn this into a 100+ KB article on OCaml. 
> The question arises whether this is appropriate for a brief overview in 
> an encyclopedia.

I cannot answer that question, but the Wikipedia is littered
with silly articles on weird Java idioms and archaic C concepts.

For example the important concept of 'control inversion' 
is described as some kind of Java idiom. Like most idiomatic usage
of lame languages, it doesn't warrant a full scale article
to describe some workaround of a lame system -- instead
of describing a general concept -- in my opinion.

My point is -- more popular languages already pollute the
Wikipedia with considerable crap, so a well written
description of properties of Ocaml is warranted to
counter this rubbish -- 100KB if necessary!. IMHO. 

But not necessarily all in one place.

For example I would love if Jacques Garrigue would
spindle, fold, and mutilate some of his notes
on Polymorphic Variants, make a separate article
on that, and hang a link off the main Ocaml article.

In addition, there are surely entries on 'polymorphism'
and 'functional programming' and other such things which
have links to other articles which describe how various
languages provide those facilities .. and one can
envisage adding a link to an Ocaml specific article
on such topics.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

RE: [Caml-list] Wikipedia

[Thomas Fischbacher]

On Mon, 7 Nov 2005, skaller wrote:

> My point is -- more popular languages already pollute the
> Wikipedia with considerable crap, so a well written
> description of properties of Ocaml is warranted to
> counter this rubbish -- 100KB if necessary!. IMHO. 
> 
> But not necessarily all in one place.
> 
> For example I would love if Jacques Garrigue would
> spindle, fold, and mutilate some of his notes
> on Polymorphic Variants, make a separate article
> on that, and hang a link off the main Ocaml article.

Yes, I fully agree with you.

-- 
regards,               tf@cip.physik.uni-muenchen.de              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

Re: [Caml-list] Wikipedia

[skaller]

On Mon, 2005-11-07 at 11:14 +0100, Andreas Rossberg wrote:
> "skaller" <skaller@users.sourceforge.net>:
> >
> > In fact there is a lot of misleading or missing stuff in Wikipoedia,
> > and there are also some excellent articles. Try
> >
> > [...]
> > Referential Transparency (totally wrong)
> 
> That article could well be more accurate and complete, but I don't see at
> all how you arrive at judging it "totally wrong".

It confuses purity with transparency. Purity is a semantic property
of functions. Transparency is a property of expressions. 

Particularly, it is applications which are transparent NOT functions.

A function is pure if it depends only on its arguments.

A expression is transparent if its evaluation does not vary
with time or place (more or less).

As I pointed out in the commentary, 'functions' are necessarily
and trivially transparent, since ALL constant terms are transparent.

The article obscures an important theorem: given 'suitable conditions',
an expression is transparent if the function part of all applications
in the expressions are pure. 

In particular, if ALL functions are pure (Haskell or Clean), then
ALL expressions are transparent.

I might add -- I know about this because at the moment,
Felix has the rule that functions may not have side effects.
But still, not all expressions are transparent because functions
can depend on variables .. so it matters when you evaluate them.
This cannot happen in Haskell because it has no variables.

I don't know what to call this kind of 'function' other
than 'impure', nor what to call the resultant expressions, 
but I am calling them 'semi-transparent'. 

What is actually interesting is that applications
of impure functions  are "transparent within a 
particular space/time locus".

The Felix optimiser uses this. For example:

	var x = 1;
	fun f(y:int)=> x + y; // impure
	val a = f 1;          // a = 2 at this point
	val b = a + a;        // b = 4
	x = 2;                // modify x
	val c = b;            // should be 4 not 6!

In this code, the optimiser knows it can inline to obtain

	val b = f 1 + f 1;

however the substitution

	val c = f 1 + f 1;

will give the wrong answer, since x has been modified -- 
so the expression f 1 is "transparent" in a certain part
of the control flow graph. Felix can follow only linear
sequence of val bindings, it gives up on branches,
assignments, and procedure calls: it is using simple
pattern matching, not a proper data flow analysis.

Anyhow, I hope my objection is clear: it is a serious
category error. 'transparency' is a property of terms
of the language -- bits of syntax. Purity is a property
of functions, not the terms which denote the function.

The coupling is vital, in the sense that transparency
allows you to examine a call such as:

	f (g (y) )

and know for sure without looking at ANY other code that
IF the expressions and subexpressions are transparent,
THEN you can rewrite the expression like, for example:

	let a = g (y) in f (a)

(or conversely) and numerous other rewrite rules,
without changing the semantics. To establish this
transparency in Ocaml you know to examine the
functions f and g to see if they're pure,
and it remains only to examine 'a' to see if it is
transparent (assuming f,g are function constants).

And here .. f and g are *of course* transparent,
when considered as expressions themselves -- whether
or not they're pure: all constants are transparent.

Transparent and pure are distinct concepts applying
to distinct categories of entity. The article is
therefore *totally* wrong. Purity and transparency
are nominally independent, they have to be connected
by a (language dependent) theorem. If you mess up
the meaning of these important words, there would
be no way to state this theorem, let alone prove it.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Re: [Caml-list] Wikipedia

[Andreas Rossberg]

"skaller" <skaller@users.sourceforge.net>:
>
> In fact there is a lot of misleading or missing stuff in Wikipoedia,
> and there are also some excellent articles. Try
>
> [...]
> Referential Transparency (totally wrong)

That article could well be more accurate and complete, but I don't see at
all how you arrive at judging it "totally wrong".

  - Andreas

Re: [Caml-list] Wikipedia

[Thomas Fischbacher]

On Sun, 6 Nov 2005, Frederic GAVA wrote:

> A little thing that I do not understand. Why the wikipedia page is not 
> written by the ocaml's team ?

While this perhaps would be a good thing, as one should expect the INRIA 
people to be able to present their work in a reasonable and balanced way, 
I'd think it's more their duty to work on the compiler than to do 
merchandising on non-scientific channels. Especially as Wikipedia is, 
well, Wikipedia. I think they are busy enough doing that kind of work for 
research grants, say.

> and a link (just one) to the ffconsultancy.com is not 
> problem is there are many other links even it is a firm.

Yes, sure. I would not have minded a single link, say to Jon's book (even 
though it might be considered bad taste if he had placed it of his own), 
on that one Wikipedia page, but if things go so badly out of proportion, 
that's a completely different issue.

-- 
regards,               tf@cip.physik.uni-muenchen.de              (o_
 Thomas Fischbacher -  http://www.cip.physik.uni-muenchen.de/~tf  //\
(lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y)           V_/_
(if (= x 0) y (g g (- x 1) (* x y)))) n 1))                  (Debian GNU)

Re: [Caml-list] Wikipedia

[skaller]

On Sun, 2005-11-06 at 21:22 +0100, Frederic GAVA wrote:
> A little thing that I do not understand. Why the wikipedia page is not
> written by the ocaml's team ? 

Why should it be? Why haven't YOU contributed some text?
The idea of a Wiki is that anyone can contribute.
Wouldn't the Ocaml team be a bit biased?

In fact there is a lot of misleading or missing stuff in Wikipoedia,
and there are also some excellent articles. Try 

Control Inversion (misleading crap about a Java idiom)
Referential Transparency (totally wrong)
Curry-Howard Isomorphism (Superb Article!)

to mention 3 articles I've looked at recently.
(I added a talk article critiquing the transparency article
in the hope some expert would rewrite the article)

Actually, the Ocaml article is not so bad, it just leaves
out some things of importance -- as mentioned there is no
description of some of the key extensions over ML, particularly
polymorphic variants and labelled arguments, nor of key
differences eg mutable fields.

-- 
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Re: [Caml-list] Wikipedia

[Frederic GAVA]

[-- Attachment #1: Type: text/plain, Size: 5661 bytes --]

A little thing that I do not understand. Why the wikipedia page is not written by the ocaml's team ? An encyclopedia has to been written by "expert" and in the wiki idea with the help and the comments of the communauty... and a link (just one) to the ffconsultancy.com is not problem is there are many other links even it is a firm. But other links to other compagnies should also added? Why not in the rubric application, a sub-case "firm using ocaml" (that develop special applications with ocaml). And why, in this page, there are so little links to applications in ocaml. For example, the soft http://www.astree.ens.fr/ which is develop by academics people and using by a firm ?





> Message du 06/11/05 20:39
> De : "Thomas Fischbacher" 
> A : "Jon Harrop" 
> Copie à : caml-list@inria.fr
> Objet : Re: [Caml-list] Wikipedia
> 
> 
> On Thu, 3 Nov 2005, Jon Harrop wrote:
> 
> ...and once again, it may be interesting to see a bit more context.
> 
> > http://en.wikipedia.org/wiki/Ocaml
> > 
> > The page gets a lot of hits and is, most likely, the first port of call by 
> > many people when trying to learn about OCaml.
> > 
> > Unfortunately, the quality of this page is substantially worse than that of 
> > the equivalent pages on SML, Haskell and so on. I have tried to improve the 
> > page myself but most of my links have been removed following complaints to 
> > admim by an anonymous, German-speaking, OCaml-using physicist with the IRC 
> > nic "tf" and all of my corrections were removed by Mike Lin. My code examples 
> > remain though.
> 
> On Oct 26, 19:10, I put a note into the "talk" section of the OCaml 
> wikipedia article, pointing out that it contained an excessively large 
> amount of links to ffconsultancy.com, which were indeed entered by Jon 
> Harrop, by now infamous also on Usenet.
> 
> http://en.wikipedia.org/w/index.php?title=Talk:O%27Caml_programming_language&oldid=26550471
> 
> To give an excerpt from the article as it was at that time:
> 
> http://en.wikipedia.org/w/index.php?title=Ocaml&oldid=26478312
> 
> Please note the number of ffconsultancy.com links:
> 
> ====>
> 
> Fun:

> 

> 
Several > href="/wiki/International_Conference_on_Functional_Programming_Contest" 
> title="International Conference on Functional Programming 
> Contest">International Conference on Functional Programming Contest 
> winners

> 
> 
A 2D maze 
> generator

> 
> class='external text' title="http://www.ffconsultancy.com/free/ray 
> tracer/">A mini ray tracer

> 
> href="http://www.ffconsultancy.com/products/ocaml_for_scientists/visualisation/" 
> class='external text' title="http://www.ffconsultancy.com/products/ocaml 
> for scientists/visualisation/">Graphical examples from a book on OCaml for 
> scientists

> 
> class='external text' 
> title="http://handhelds.freshmeat.net/projects/planets/">Gravity 
> simulator

> 

> 
> Education:

> 

> 
> title="http://home.gna.org/geocaml/">Drgeocaml, a dynamic geometry 
> software

> 

> 
> Scientific:

> 
> 

> 
> href="http://www.ffconsultancy.com/products/ocaml_for_scientists/complete/" 
> class='external text' title="http://www.ffconsultancy.com/products/ocaml 
> for scientists/complete/">Examples from a book on OCaml for 
> scientists

> 

> 
> Commercial:

> 

> 
> class='external text' 
> title="http://www.ffconsultancy.com/products/presenta">Presenta 
> technical presentation software

> 

> 
> Engineering:

> 

> 
> title="http://www.confluent.org">Confluence is a language for 
> synchronous reactive system design. A Confluence program can generate 
> digital logic for an FPGA or ASIC platform, or C code for hard real-time 
> software.

> 
> 

> <====
> 
> This was the state of the article when I put that comment into the 
> discussion section. I did not re-enter that discussion from then on, so 
> when Jon says:
> 
> > I have tried to improve the 
> > page myself but most of my links have been removed following complaints to 
> > admim by (...) tf
> 
> his "improvements" which were removed following my "complaints" can only 
> refer to his act of Wikipedia vandalism (as I would call it), putting 
> excessive link spam (see above) into the article.
> 
> 
> Let's face it: if one takes Google rank as a a rough first measure of 
> relevance (in the sense of "how much do people talk about it") in the 
> non-academic world, in a search for "OCaml", none of Jon's material 
> (despite all his efforts) is in the top 10 (note: google searches may 
> depend on region/country), a link to his book is #17, and not one further 
> single ffconsultancy link is among the top 50. 
> 
> So, if a single individual makes about half of all the external links in a 
> Wikipedia page point to an irrelevant, obscure web site of his own(!), 
> that hardly can be considered an "improvement", can it?
> 
> But unfortunately, such distorted view and presentation of reality seems 
> to be not too uncommon in Jon's writings.
> 
> 
> -- 
> regards, tf@cip.physik.uni-muenchen.de (o_
> Thomas Fischbacher - http://www.cip.physik.uni-muenchen.de/~tf //\
> (lambda (n) ((lambda (p q r) (p p q r)) (lambda (g x y) V_/_
> (if (= x 0) y (g g (- x 1) (* x y)))) n 1)) (Debian GNU)
> 
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
> 
>

[-- Attachment #2: Type: text/html, Size: 7273 bytes --]