Smoke Vector Graphics: source code licenses for sale

Smoke Vector Graphics: source code licenses for sale

[Jon Harrop]

We are releasing our high-performance hardware-accelerated vector graphics 
software under a commercial source code license alongside the existing free 
edition:

  http://www.ffconsultancy.com/products/smoke_vector_graphics/?ol

Until the license is put on the site, e-mail me if you'd like to read a copy.

As an aside, several users asked us to provide bytecode versions for use with 
the latest OCaml 3.10 as well as 3.09 (still the latest in Debian testing). 
Our download statistics for last month show 94 hits for the 3.09 version and 
56 hits for the 3.10 version, so many people seem to be pushing ahead with 
the latest technology!

All the best,
-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Oliver Bandel]

Looks, like the Spam-Filter should be re-adjusted ;-)

Ciao,
   Oliver

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[skaller]

On Sun, 2007-11-04 at 16:06 +0100, Oliver Bandel wrote:
> Looks, like the Spam-Filter should be re-adjusted ;-)
> 
> Ciao,
>    Oliver

Why is that? This list is for Ocaml stuff. No one said
any particular licence. No one said free, pretend free,
or downright encumbered by a stupid licence.

In case you didn't realise .. programming is *primarily* a 
commercial (for money) activity which is *primarily* intended
to support commercial or governmental activity (including
academic and semi-academic institutions).

Lack of commercial support -- the kind Jon is offering --
is one of the impediments to industry taking Ocaml seriously.

It seems around 100 readers agree, if you can believe Jon's stats:

"Our download statistics for last month show 94 hits for the 3.09
version and 56 hits for the 3.10 version"

Lets hope Jon gets some sales .. he's one of the few commercial
support providers around.

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

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Oliver Bandel]

Zitat von skaller <skaller@users.sourceforge.net>:

>
> On Sun, 2007-11-04 at 16:06 +0100, Oliver Bandel wrote:
> > Looks, like the Spam-Filter should be re-adjusted ;-)
> >
> > Ciao,
> >    Oliver
>
> Why is that? This list is for Ocaml stuff. No one said
> any particular licence. No one said free, pretend free,
> or downright encumbered by a stupid licence.
>
> In case you didn't realise .. programming is *primarily* a
> commercial (for money) activity which is *primarily* intended
> to support commercial or governmental activity (including
> academic and semi-academic institutions).
[...]

if it is *primarily* this or that might be end in endless discussions.

But it also is a commercial thing.
I earn my money with programming too.

But I doubt this list is the right area to
post advertising-only posts.


>
> Lack of commercial support -- the kind Jon is offering --
> is one of the impediments to industry taking Ocaml seriously.
>
> It seems around 100 readers agree, if you can believe Jon's stats:
>
> "Our download statistics for last month show 94 hits for the 3.09
> version and 56 hits for the 3.10 version"
[...]

And what does theses hits show?
What do they prove?
Sorry, you mighht explain in more detail?!


>
> Lets hope Jon gets some sales .. he's one of the few commercial
> support providers around.

Yes, it's fine, if he makes money with OCaml.

But nevrtheless I doubt, this list is the forum to
make avertising-posts.

We could end up in "Buy this new Vi**gra-stuff; our website is OCaml-driven!".

Ciao,
   Oliver

Industry looks for commercial OCaml support?

[Oliver Bandel]

Zitat von skaller <skaller@users.sourceforge.net>:
[...]
> Lack of commercial support -- the kind Jon is offering --
> is one of the impediments to industry taking Ocaml seriously.
[...]

I doubt one or some programmers that offer their programming service is,
what industry is looking for, when it's looking for commercial support.

Commercial support means, that big companies offer that service, because
industry mistrusts single individuals.

So, I doubt Jon can provide this.

And: is this really a show-stopper for Ocaml, that there is no
"commercial support"?

I think, there are companies that already using Ocaml,
but do not mention it. They see the advantages and use it.

I don't miss any commercial support in OCaml, and
the industry, which not knew anything about OCaml
also does not miss support for a language that they don't know.


The need for such support may come,
when more people are interested in OCaml.


Ciao,
   Oliver

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[skaller]

On Sun, 2007-11-04 at 17:48 +0100, Oliver Bandel wrote:
> Zitat von skaller <skaller@users.sourceforge.net>:

> But I doubt this list is the right area to
> post advertising-only posts.

People do it ALL THE TIME. they announce new Ocaml products.
They even announce only marginally related conferences.
All of that is advertising.

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

RE: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[David Allsopp]

Oliver Bandel wrote:
> But I doubt this list is the right area to
> post advertising-only posts.

<quote src="http://caml.inria.fr/resources/forums.en.html">
The Caml Mailing List
... The purpose of this list is to share experience, exchange ideas and
code, and report on applications of the Caml language.

...

Caml Announcements 
This is a low-traffic, moderated list for announcements of Caml releases and
new Caml-related software, libraries, documents, etc. Please note that all
messages sent to caml-announce@inria.fr are also automatically forwarded to
caml-list@inria.fr.
</quote>

So perhaps Jon should post to caml-announce@inria.fr for moderation, but
it's safe to say from the above that O'Caml related software announcements
have a place on this list - indeed, they're reasonably common. Is your
problem just that Smoke costs money?

and...

> We could end up in "Buy this new Vi**gra-stuff; our website is OCaml-
> driven!".

... is just puerile. Can you pay attention to the results your post
http://tinyurl.com/2bzyre got and post something useful?

Re: [Caml-list] Industry looks for commercial OCaml support?

[skaller]

On Sun, 2007-11-04 at 17:58 +0100, Oliver Bandel wrote:
> Zitat von skaller <skaller@users.sourceforge.net>:
> [...]
> > Lack of commercial support -- the kind Jon is offering --
> > is one of the impediments to industry taking Ocaml seriously.
> [...]
> 
> I doubt one or some programmers that offer their programming service is,
> what industry is looking for, when it's looking for commercial support.
> 
> Commercial support means, that big companies offer that service, because
> industry mistrusts single individuals.

I believe you are right that big companies don't trust individuals,
but that doesn't mean Jon isn't offering anyhow.

> And: is this really a show-stopper for Ocaml, that there is no
> "commercial support"?

I think so, but I'm only guessing. Ubuntu Linux has commercial
support by Canonical, Fedora by Red-Hat, I believe this has
some impact on their popularity. C# is supported by MS,
Java by Sun.

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

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Vincent Hanquez]

On Mon, Nov 05, 2007 at 03:12:01AM +1100, skaller wrote:
> On Sun, 2007-11-04 at 16:06 +0100, Oliver Bandel wrote:
> > Looks, like the Spam-Filter should be re-adjusted ;-)
> > 
> > Ciao,
> >    Oliver
> 
> Why is that? This list is for Ocaml stuff. No one said
> any particular licence. No one said free, pretend free,
> or downright encumbered by a stupid licence.

There's advertising and too much advertising (spam).
Jon's posts are definitely in the latter.

-- 
Vincent Hanquez

RE: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Oliver Bandel]

Zitat von David Allsopp <dra-news@metastack.com>:

> Oliver Bandel wrote:
> > But I doubt this list is the right area to
> > post advertising-only posts.
>
> <quote src="http://caml.inria.fr/resources/forums.en.html">
> The Caml Mailing List
> ... The purpose of this list is to share experience, exchange ideas and
> code, and report on applications of the Caml language.

OK.


I thought the list is only for discussion on the language
and not for advertising things in a commercial manner.

As a catalogue of software I thought, the Caml-Hump is
the right area.

So, sorry, when I thought the focus of this list is narrower
than it officially is.

Since a while I have avoided to write messages with announcements here,
because I thought it's only for discussion, not for announcement
of software.


Ciao,
   Oliver

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Oliver Bandel]

Zitat von skaller <skaller@users.sourceforge.net>:

>
> On Sun, 2007-11-04 at 17:48 +0100, Oliver Bandel wrote:
> > Zitat von skaller <skaller@users.sourceforge.net>:
>
> > But I doubt this list is the right area to
> > post advertising-only posts.
>
> People do it ALL THE TIME. they announce new Ocaml products.

I thought that's, what the Caml-Hump is intended for
and the website on commercial caml-applications.


> They even announce only marginally related conferences.

This is discussion on the language, and I thought,
this is the main intention of this list.


> All of that is advertising.

Well, of different kind, IMHO.


But it seems I have seen the list in the wrong way.
Sorry.

Ciao,
   Oliver

Re: [Caml-list] Industry looks for commercial OCaml support?

[Oliver Bandel]

Zitat von skaller <skaller@users.sourceforge.net>:

>
> On Sun, 2007-11-04 at 17:58 +0100, Oliver Bandel wrote:
[...]

>
> > And: is this really a show-stopper for Ocaml, that there is no
> > "commercial support"?
>
> I think so, but I'm only guessing. Ubuntu Linux has commercial
> support by Canonical, Fedora by Red-Hat, I believe this has
> some impact on their popularity. C# is supported by MS,
> Java by Sun.
[...]

What's with C, C++, Perl?

Ciao,
   Oliver

Re: [Caml-list] Industry looks for commercial OCaml support?

[skaller]

On Sun, 2007-11-04 at 22:39 +0100, Oliver Bandel wrote:
> Zitat von skaller <skaller@users.sourceforge.net>:

> >
> > I think so, but I'm only guessing. Ubuntu Linux has commercial
> > support by Canonical, Fedora by Red-Hat, I believe this has
> > some impact on their popularity. C# is supported by MS,
> > Java by Sun.
> [...]
> 
> What's with C, C++, Perl?

C++ was developed by AT&T. HP pushed it into ANSI Standardisation
so they could us it in certain contracts. It went up to a joint
ANSI/ISO process later. There are many many commercial supporters
of C and C++ software.

Perl is dead... maybe *because* it lacked commercial support
as a language. 

There are certainly popular Open Source languages without
commercial support for the language development though: 
Python and Ruby for example.


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

OT: Commercial Support and Programming Languages

[Robert Fischer]

[-- Attachment #1: Type: text/html, Size: 3148 bytes --]

Re: [Caml-list] OT: Commercial Support and Programming Languages

[Jon Harrop]

On Monday 05 November 2007 03:58, Robert Fischer wrote:
> And O'Reilly had a big hand in evangelizing Perl, although it didn't
> offer "commercial support" in the form of consulting.

Indeed. Programming books are an often-forgotten but quite substantial 
commercial market. I'm still yearning for a book that covers ocamlbuild, 
IDEs, lablgtk, C FFI and so on...

Are there even any good tutorials on how to use Emacs/XEmacs as an IDE for 
OCaml, covering integrated top-level evaluation and so on?

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] Industry looks for commercial OCaml support?

[Jon Harrop]

On Monday 05 November 2007 03:32, skaller wrote:
> Perl is dead... maybe *because* it lacked commercial support
> as a language.

Do you mean Perl's use in industry has declined or that it has lost market 
share even among free software languages?

We had an interesting discussion about the choice of GL library that underpins 
LablGTKGL on the LablGTK mailing list a while back. According to the Debian 
package popularity contest, Perl (and OCaml) programmers use the 
arguably-deprecated GLArea library but other (primarily Ruby) programmers use 
the newer but arguably-worse GLExt library. I concluded that Perl's 
popularity is keeping GLArea alive so there isn't too much to worry about 
LablGTKGL growing out of date.

If Perl is in decline then maybe LablGTK should move on to GLExt. Looking at 
the Ubuntu package popularity contest now, GLExt is 5x more popular than 
GLArea (33k vs 6.4k installations).

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] OT: Commercial Support and Programming Languages

[skaller]

On Sun, 2007-11-04 at 21:58 -0600, Robert Fischer wrote:

> The announcement of Perl's death has been widely exaggerated: go ask
> your local Unix admin if Perl is dead.

Perl is dead in the sense of a developing language: sure, there
are still Perl scripts around, and people use it, and maybe
Perl 6 will actually happen ..

> Python is the interesting case, because it came out after Perl, but
> still managed to gain a fairly significant following.  It's still not
> terribly widely adopted and particularly not widely adopted in
> industry, so it doesn't manage to be a counterpoint to the basic
> argument.

That's not so clear to me: at least here in Australia Python has
a small but significant commercial toehold, certainly much larger
than Perl. However PHP is now bigger and Ruby with Rails is set
to wipe both out.

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

Re: OT: Commercial Support and Programming Languages

[Alan Falloon]

Jon Harrop wrote:
> On Monday 05 November 2007 03:58, Robert Fischer wrote:
>> And O'Reilly had a big hand in evangelizing Perl, although it didn't
>> offer "commercial support" in the form of consulting.
> 
> Indeed. Programming books are an often-forgotten but quite substantial 
> commercial market. I'm still yearning for a book that covers ocamlbuild, 
> IDEs, lablgtk, C FFI and so on...
> 
> Are there even any good tutorials on how to use Emacs/XEmacs as an IDE for 
> OCaml, covering integrated top-level evaluation and so on?

If you write one of those books, I will buy it.

Specifically, I want books covering: packaging ocaml for release, camlp4 
in detail, advanced ocamlbuild, and getting the most out of emacs and/or 
vim.

Maybe you could get Nicolas Pouillard to consult on camlp4 and ocamlbuild.

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Jon Harrop]

On Wednesday 07 November 2007 15:55, Alan Falloon wrote:
> If you write one of those books, I will buy it.
>
> Specifically, I want books covering: packaging ocaml for release, camlp4
> in detail, advanced ocamlbuild, and getting the most out of emacs and/or
> vim.
>
> Maybe you could get Nicolas Pouillard to consult on camlp4 and ocamlbuild.

This is a chicken and egg problem. Without tutorial information on these 
subjects I cannot learn about them in order to write a book on them.

Perhaps it would be best if everyone collaborated in writing wiki pages:

  http://www.ocaml-tutorial.org/introduction_to_gtk
  http://www.ocaml-tutorial.org/camlp4_3.10
  http://www.ocaml-tutorial.org/ocaml_and_the_web
  http://www.ocaml-tutorial.org/compiling_ocaml_projects

Another subject that I have discovered from my work on F# is the usefulness of 
interoperability, specifically being able to manipulate Excel spreadsheets 
from F#. I'm writing web analytics software for our company in OCaml and I'd 
like to inject the results into OpenOffice's spreadsheet. Any idea how to do 
that?

As an aside, although I am very tempted by the idea of writing a mainstream 
book on OCaml I am concerned that I might undercut our sales of OCaml for 
Scientists. This is particularly worrying because that book still accounts 
for 50% of our revenue from sales...

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Edgar Friendly]

Jon Harrop wrote:
> Another subject that I have discovered from my work on F# is the usefulness of 
> interoperability, specifically being able to manipulate Excel spreadsheets 
> from F#. I'm writing web analytics software for our company in OCaml and I'd 
> like to inject the results into OpenOffice's spreadsheet. Any idea how to do 
> that?
> 
Depending on the complexity of your results, you *might* succeed at
outputting a CSV file and having OO.org import that.  It probably won't
look like a final product, but might serve as a base for some OO macros
to clean up.

E.

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Jon Harrop]

On Wednesday 07 November 2007 16:35, you wrote:
> Depending on the complexity of your results, you *might* succeed at
> outputting a CSV file and having OO.org import that.  It probably won't
> look like a final product, but might serve as a base for some OO macros
> to clean up.

Can an OCaml program command OpenOffice to overwrite a current worksheet with 
the contents of a CSV file? I'm doing that by hand at the moment...

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Edgar Friendly]

Jon Harrop wrote:
> On Wednesday 07 November 2007 16:35, you wrote:
>> Depending on the complexity of your results, you *might* succeed at
>> outputting a CSV file and having OO.org import that.  It probably won't
>> look like a final product, but might serve as a base for some OO macros
>> to clean up.
> 
> Can an OCaml program command OpenOffice to overwrite a current worksheet with 
> the contents of a CSV file? I'm doing that by hand at the moment...
> 
OO.org has UNO, which is available through Java, OO Basic, MS OLE and
the CLI.  There's supposedly Python bindings too.  To do what you want,
I expect you'll need some bindings to UNO, which don't exist ATM, but
CamlIDL might get you to OLE easy enough (never used it, tho), or some
of the new caml-java might get you through the java route.  A weak
binding might not take that much work to write.

E.

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Christophe TROESTLER]

On Wed, 7 Nov 2007 16:41:24 +0000, Jon Harrop wrote:
> 
> On Wednesday 07 November 2007 16:35, you wrote:
> > Depending on the complexity of your results, you *might* succeed at
> > outputting a CSV file and having OO.org import that.  It probably won't
> > look like a final product, but might serve as a base for some OO macros
> > to clean up.
> 
> Can an OCaml program command OpenOffice to overwrite a current worksheet with 
> the contents of a CSV file? I'm doing that by hand at the moment...

This may interest you as a starting point:

  http://www.linuxjournal.com/article/8608

My 2 cents,
ChriS

Re: [Caml-list] Re: OT: Commercial Support and Programming Languages

[Richard Jones]

On Wed, Nov 07, 2007 at 04:07:51PM +0000, Jon Harrop wrote:
> Another subject that I have discovered from my work on F# is the usefulness of 
> interoperability, specifically being able to manipulate Excel spreadsheets 
> from F#. I'm writing web analytics software for our company in OCaml and I'd 
> like to inject the results into OpenOffice's spreadsheet. Any idea how to do 
> that?

It's funny you should mention these, because all three are possible
under OCaml (and I've done two and a half of them).

You can manipulate Excel using http://tech.motion-twin.com/ocamole.html
(only OCaml on Windows, mind you).

You can write Excel spreadsheets via a number of routes.  Probably
easiest is to use the CSV library from http://merjis.com/developers/csv
to write CSV files, which is exactly what we did for web log analysis
(http://merjis.com/developers/weblogs).  Another alternative is to go
via perl4caml & use the Perl module which does this.

Writing OpenOffice spreadsheets is easier because you can write ODF
directly, and put it into a ZIP file in the right format using
something like CamlZip.  I did this for OO writer documents, and have
some code if you want to take a look at it.

Rich.

-- 
Richard Jones
Red Hat

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Alexandre Pilkiewicz]

Le Sunday 04 November 2007 17:12:01 skaller, vous avez écrit :
> Lack of commercial support -- the kind Jon is offering --
> is one of the impediments to industry taking Ocaml seriously.


Don't you think the worst problem for the industry is the lack of 
retro-compatibility ? 

Between 3.09 and 3.10 (a *minor* version number change), a lot of program 
using camlp4 stopped compiling. If a company has a 100 000 lines code (or 
more) to revise just for that, it could be kind of a problem.

And so many things are just unspecified. I know it's a bad idea, whatever 
language you use, to rely on the order of evaluation of the argument of a 
function, but to say "this order may change one day" is to tell 
industrialists : "if you have some "not so good" programmers, even if you 
make all the test you want on your program to check it works, one day it may 
just stop working because we changed the order or evaluation, or worst, a lot 
of silent bugs can appear". 
Do you imagine that a car constructor with nice screw-down-robots would accept 
if a provider says "I have very nice and efficient screws, but it is possible 
that one day, you have to turn them anticlockwise if I realize it's more 
efficient." ?

IMO it is the real show-stopper for an industrial use of OCaml


Alexandre Pilkiewicz

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Luc Maranget]

> And so many things are just unspecified. I know it's a bad idea, whatever 
> language you use, to rely on the order of evaluation of the argument of a 
> function, but to say "this order may change one day" is to tell 
> industrialists : "if you have some "not so good" programmers, even if you 
> make all the test you want on your program to check it works, one day it may 
> just stop working because we changed the order or evaluation, or worst, a lot 
> of silent bugs can appear". 

I cannot resist to remind you that the order of argument evaluation is
unspecified in C.

Arguably a bad design, but not sufficient to frighten industry.


-- Luc

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Brian Hurt]

[-- Attachment #1: Type: text/plain, Size: 1726 bytes --]

Alexandre Pilkiewicz wrote:

>Le Sunday 04 November 2007 17:12:01 skaller, vous avez écrit :
>  
>
>>Lack of commercial support -- the kind Jon is offering --
>>is one of the impediments to industry taking Ocaml seriously.
>>    
>>
>
>
>Don't you think the worst problem for the industry is the lack of 
>retro-compatibility ? 
>
>Between 3.09 and 3.10 (a *minor* version number change), a lot of program 
>using camlp4 stopped compiling. If a company has a 100 000 lines code (or 
>more) to revise just for that, it could be kind of a problem.
>
>And so many things are just unspecified. I know it's a bad idea, whatever 
>language you use, to rely on the order of evaluation of the argument of a 
>function, but to say "this order may change one day" is to tell 
>industrialists : "if you have some "not so good" programmers, even if you 
>make all the test you want on your program to check it works, one day it may 
>just stop working because we changed the order or evaluation, or worst, a lot 
>of silent bugs can appear". 
>  
>
Which explains why industry is so wary of languages like Java and C++, 
and refuses to adopt them.

Oh, wait.

Seriously, you can protect against a "not so good" programmer.  Java 
tried, they really tried.  But you can't have decent performance and 
absolute reproducability- as Java discovered with floating point numbers 
and memory behavior in a multi-threaded program, among other problems.  
Witness all the rants about Java's "Write once, debug everywhere" problems.

One thing I really like about Ocaml is that if order of evaluation is 
important, it gives me a way to enforce a specific ordering.  And it's 
not even an obscure part of the language (let/in definitions).

Brian


[-- Attachment #2: Type: text/html, Size: 2220 bytes --]

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Alexandre Pilkiewicz]

Le Thursday 08 November 2007 20:09:03 Luc Maranget, vous avez écrit :
> > And so many things are just unspecified. I know it's a bad idea, whatever
> > language you use, to rely on the order of evaluation of the argument of a
> > function, but to say "this order may change one day" is to tell
> > industrialists : "if you have some "not so good" programmers, even if you
> > make all the test you want on your program to check it works, one day it
> > may just stop working because we changed the order or evaluation, or
> > worst, a lot of silent bugs can appear".
>
> I cannot resist to remind you that the order of argument evaluation is
> unspecified in C.
>
> Arguably a bad design, but not sufficient to frighten industry.
>
>
> -- Luc


My mistake. This was definitely a bad example. I never realized that the same 
problem existed in other languages, probably because we use so much more 
function evaluations in OCaml (and also probably because I program almost 
only in OCaml :-) ). Sorry for the noise :-\

Do you think it may be possible one day for OCaml to have a normalisation, 
like Haskell had in 1998, and if it's allready the case, where can I find 
it ?


-- 
Alexandre Pilkiewicz

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Edgar Friendly]

Alexandre Pilkiewicz wrote:
> Do you think it may be possible one day for OCaml to have a normalisation, 
> like Haskell had in 1998, and if it's allready the case, where can I find 
> it ?
> 
I don't see the need for formalizing the OCaml language - If we tried to
characterize what the current compiler did, we'd fail in many many
details.  And if we tried to write a spec independent of the compiler,
all of a sudden we'd just introduce hundreds of bugs into the compiler
because of its deviation from the spec.

Formal specifications have use in strictly "research" languages, but I
see their ability to stifle growth and improvement as more negative than
their ability to help people understand the proper operation of OCaml.

E.

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Oliver Bandel]

Zitat von Edgar Friendly <thelema314@gmail.com>:

> Alexandre Pilkiewicz wrote:
> > Do you think it may be possible one day for OCaml to have a normalisation,
> > like Haskell had in 1998, and if it's allready the case, where can I find
> > it ?
> >
> I don't see the need for formalizing the OCaml language - If we tried to
> characterize what the current compiler did, we'd fail in many many
> details.  And if we tried to write a spec independent of the compiler,
> all of a sudden we'd just introduce hundreds of bugs into the compiler
> because of its deviation from the spec.
>
> Formal specifications have use in strictly "research" languages, but I
> see their ability to stifle growth and improvement as more negative than
> their ability to help people understand the proper operation of OCaml.
>
[...]

It seems there are so many wishes to OCaml,
which are similar to wishes that long-time Word-users
that tried LaTeX the first time had for LaTeX.

Things like "how to change the layout freely, as I can do with Word?"
and they are looking for so many things of freedom that result
in ugly layouts, ugly typography, ...

The don't know about microtypography, kerning, grey-value of the
page, the number of characters per line that are easily readable,
the peace for the eye (and mind), when there are not too many fonts and colors,
and the many other typographic details, that are built-in in TeX/LaTeX.

And: no, TeX/LaTeX is/are not perfect. But they are always a good choice. :)

And the discussions on this list often remind me on those discussions about
LaTeX ;-)

I don't want to say, feature-wishes are nonsense... I only saw a
similarity in the kind of feature-wishes. :)


Ciao,
   Oliver


P.S.: This mail is not intended to offend someone; only a note
      to - maybe - contemplate about, when you have some minutes
      of free time ;-)
      (and sorry, if I'm too much off-topic ;-))

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Richard Jones]

On Thu, Nov 08, 2007 at 07:28:11PM +0100, Alexandre Pilkiewicz wrote:
> Don't you think the worst problem for the industry is the lack of 
> retro-compatibility ? 
> 
> Between 3.09 and 3.10 (a *minor* version number change), a lot of program 
> using camlp4 stopped compiling. If a company has a 100 000 lines code (or 
> more) to revise just for that, it could be kind of a problem.

Granted, it wasn't OCaml's finest moment, but both Debian & Fedora/
RHEL are shipping the 'old' camlp4 (now called camlp5) along with the
new one.  So commercial enterprises depending on camlp4 from 3.09, if
there are any, shouldn't have problems.

Rich.

-- 
Richard Jones
Red Hat

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Gerd Stolpmann]

Am Donnerstag, den 08.11.2007, 23:08 +0000 schrieb Richard Jones:
> On Thu, Nov 08, 2007 at 07:28:11PM +0100, Alexandre Pilkiewicz wrote:
> > Don't you think the worst problem for the industry is the lack of 
> > retro-compatibility ? 
> > 
> > Between 3.09 and 3.10 (a *minor* version number change), a lot of program 
> > using camlp4 stopped compiling. If a company has a 100 000 lines code (or 
> > more) to revise just for that, it could be kind of a problem.
> 
> Granted, it wasn't OCaml's finest moment, but both Debian & Fedora/
> RHEL are shipping the 'old' camlp4 (now called camlp5) along with the
> new one.  So commercial enterprises depending on camlp4 from 3.09, if
> there are any, shouldn't have problems.

There is another solution: continue to use 3.09 (that's what we are
doing at Wink) and wait until the problems in the external libraries are
solved. There's no real disadvantage (for us) - 3.09 is one of the best
versions that ever existed.

If the new camlp4 was really a problem for enterprises they could also
invest time in upgrading the code. I don't see any problem here since
they got the code for free. I've already upgraded stuff to 3.10 - it's
not really that much time, very affordable for companies. Actually, this
is the type of development that tends to be boring, and companies would
be very welcome improving the motivation with money.

Before this thread continues in this style: There is nothing wrong with
O'Caml in industry - all languages have their good and bad sides, and
industry knows this. We are currently seeing a very slow adoption by
industrial pioneers. All in all this is encouraging. For speeding this
process up, the most urgent things we need are people who convince their
bosses to try this "new" technology out. Where are all the companies
that jump at every trend, and that want to present themselves as the
earliest adopters of any stuff that is ever released to mankind? - I
think it is the right moment to advertise O'Caml in this way. Make a
trend of it. A trend does not need real arguments (this can come later),
it lives for some time by itself because people don't want to miss it.

I think we should stop throwing at every O'Caml error with the
industrial argument ("this and this prevents O'Caml from being used in
industry"). We wouldn't have pioneering companies if that were true in
any way. And saying this argument too loud could really prevent possible
uses in industry - evaluations are often not based on facts, but on
reputation.

Gerd
-- 
------------------------------------------------------------
Gerd Stolpmann * Viktoriastr. 45 * 64293 Darmstadt * Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
Phone: +49-6151-153855                  Fax: +49-6151-997714
------------------------------------------------------------

Re: [Caml-list] Smoke Vector Graphics: source code licenses for sale

[Richard Jones]

On Fri, Nov 09, 2007 at 01:16:28PM +0100, Gerd Stolpmann wrote:
> Before this thread continues in this style: There is nothing wrong with
> O'Caml in industry - all languages have their good and bad sides, and
> industry knows this. We are currently seeing a very slow adoption by
> industrial pioneers. All in all this is encouraging. For speeding this
> process up, the most urgent things we need are people who convince their
> bosses to try this "new" technology out. Where are all the companies
> that jump at every trend, and that want to present themselves as the
> earliest adopters of any stuff that is ever released to mankind? - I
> think it is the right moment to advertise O'Caml in this way. Make a
> trend of it. A trend does not need real arguments (this can come later),
> it lives for some time by itself because people don't want to miss it.

A small anecdote from a friend of mine who works at $BIG_BANK is that
the Jane Street paper[1] has been doing the rounds there and is
proving to be quite popular / influential.

Rich.

[1] From page 23 here
http://www.haskell.org/sitewiki/images/0/03/TMR-Issue7.pdf, there is
also a standalone document which I can't find now.

-- 
Richard Jones
Red Hat