Extra Widgets

0 views
Skip to first unread message

Adrian Davis

unread,
Apr 6, 1999, 3:00:00 AM4/6/99
to
A few weeks back I seem to remember somebody announced a set of addtional
widgets, I think the chap had a French name?? Can anybody point me to his
site??

Many Thanks,
=Adrian=

Cameron Laird

unread,
Apr 6, 1999, 3:00:00 AM4/6/99
to
In article <01be8025$0a1c3a80$8c03010a@mv292>,

Adrian Davis <Adrian...@lewisham.gov.uk> wrote:
>A few weeks back I seem to remember somebody announced a set of addtional
>widgets, I think the chap had a French name?? Can anybody point me to his
>site??
.
.
.
BWidget <URL:http://www.unifix-online.com/BWidget/>. I'm in
a dark mood about BWidget now because I don't understand its
licensing <URL:http://x3.dejanews.com/getdoc.xp?AN=461342750>.

Incidentally, do you know how to use DejaNews <URL:
http://www.dejanews.com/home_ps.shtml> to do your own search-
ing?
--

Cameron Laird http://starbase.neosoft.com/~claird/home.html
cla...@NeoSoft.com +1 281 996 8546 FAX

Daniel

unread,
Apr 9, 1999, 3:00:00 AM4/9/99
to

> BWidget <URL:http://www.unifix-online.com/BWidget/>. I'm in
> a dark mood about BWidget now because I don't understand its
> licensing <URL:http://x3.dejanews.com/getdoc.xp?AN=461342750>.

Hola Cameron,

It is LGPL. You can use for whatever purposes (including commercial ones) but
changes to the library itself must be freely redistributable

From hte FAQ:

1.The BWidget Toolkit is it really a free package ?

Yes the BWidget package is free software under the
terms of the GNU Library General Public License. See the license file
(license.txt)
comming with the package or contact us for
informations.


Greetings

Daniel

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own

Cameron Laird

unread,
Apr 9, 1999, 3:00:00 AM4/9/99
to
In article <7ekblt$cg$1...@nnrp1.dejanews.com>,

Daniel <ridr...@esi.us.es> wrote:
>
>
>> BWidget <URL:http://www.unifix-online.com/BWidget/>. I'm in
>> a dark mood about BWidget now because I don't understand its
>> licensing <URL:http://x3.dejanews.com/getdoc.xp?AN=461342750>.
>
>Hola Cameron,
>
>It is LGPL. You can use for whatever purposes (including commercial ones) but
>changes to the library itself must be freely redistributable
>
>From hte FAQ:
>
>1.The BWidget Toolkit is it really a free package ?
>
> Yes the BWidget package is free software under the
>terms of the GNU Library General Public License. See the license file
>(license.txt)
> comming with the package or contact us for
>informations.
.
.
.
Thanks, Daniel.

I'm not satisfied. I appreciate your efforts, and Mr.
Chassany's, to clear this up for me. I remain uneasy.

Mr. Chassany has told everyone whom I know to have writ-
ten him that he wants to give the community a useful
toolkit, and that the only restriction he intends to
apply is that commercial applications include the
license.txt <URL:http://www.unifix-online.com/BWidget/license.html>.
This is more than reasonable. I can only applaud his
motives and professionalism.

I know he chose LGPL <URL:http://www.gnu.org/copyleft/lgpl.txt>
(which he sometimes writes as GPL; that's the sort of
thing that worries me, just because I've seen how much
can unravel when lawyers arrive on the scene) precisely
so as not "to create problems about BWidget use". However,
(L)GPL troubles me:
1. Is there precedent for applying LGPL to a
corpus of source-code scripts?
2. How does a LGPLed BWidget affect the rest
of any commercial applications I deliver?
3. What does it mean that I "must provide
complete object files to the recipients?"
3. What's an example of something LGPL does
NOT allow me to do?

I hope this is all just a misunderstanding on my part.
Maybe the reality is as simple as Mr. Chassany quite
evidently wants it to be. I surely am no expert in
LGPL interpretation. Is there someone who can help me
with this?

Incidentally, I'm certain that composition of the LGPL
was a good thing, for it applies to many common situations
better than GPL. Is anyone thinking about whether we now
need also a ScriptingGPL?

Steven G. Johnson

unread,
Apr 10, 1999, 3:00:00 AM4/10/99
to
cla...@Starbase.NeoSoft.COM (Cameron Laird) wrote:
> [...] However, (L)GPL troubles me:
> [...]

> 2. How does a LGPLed BWidget affect the rest
> of any commercial applications I deliver?

It doesn't, except as in (3) below.

> 3. What does it mean that I "must provide
> complete object files to the recipients?"

It means that you should provide the program in a form such that users can
swap in a newer version of BWidget if they choose. This is normally
satisfied by linking to the LGPLed code as a shared library (DLL), since
users can replace/update shared libraries with no problem. (It sounds
like BWidget is just a script that you call; in this case, I imagine that
the script is easily replaceable by nature.)

> 4. What's an example of something LGPL does


> NOT allow me to do?

You can't modify the BWidget source code and then distribute the modified
version only in binary (or otherwise obfuscated) form.

You can't forbid your users from redistributing the BWidget library, or
refuse to give them the BWidget source code (or modified versions
thereof).

You can't statically link to (or otherwise use/call) BWidget in such a way
as to prevent users from swapping in a newer/modified version of BWidget.

Cordially,
Steven G. Johnson

Cameron Laird

unread,
Apr 11, 1999, 3:00:00 AM4/11/99
to
In article <stevenj-1004...@wetelectron.mit.edu>,

Steven G. Johnson <ste...@alum.mit.edu> wrote:
>cla...@Starbase.NeoSoft.COM (Cameron Laird) wrote:
>> [...] However, (L)GPL troubles me:
>> [...]
>> 2. How does a LGPLed BWidget affect the rest
>> of any commercial applications I deliver?
>
>It doesn't, except as in (3) below.
>
>> 3. What does it mean that I "must provide
>> complete object files to the recipients?"
>
>It means that you should provide the program in a form such that users can
>swap in a newer version of BWidget if they choose. This is normally
>satisfied by linking to the LGPLed code as a shared library (DLL), since
>users can replace/update shared libraries with no problem. (It sounds
>like BWidget is just a script that you call; in this case, I imagine that
>the script is easily replaceable by nature.)
Yes.

That is, BWidget is "just a script" formally, but
one of considerable craftsmanship and artistry.


>
>> 4. What's an example of something LGPL does
>> NOT allow me to do?
>
>You can't modify the BWidget source code and then distribute the modified
>version only in binary (or otherwise obfuscated) form.
>
>You can't forbid your users from redistributing the BWidget library, or
>refuse to give them the BWidget source code (or modified versions
>thereof).
>
>You can't statically link to (or otherwise use/call) BWidget in such a way
>as to prevent users from swapping in a newer/modified version of BWidget.

.
.
.
I find this interpretation thoroughly consistent with
everything I know of Mr. Chassany's intentions, entire-
ly satisfying to me as a commercial developer, and well
grounded in what I know of GPL's historical development.
I've provisionally decided that I'll stop looking for
trouble in this area, and trust that the ambiguities I
perceive in application of LGPL to scripting source are
only "theoretical"--no reasonable person would read it
in any way other than the one explained above. Thanks
for your help.

Munagala V. S. Ramanath

unread,
Apr 14, 1999, 3:00:00 AM4/14/99
to
cla...@Starbase.NeoSoft.COM (Cameron Laird) writes:

>3. What's an example of something LGPL does


> NOT allow me to do?

In the context of a script library, if you were to embed the LGPL'd
library as a C string in your static executable you would be violating
LGPL since the user has no way of replacing the LGPL part with a fixed
version.

Ram

Bryan Oakley

unread,
Apr 14, 1999, 3:00:00 AM4/14/99
to

Hmmm. So I guess that means you can't use LGPL's code in a wrapped
script. Unless, I suppose, you put in some mechanism to override the
wrapped code if a more recent version is found on disk somewhere.

*sigh*. Why doesn't everyone just use a BSD style license? It's just
soooo much simpler on the whole world, isn't it? Sure, it doesn't
promote any political agenda like the GPL, but so what? has the GPL
really caused any change in the way people do business, other than to
discourage the use of GPL'd code by some people?

I'm all for GNU and use some of their tools constantly, but I hate the
viral effects of their licensing scheme.

Not that I want to get into a licensing debate. I just don't understand
why free software can't be completely and totally free in every sense of
the word.

--
Bryan Oakley mailto:oak...@channelpoint.com
ChannelPoint, Inc. http://purl.oclc.org/net/oakley

Joshua E. Rodd

unread,
Apr 14, 1999, 3:00:00 AM4/14/99
to
Bryan Oakley wrote:
> *sigh*. Why doesn't everyone just use a BSD style license? It's just
> soooo much simpler on the whole world, isn't it?

Because your code ends up wrapped up in orphans like SunOS 4 or IBM's
PMX.

> Sure, it doesn't
> promote any political agenda like the GPL, but so what? has the GPL
> really caused any change in the way people do business, other than to
> discourage the use of GPL'd code by some people?

Yes. Look at all the freed code we get from Cygnus for use in our own
freed compiler. There have been derivitates of the BSD compiler as well,
but are we getting code back from them?

The GPL prevents others from taking away my freedom by taking my code,
tweaking it, and then selling it back to me and restricting what I can
do with it!

> I'm all for GNU and use some of their tools constantly, but I hate the
> viral effects of their licensing scheme.

Then don't use the GPL. I personally love the effect.

> Not that I want to get into a licensing debate. I just don't understand
> why free software can't be completely and totally free in every sense of
> the word.

That's like saying ``Why can't a society be totally free, where I have the
right to steal and kill from other people?''

BSD style licences leave you naked to commercial exploitation. In the
case of some types of software, that's fine--e.g. X11. In the case of
other types of software, that's not fine at all--e.g. gcc.

Chang LI

unread,
Apr 14, 1999, 3:00:00 AM4/14/99
to
Bryan Oakley wrote:
>

> Hmmm. So I guess that means you can't use LGPL's code in a wrapped
> script. Unless, I suppose, you put in some mechanism to override the
> wrapped code if a more recent version is found on disk somewhere.
>

That is to say when you use the LGPL code you have to call it
with the source command, is it?



> *sigh*. Why doesn't everyone just use a BSD style license? It's just

> soooo much simpler on the whole world, isn't it? Sure, it doesn't


> promote any political agenda like the GPL, but so what? has the GPL
> really caused any change in the way people do business, other than to
> discourage the use of GPL'd code by some people?
>

How about the Artistic License used by Perl? It is better than LGPL?



> I'm all for GNU and use some of their tools constantly, but I hate the
> viral effects of their licensing scheme.
>

Ya. It is like virus.



> Not that I want to get into a licensing debate. I just don't understand
> why free software can't be completely and totally free in every sense of
> the word.
>

> --
> Bryan Oakley mailto:oak...@channelpoint.com
> ChannelPoint, Inc. http://purl.oclc.org/net/oakley

--
--------------------------------------------------------------
Chang LI, Neatware
email: cha...@neatware.com
--------------------------------------------------------------

Joshua E. Rodd

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
Chang LI wrote:
> That is to say when you use the LGPL code you have to call it
> with the source command, is it?

I'm afraid I don't understand what you're saying. The LGPL is intended
for binary libraries anyway, not scripting libraries.

> How about the Artistic License used by Perl? It is better than LGPL?

The Artistic licence is somewhat more restrictive than the LGPL.

It's not as bad as the qmail licence though; of course, the author of
qmail has every right to licence his software as he sees fit.

> > I'm all for GNU and use some of their tools constantly, but I hate the
> > viral effects of their licensing scheme.
> Ya. It is like virus.

One could as well argue that traditional EULAs are viral because once
I install (for example) IBM DB2 on a machine, I can't legally copy its
disk to another machine, or archive my filesystem and send it to someone
else. It ``infects'' my filesystem.

On the topic of GPL vs. BSD/MIT licencing, it's interesting to note that
some of the main proponents of it because it's more ``free'' are the
very people who distribute extremely non-free books (including well-known
publications with depictions of various fauna) and some of whom (though
certanily

Chang LI

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
Joshua E. Rodd wrote:
>
> Chang LI wrote:
> > That is to say when you use the LGPL code you have to call it
> > with the source command, is it?
>
> I'm afraid I don't understand what you're saying. The LGPL is intended
> for binary libraries anyway, not scripting libraries.
>

???. Someone applied LGPL on Tcl scripts. That is I can not wrap this
scripts into a single .exe file. That is my question. What I can do is
to use "source LGPLLIB.tcl" when I try to use the LGPL library.

> > How about the Artistic License used by Perl? It is better than LGPL?
>
> The Artistic licence is somewhat more restrictive than the LGPL.
>

???. I do not know LGPL well. The Artistic License seems more like BSD
except try to keep the unify of the "Standard Version". And it seems
to suitable for script library.

[snip]

> On the topic of GPL vs. BSD/MIT licencing, it's interesting to note that
> some of the main proponents of it because it's more ``free'' are the
> very people who distribute extremely non-free books (including well-known
> publications with depictions of various fauna) and some of whom (though
> certanily

GPL's big problem is that you can not apply them for commercial
applications.
It ask you to exposue your own extensions. It deprives your private
rights.

LGPL is more complicated. The problem is that LGPL makes people confuse
with
the GPL. And it is not very clear what are its special limitations.

Joshua E. Rodd

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
Chang LI wrote:
> ???. Someone applied LGPL on Tcl scripts. That is I can not wrap this
> scripts into a single .exe file. That is my question. What I can do is
> to use "source LGPLLIB.tcl" when I try to use the LGPL library.

You could; simply release your code under the (L)GPL as well. If you
won't want to do that, then don't use the LGPL'd code, or use it in
a seperate, well-defined library with dynamic linking.

> ???. I do not know LGPL well. The Artistic License seems more like BSD
> except try to keep the unify of the "Standard Version". And it seems
> to suitable for script library.

The Artistic licence prevents you from making substantial changes
and redistributing the code, in essence. It was a good attempt at
a GPL-like licence, but I find it's too restrictive--OTOH, it does
keep vendors from shipping extremely broken implementations of Perl,
as they're wont to do with typical BSD'd software.

> GPL's big problem is that you can not apply them for commercial
> applications.
> It ask you to exposue your own extensions. It deprives your private
> rights.

Not really. You can do anything you want commercially with GPL'd code,
except take that code, link it with your own code, and then deprive
other people of their rights to use that code.

You don't have to use GPL'd code if you don't want to. One reason I
release my OS/2 code under the GPL (or will release; I don't have
anything even alpha-quality yet but will RSN) is because I don't want
Stardock et al. taking BSD'd code and rereleasing it with high
prices and minor tweaks.

> LGPL is more complicated. The problem is that LGPL makes people confuse
> with
> the GPL. And it is not very clear what are its special limitations.

It's quite clear to me.

Jochem Huhmann

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
In article <37161C52...@noah.dhs.org>,
"Joshua E. Rodd" <jer...@noah.dhs.org> writes:

JER> Chang LI wrote:
>> ???. Someone applied LGPL on Tcl scripts. That is I can not wrap this
>> scripts into a single .exe file. That is my question. What I can do is
>> to use "source LGPLLIB.tcl" when I try to use the LGPL library.
JER>
JER> You could; simply release your code under the (L)GPL as well. If you
JER> won't want to do that, then don't use the LGPL'd code, or use it in
JER> a seperate, well-defined library with dynamic linking.

You may even wrap a LGPL'ed library into a single application
file. But you have to provide some hook, so that the user/customer can
install a newer version of that library which will then be used
instead of the wrapped one. As a customer I would like such a feature
anyway.

>> GPL's big problem is that you can not apply them for commercial
>> applications.
>> It ask you to exposue your own extensions. It deprives your private
>> rights.

JER>
JER> Not really. You can do anything you want commercially with GPL'd code,
JER> except take that code, link it with your own code, and then deprive
JER> other people of their rights to use that code.

Yes. Who writes the code has the right to choose the license for that
code. That's his freedom. If he gives his code away under the LGPL,
you have to accept this license, which says that you may not restrict
the users of this code to modify or replace it with a newer
version. This may raise some technical difficulties sometimes, but it
is a fine and fair solution for developers, vendors and users. Freedom
without liabilities and commitments has not been seen yet.

JER> It's quite clear to me.

Yes, it's always a good idea to read the actual licence before
objecting it ;-)


Jochem

--
# Jochem Huhmann <j...@gmx.net> Duisburg (Germany)

Steven G. Johnson

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
"Joshua E. Rodd" <jer...@noah.dhs.org> wrote:
> The Artistic licence prevents you from making substantial changes
> and redistributing the code, in essence. It was a good attempt at
> a GPL-like licence, but I find it's too restrictive--OTOH, it does
> keep vendors from shipping extremely broken implementations of Perl,
> as they're wont to do with typical BSD'd software.

I don't think this is very accurate. The Artistic license allows you to
do essentially anything you want, provided you change the name of the
program. The whole license is full of loopholes of this sort (see clauses
3c and 4c in http://www.opensource.org/artistic-license.html). Which is
fine if that's what you want in a license. It's just not what I would
consider "GPL-like", and is more similar in effect to X11 or BSD, as far
as I can tell.

Cordially,
Steven G. Johnson

PS. I am not a lawyer.

Joshua E. Rodd

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
"Steven G. Johnson" wrote:
> I don't think this is very accurate. The Artistic license allows you to
> do essentially anything you want, provided you change the name of the
> program. The whole license is full of loopholes of this sort (see clauses
> 3c and 4c in http://www.opensource.org/artistic-license.html). Which is
> fine if that's what you want in a license. It's just not what I would
> consider "GPL-like", and is more similar in effect to X11 or BSD, as far
> as I can tell.

True. IIRC the Artistic was intended to allow vendors to make minor
tweaks to Perl and then ship binaries with their UNIX platforms without
having to release the source to the tweaks, without "giving away the
store" as happened with BSD which then morphed into SunOS/BSDi/et al.
or X11 which has morphed into dozens (hundreds?) of vendor-specific
X implementations.

I did go back and re-read the licence; I didn't realise it was X11 like.
As someone else just said, it always pays to read a licence before you
complain about it. =)

Cameron Laird

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
In article <ramFA6...@netcom.com>,

Munagala V. S. Ramanath <r...@netcom.com> wrote:
>cla...@Starbase.NeoSoft.COM (Cameron Laird) writes:
>
>>3. What's an example of something LGPL does
>> NOT allow me to do?
>
>In the context of a script library, if you were to embed the LGPL'd
>library as a C string in your static executable you would be violating
>LGPL since the user has no way of replacing the LGPL part with a fixed
>version.
>
>Ram

I'm angry.

With a situation that we collectively create,
not with any individual. I have a lot of con-
fidence that everyone in this thread, for
example, is sincere and well-meaning.

I originally intended to write a long diatribe
about the absurdities of existing licenses,
our discussions of them, our discussions about
those discussions, ... I've cooled off enough
to forego that. Please, though, don't try to
persuade me that "all I have to do is read the
license" or some comparable platitude.

One of the best parts of this thread is Mr.
Ramanath's statement above,


In the context of a script library,
if you were to embed the LGPL'd
library as a C string in your static
executable you would be violating
LGPL since the user has no way of
replacing the LGPL part with a
fixed version.

This is good because I can understand what it
means.

It's not so good, because I don't agree with it.
In fact, I have weak plans for a future project
that will involve the LGPLed BWidget, compiled
into a C string (or something very close), and
I anticipated no controversy that this would NOT
violate the license. I'd also supply the customer
with a generation procedure that would suffice,
in principle, to re-create the deliverable, so I
concluded I'm complying with the license.

Rather than detailing what compilers are involved,
whether *their* licenses conflict with this inter-
pretation, the ontology of object files, ..., I'll
just conclude by saying this stuff feels very,
very wrong to me. I like to spend more of my time
designing good software solutions than fussing with
licenses, and my experience is that I often don't
have that privilege.

Jochem Huhmann

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
In article <7f5is2$klp$1...@starbase.neosoft.com>,
cla...@Starbase.NeoSoft.COM (Cameron Laird) writes:
CL> It's not so good, because I don't agree with it.
CL> In fact, I have weak plans for a future project
CL> that will involve the LGPLed BWidget, compiled
CL> into a C string (or something very close), and
CL> I anticipated no controversy that this would NOT
CL> violate the license. I'd also supply the customer
CL> with a generation procedure that would suffice,
CL> in principle, to re-create the deliverable, so I
CL> concluded I'm complying with the license.
CL>
CL> Rather than detailing what compilers are involved,
CL> whether *their* licenses conflict with this inter-
CL> pretation, the ontology of object files, ..., I'll
CL> just conclude by saying this stuff feels very,
CL> very wrong to me. I like to spend more of my time
CL> designing good software solutions than fussing with
CL> licenses, and my experience is that I often don't
CL> have that privilege.

Just as a note: You are always free to contact the author of the
library (unifix in this case) and ask for a special license. They
might say "well, do as you like, it's ok" or offer to sell a
commercial license to you. Very often it's much more simpler just to
ask than to fuss with license stuff. If there are enough of such
questions, they may even decide to change the license to something
simpler. Sending them a note is a quick and cheap way to clarify this.

Chang LI

unread,
Apr 15, 1999, 3:00:00 AM4/15/99
to
Joshua E. Rodd wrote:
>
> Chang LI wrote:
> > ???. Someone applied LGPL on Tcl scripts. That is I can not wrap this
> > scripts into a single .exe file. That is my question. What I can do is
> > to use "source LGPLLIB.tcl" when I try to use the LGPL library.
>
> You could; simply release your code under the (L)GPL as well. If you
> won't want to do that, then don't use the LGPL'd code, or use it in
> a seperate, well-defined library with dynamic linking.
>

Is it possible to wrap the modified LGPL library in a exe file and
distribute
the unmodified LGPL library together?

It seems no way to wrap a LGPL libray into a single executable file.
But I think use it as an extension is free in any meaning.

lvi...@cas.org

unread,
Apr 16, 1999, 3:00:00 AM4/16/99
to

According to Cameron Laird <cla...@Starbase.NeoSoft.COM>:
:very wrong to me. I like to spend more of my time
:designing good software solutions than fussing with
:licenses, and my experience is that I often don't
:have that privilege.

Cameron, for those programming in a commercial environment, an even worse
(in my opinion) situation is that of software patents. _YOU_ try to research
whether any of the _algorithms_ you are using are not patented! It's not going
to be an easy job, but to ''safely'' market a product, without worry that
someone is going to come along and slap you with a restraining order right
after you've spent several man-years of work on a product, you better know
whether you are safe or not...

What a nightmare.

Most companies today ignore the whole issue - hoping that IBM, Microsoft,
etc. don't decide to press the issue. But every now and then someone
plays the bad guy (example - Unisys and the LZW algorithm vs GIF ...) and
the whole industry goes into a spasm.

--
<URL: mailto:lvi...@cas.org> Quote: Saving the world before bedtime.
<*> O- <URL: http://www.purl.org/NET/lvirden/>
Unless explicitly stated to the contrary, nothing in this posting
should be construed as representing my employer's opinions.

Ottavio G. Rizzo

unread,
Apr 16, 1999, 3:00:00 AM4/16/99
to
"Joshua E. Rodd" <jer...@noah.dhs.org> writes:

> On the topic of GPL vs. BSD/MIT licencing, it's interesting to note that
> some of the main proponents of it because it's more ``free'' are the
> very people who distribute extremely non-free books (including well-known
> publications with depictions of various fauna) and some of whom (though
> certanily


That same very people have stated many times that it is up to the
author to decide about the "closeness" of the book (even if they
recommend against it): the NAG is free, and LDD is (or is going to be:
I don't know, since I got a free copy from the author anyhow :-) free,
too.

--
Ottavio Rizzo IRMAR, Campus de Beaulieu
ri...@maths.univ-rennes1.fr Université de Rennes 1
Tél +33 (0)2 99 28 67 92 35042 RENNES cedex
Fax +33 (0)2 99 28 67 90 FRANCE

DUPERVAL, LAURENT

unread,
Apr 16, 1999, 3:00:00 AM4/16/99
to
lvi...@cas.org wrote:
>
> Cameron, for those programming in a commercial environment, an even worse
> (in my opinion) situation is that of software patents.
>
<snip>

> What a nightmare.
>

No kidding...

> Most companies today ignore the whole issue - hoping that IBM, Microsoft,
> etc. don't decide to press the issue. But every now and then someone
> plays the bad guy (example - Unisys and the LZW algorithm vs GIF ...) and
> the whole industry goes into a spasm.
>

Yeah. Just check out what's going on with the XV program. It's totally
idiotic. Algorithms should not be patentable. And if they must be, the
length of the patent should be much smaller than the default (50 years?
Should be closer to 2 years).

All this 'cuz lawyers find they don't make enough money...

L

--
Penguin Power!

Laurent Duperval - CGI
5 Place Ville Marie, Suite 1600 South, Montreal H3B 2G2
Tel: (514) 870-0879 Fax: (514) 391-2212
EMAIL: mailto:laurent....@cgi.ca

Frederic BONNET

unread,
Apr 16, 1999, 3:00:00 AM4/16/99
to
Hi Larry,

lvi...@cas.org wrote:
> Cameron, for those programming in a commercial environment, an even worse

> (in my opinion) situation is that of software patents. _YOU_ try to research
> whether any of the _algorithms_ you are using are not patented! It's not going
> to be an easy job, but to ''safely'' market a product, without worry that
> someone is going to come along and slap you with a restraining order right
> after you've spent several man-years of work on a product, you better know
> whether you are safe or not...
>
> What a nightmare.

I wholeheartly agree. Software patenting is a total nonsense. But it
seems to be a US-specific problem. Here in France, software is regarded
as a mathematical method, and thus not patentable as is, it must be part
of a complete industrial system. I think this is the same in the whole
EU, as the European Patent Office try to harmonize the many EU members'
patents regulations. This needs to be confirmed, but I've never heard of
any granted software patent on this side of the pond.
I personally think the US Patents Office is a bunch of incompetent
people. They have the reputation to grant the most improbable patents.
Look at the patent granted to Microsoft for style sheet technology,
without even checking for prior art. I know there are other significant
examples.

See you, Fred
--
Frédéric BONNET frederi...@ciril.fr
---------------------------------------------------------------
"Theory may inform, but Practice convinces"
George Bain

Donal K. Fellows

unread,
Apr 19, 1999, 3:00:00 AM4/19/99
to
In article <3717457A...@cgi.ca>,

DUPERVAL, LAURENT <laurent....@cgi.ca> wrote:
> Yeah. Just check out what's going on with the XV program. It's totally
> idiotic. Algorithms should not be patentable. And if they must be, the
> length of the patent should be much smaller than the default (50 years?
> Should be closer to 2 years).

18 years is the current length of patent, IIRC. The Unisys LZW-patent
runs out fairly soon.

Donal.
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ fell...@cs.man.ac.uk
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>

Paul Duffin

unread,
Apr 19, 1999, 3:00:00 AM4/19/99
to
Frederic BONNET wrote:
>
> Hi Larry,
>
> lvi...@cas.org wrote:
> > Cameron, for those programming in a commercial environment, an even worse
> > (in my opinion) situation is that of software patents. _YOU_ try to research
> > whether any of the _algorithms_ you are using are not patented! It's not going
> > to be an easy job, but to ''safely'' market a product, without worry that
> > someone is going to come along and slap you with a restraining order right
> > after you've spent several man-years of work on a product, you better know
> > whether you are safe or not...
> >
> > What a nightmare.
>
> I wholeheartly agree. Software patenting is a total nonsense. But it
> seems to be a US-specific problem. Here in France, software is regarded
> as a mathematical method, and thus not patentable as is, it must be part
> of a complete industrial system. I think this is the same in the whole
> EU, as the European Patent Office try to harmonize the many EU members'
> patents regulations. This needs to be confirmed, but I've never heard of
> any granted software patent on this side of the pond.
> I personally think the US Patents Office is a bunch of incompetent
> people. They have the reputation to grant the most improbable patents.
> Look at the patent granted to Microsoft for style sheet technology,
> without even checking for prior art. I know there are other significant
> examples.
>

I am afraid that I am one of those people who betrays the whole
community for a fast buck. (I have just received a cheque from IBM for
submitting a patent application). IBM is obviously the biggest culprit
as it has had the most patents granted each year for the last 6 years
and earns over $1 billion a year from licensing its patents.

While I understand the need for companies who have invested a lot of
their money to protect their intellectual property I do not agree with
the way it happens. I think that there should definitely be a small time
limit on the length a patent is protected, possibly with some slow
degredation of its 'value' over time. Also the cost of licensing the
patent should be some multiple of the sale price of a single item so
free software would not have to pay. However this will never happen so
we have to live with it and do what we can outside it.

P.S. I could possibly have patented the Stubs mechanism but did not
because I though that it was so important to the community.

--
Paul Duffin
DT/6000 Development Email: pdu...@hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880 International: +44 1962-816880

DUPERVAL, LAURENT

unread,
Apr 19, 1999, 3:00:00 AM4/19/99
to
Paul Duffin wrote:
>
>
> P.S. I could possibly have patented the Stubs mechanism but did not
> because I though that it was so important to the community.
>

Hmmm.. But won't someone else be allowed to come around and patent it?
Just like some company patented the email feedback mechanism? You know:
you subscribe to something and get the information periodically by
email. Yes, it was patented even though that's been around for ages.

Laurent

Paul Duffin

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to
DUPERVAL, LAURENT wrote:
>
> Paul Duffin wrote:
> >
> >
> > P.S. I could possibly have patented the Stubs mechanism but did not
> > because I though that it was so important to the community.
> >
>
> Hmmm.. But won't someone else be allowed to come around and patent it?
> Just like some company patented the email feedback mechanism? You know:
> you subscribe to something and get the information periodically by
> email. Yes, it was patented even though that's been around for ages.
>

While it may be possible to patent something which already exists (the
patent office does not know everything) I do not think that it is
possible to enforce the patent. All that Scriptics (or anyone else who
uses it) would have to do would be to prove that there was prior art.
How they prove it is another matter.

This is all how I think it should / does work but I could be totally
wrong.

Donal K. Fellows

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to
In article <371B3B69...@cgi.ca>,

DUPERVAL, LAURENT <laurent....@cgi.ca> wrote:
> Hmmm.. But won't someone else be allowed to come around and patent it?
> Just like some company patented the email feedback mechanism? You know:
> you subscribe to something and get the information periodically by
> email. Yes, it was patented even though that's been around for ages.

You (should) only get granted a patent if there is no Prior Art (i.e.
if nobody else has done it before) and if the patent does mistakenly
get granted, the existence of Prior Art makes demolishing the patent
in court relatively easy.

I'm not sure if the overall concept of the stub mechanism is due to
Paul or not anyway...

Wasn't there an archive of all traffic on clt a while back? Probably
1997, but maybe earlier than that, or even possibly sometime in early
1998. Before DejaNews was all-powerful...

Cameron Laird

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to
In article <7fi02s$spg$1...@m1.cs.man.ac.uk>,
Donal K. Fellows <fell...@cs.man.ac.uk> wrote:
.
.

.
>I'm not sure if the overall concept of the stub mechanism is due to
>Paul or not anyway...
There are at least a couple of distinct ideas
in Stubs. I believe at least one of them ar-
rived with Paul first by way of private e-mail.
I also believe, of course, that Paul deserves a
great deal of credit for pushing Stubs forward
as he has done.

>
>Wasn't there an archive of all traffic on clt a while back? Probably
>1997, but maybe earlier than that, or even possibly sometime in early
>1998. Before DejaNews was all-powerful...
.
.
.
<URL:http://
http://Starbase.NeoSoft.COM/~claird/news.lists/compnewsgroup_archives.html#tcl>

Chang LI

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to
Donal K. Fellows wrote:
>
>

Paul is credit for the real stub design and implementation.
It is good for Tcl community he published this.
The real use of stub is coming soon.


> I'm not sure if the overall concept of the stub mechanism is due to
> Paul or not anyway...
>

> Wasn't there an archive of all traffic on clt a while back? Probably
> 1997, but maybe earlier than that, or even possibly sometime in early
> 1998. Before DejaNews was all-powerful...
>

> Donal.
> --
> Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ fell...@cs.man.ac.uk
> -- The small advantage of not having California being part of my country would
> be overweighed by having California as a heavily-armed rabid weasel on our
> borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>

--

Joshua E. Rodd

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to
Chang LI wrote:
> Donal K. Fellows wrote:
> Paul is credit for the real stub design and implementation.
> It is good for Tcl community he published this.
> The real use of stub is coming soon.

As one who knows nought about Tcl, what is stub? And how would
IBM's numerous patents relate to it?

Chang LI

unread,
Apr 20, 1999, 3:00:00 AM4/20/99
to

This is the description from the TEA Summit:

"The Tcl stubs mechanism defines a way to dynamically bind extensions
to a particular Tcl implementation at run time. This is
accomplished by exporting function tables that define a particular
interface. All access to the interface in question is done through
offsets into the function table so there are no direct references to
any of the library's symbols."

Without stub, for example, suppose you build a .exe file that includes
Tcl scripts, then you can not load a .dll extension with the load or
package command in your Tcl scripts. This will greatly limit the
extensibility. It is not necessary for extension author to provide
.lib file to you. And it is not necessary for you to use the .lib
in your link. The idea of stub is cool. It looks like "give me your
key I can get all in you room".

I do not know if IBM had applied the stub patent. But if they do
before (not now) it is very possible they can get it. That may damage
the Tcl community. Stub first came from Tcl community because there
were strong needs here. Others may follow later.

I am not sure how other patents relate to it. Stub itself is a
simple and complete idea.

It is hard to say the software patent. Many applications are ridiculous.
But the tendency is more and more companies may use it to protect their
IPs.
Many people may have to shut up their mouth.

Donal K. Fellows

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
In article <7fid2j$7ft$1...@Starbase.NeoSoft.COM>,

Cameron Laird <cla...@Starbase.NeoSoft.COM> wrote:
> Donal K. Fellows <fell...@cs.man.ac.uk> wrote:
>> I'm not sure if the overall concept of the stub mechanism is due to
>> Paul or not anyway...
>
> There are at least a couple of distinct ideas in Stubs. I believe
> at least one of them ar- rived with Paul first by way of private
> e-mail. I also believe, of course, that Paul deserves a great deal
> of credit for pushing Stubs forward as he has done.

Oh, all credit to Paul for the design and implementation of what we
have. But I suspect that the idea of stubs for libraries was mooted
earlier, which would make a patent difficult to uphold. Which was my
point! :^)

>> Wasn't there an archive of all traffic on clt a while back? Probably
>> 1997, but maybe earlier than that, or even possibly sometime in early
>> 1998. Before DejaNews was all-powerful...
>

> http://Starbase.NeoSoft.COM/~claird/news.lists/compnewsgroup_archives.html

FYI, all the links there to tcl-related stuff seem to be dead.

Donal K. Fellows

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
In article <371D0F...@neatware.com>, Chang LI <cha...@neatware.com> wrote:
> Paul is credit for the real stub design and implementation.

Oh, I know this in great detail. I was on the mailing list and
contributed some early bits of code to illustrate things. Jean-Claude
Wippler and Paul Duffin can verify this...

Donal K. Fellows

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
In article <371D22CA...@noah.dhs.org>,

Joshua E. Rodd <jer...@noah.dhs.org> wrote:
> As one who knows nought about Tcl, what is stub?

Super-duper dynamic linking mechanism that permits linking in of new
code at run-time in such a way that all existing code gets to use the
updated stuff. It may even be possible for different interpreters
within the same executable to be running different versions of the
same code (IIRC.)

It also manages to make dynamic linking work properly on all platforms
that can actually do the loading of the code. This is a big deal,
since many platforms are quite subtly broken...

> And how would IBM's numerous patents relate to it?

Tut, tut. If you'd read the whole of this thread, you would know that
they don't. :^)

Paul Duffin

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
Donal K. Fellows wrote:
>
> I'm not sure if the overall concept of the stub mechanism is due to
> Paul or not anyway...
>

It is and it isn't. The original spark was the postscript in the
following posting from Jean-Claude Wippler. As you can see it has the
basic idea behind Stubs which I took and generalised and built upon with
help from both Jean-Claude and Jan Nijtmans.

http://www.dejanews.com/[ST_rn=ps]/getdoc.xp?AN=362333629

Paul Duffin

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
Chang LI wrote:
>
> I do not know if IBM had applied the stub patent. But if they do
> before (not now) it is very possible they can get it. That may damage
> the Tcl community. Stub first came from Tcl community because there
> were strong needs here. Others may follow later.
>

Nobody has a patent on the Stubs mechanism.

Paul Duffin

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
Donal K. Fellows wrote:
>
> In article <371D22CA...@noah.dhs.org>,
> Joshua E. Rodd <jer...@noah.dhs.org> wrote:
> > As one who knows nought about Tcl, what is stub?
>
> Super-duper dynamic linking mechanism that permits linking in of new
> code at run-time in such a way that all existing code gets to use the
> updated stuff. It may even be possible for different interpreters
> within the same executable to be running different versions of the
> same code (IIRC.)
>
> It also manages to make dynamic linking work properly on all platforms
> that can actually do the loading of the code. This is a big deal,
> since many platforms are quite subtly broken...
>
> > And how would IBM's numerous patents relate to it?
>
> Tut, tut. If you'd read the whole of this thread, you would know that
> they don't. :^)
>

I have fallen into that trap before, it is really strange how these
newservers of ours propagate messages around. I wouldn't be at all
surprised if there were purposely delaying messages just to confuse us.
;^).

Chang LI

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
Paul Duffin wrote:
>

European group started and finished the stub.
Anyway no patent on stub is very good for Tcl community.
Thanks for their innovation and contribution.



> Donal K. Fellows wrote:
> >
> > I'm not sure if the overall concept of the stub mechanism is due to
> > Paul or not anyway...
> >
>
> It is and it isn't. The original spark was the postscript in the
> following posting from Jean-Claude Wippler. As you can see it has the
> basic idea behind Stubs which I took and generalised and built upon with
> help from both Jean-Claude and Jan Nijtmans.
>
> http://www.dejanews.com/[ST_rn=ps]/getdoc.xp?AN=362333629
>

> --
> Paul Duffin
> DT/6000 Development Email: pdu...@hursley.ibm.com
> IBM UK Laboratories Ltd., Hursley Park nr. Winchester
> Internal: 7-246880 International: +44 1962-816880

--

Paul Duffin

unread,
Apr 21, 1999, 3:00:00 AM4/21/99
to
Donal K. Fellows wrote:
>
> In article <7fid2j$7ft$1...@Starbase.NeoSoft.COM>,
> Cameron Laird <cla...@Starbase.NeoSoft.COM> wrote:
> > Donal K. Fellows <fell...@cs.man.ac.uk> wrote:
> >> I'm not sure if the overall concept of the stub mechanism is due to
> >> Paul or not anyway...
> >
> > There are at least a couple of distinct ideas in Stubs. I believe
> > at least one of them ar- rived with Paul first by way of private
> > e-mail. I also believe, of course, that Paul deserves a great deal
> > of credit for pushing Stubs forward as he has done.
>
> Oh, all credit to Paul for the design and implementation of what we
> have. But I suspect that the idea of stubs for libraries was mooted
> earlier, which would make a patent difficult to uphold. Which was my
> point! :^)
>

True but as I get paid when a patent application gets accepted by the
patent office and not when IBM gets money from it I don't mind ;-).

Bruce S. O. Adams

unread,
Apr 22, 1999, 3:00:00 AM4/22/99
to

Frederic BONNET wrote:

> Hi Larry,
>
> lvi...@cas.org wrote:
> > Cameron, for those programming in a commercial environment, an even worse
> > (in my opinion) situation is that of software patents. _YOU_ try to research
> > whether any of the _algorithms_ you are using are not patented! It's not going
> > to be an easy job, but to ''safely'' market a product, without worry that
> > someone is going to come along and slap you with a restraining order right
> > after you've spent several man-years of work on a product, you better know
> > whether you are safe or not...
> >
> > What a nightmare.
>
> I wholeheartly agree. Software patenting is a total nonsense. But it
> seems to be a US-specific problem. Here in France, software is regarded
> as a mathematical method, and thus not patentable as is, it must be part
> of a complete industrial system. I think this is the same in the whole
> EU, as the European Patent Office try to harmonize the many EU members'
> patents regulations. This needs to be confirmed, but I've never heard of
> any granted software patent on this side of the pond.
> I personally think the US Patents Office is a bunch of incompetent
> people. They have the reputation to grant the most improbable patents.
> Look at the patent granted to Microsoft for style sheet technology,
> without even checking for prior art. I know there are other significant
> examples.
>

> See you, Fred
> --
> Frédéric BONNET frederi...@ciril.fr
> ---------------------------------------------------------------
> "Theory may inform, but Practice convinces"
> George Bain

I think you can patent algorithms in the UK too, in certain circumstances. I
certainly remember seeing the odd article in the patents section of new scientist
(a magazine for those that don't know).
Regards,
Bruce A.

Russ Allbery

unread,
Apr 22, 1999, 3:00:00 AM4/22/99
to
In gnu.misc.discuss, Donal K Fellows <fell...@cs.man.ac.uk> writes:

> You (should) only get granted a patent if there is no Prior Art (i.e.
> if nobody else has done it before)

Emphasis on the word *should*. The patent office apparently no longer
considers this to be their job, given the sort of crap they're giving
patents for these days. Apparently they think the courts should do all of
their research for them. I'd really like to see a legal judgement against
the patent office for incompetence.

> and if the patent does mistakenly get granted, the existence of Prior
> Art makes demolishing the patent in court relatively easy.

Nothing about going to court is ever really easy, though.

--
Russ Allbery (r...@stanford.edu) <URL:http://www.eyrie.org/~eagle/>

Bruce S. O. Adams

unread,
Apr 22, 1999, 3:00:00 AM4/22/99
to

"DUPERVAL, LAURENT" wrote:

> Paul Duffin wrote:
> >
> >
> > P.S. I could possibly have patented the Stubs mechanism but did not
> > because I though that it was so important to the community.
> >
>

> Hmmm.. But won't someone else be allowed to come around and patent it?
> Just like some company patented the email feedback mechanism? You know:
> you subscribe to something and get the information periodically by
> email. Yes, it was patented even though that's been around for ages.
>

> Laurent

Technically you aren't allowed to patent 'prior art', but sometimes
something slips through
the net.
Regards,
Bruce A.


Paul Duffin

unread,
Apr 23, 1999, 3:00:00 AM4/23/99
to
Jonathan Stott wrote:
>
> In article <371D5A...@neatware.com>, Chang LI <cha...@neatware.com> wrote:

> >Joshua E. Rodd wrote:
> >>
> >
> >This is the description from the TEA Summit:
> >
> >"The Tcl stubs mechanism defines a way to dynamically bind extensions
> >to a particular Tcl implementation at run time. This is
> >accomplished by exporting function tables that define a particular
> >interface. All access to the interface in question is done through
> >offsets into the function table so there are no direct references to
> >any of the library's symbols."
>
> This sounds like what Amiga's had to handle shared libraries 10+ years
> ago. There was exactly one officially fixed address in the machine
> (0x00000004) and it contained a pointer to the function table of the
> exec.library. Exec.library had the openlibrary() function, which took
> care of opening all other shared libraries. It also had a routine to
> [safely] replace system functions by putting a new address into the
> table.
>

Ah yes but I think I could still get it past the patent lawyers, I
seriously doubt that they would know the internals of an Amiga.

ray_j...@my-dejanews.com

unread,
Apr 26, 1999, 3:00:00 AM4/26/99
to
In article <7fi02s$spg$1...@m1.cs.man.ac.uk>,

fell...@cs.man.ac.uk (Donal K. Fellows) wrote:
> In article <371B3B69...@cgi.ca>,
> DUPERVAL, LAURENT <laurent....@cgi.ca> wrote:
> > Hmmm.. But won't someone else be allowed to come around and patent it?
> > Just like some company patented the email feedback mechanism? You know:
> > you subscribe to something and get the information periodically by
> > email. Yes, it was patented even though that's been around for ages.
>
> You (should) only get granted a patent if there is no Prior Art (i.e.
> if nobody else has done it before) and if the patent does mistakenly

> get granted, the existence of Prior Art makes demolishing the patent
> in court relatively easy.
>
> I'm not sure if the overall concept of the stub mechanism is due to
> Paul or not anyway...
>
> Wasn't there an archive of all traffic on clt a while back? Probably
> 1997, but maybe earlier than that, or even possibly sometime in early
> 1998. Before DejaNews was all-powerful...

The stubbs mechanism Paul & company came up with is actually very much
like the Adobe's extension mechanism on the Mac and other extension
mechanisms made before dynamic loading came into existance. In fact,
that approach was something we originally considered for the Tcl load
mechanism - but the quick fix of just using native dynamic loading
won out...

Ray


-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own

Reply all
Reply to author
Forward
0 new messages