[A bit rambling] Open source licensing being questioned byanti-copyright types

[Arne Vajhøj]

Norway. The English (non-authoritative) version of the law is here:
http://www.regjeringen.no/upload/KK...ns/Aandsverkloven_engelsk_versjon_nov2008.pdf

in particular para 3:
(...)

"The author may not waive his rights under the first and second
paragraphs, unless the use of the work in question is limited in
nature and extent."

"(...) This right may not be waived by the author."


also para 38c:
(...)

"The right under the first paragraph can be neither waived not
transferred. (...)"


and para 39e:
(...)

"Any agreement that to the detriment of the author significantly
deviates from the provisions of the second, sixth and seventh
paragraph cannot be applied. The author may not waive his right
pursuant to the fifth paragraph."

(...)


There may be other examples, these are the ones I could easily find by
searching on "waive" in the document. Not all of the above may apply
to software, I didn't check. Some specifics of software are treated in
paras 39g,h,i.

FireFox and Thunderbird uses SQLite.

The authors of SQLite has released it as public domain.

If such a release is not valid in Norway, then all users
of FF and TB in Norway are breaking copyright.

I think a lot of IT people would be willing to testify about
"industry practice", "common sense", "the intention of the law"
etc. if that came to a trial.

Arne

 

[Arne Vajhøj]

I don't think so. Open source with the BSD or Apache license has so
little restriction on distribution that if all software code was public
domain, the BSD license and Apache license using development/business
models would pretty much be unaffected.

They seem to think otherwise, because they chose their licenses
over public domain.

If copyright disappeared tomorrow, the whole software industry would
eventually reorganize along the lines of such projects.

Unlikely.

Apache is getting a lot of money from IBM.

IBM making less money => less resources to Apache, Eclipse etc..

Now, since you seem to need some real hand-holding here, I'll post the
links:

http://www.againstmonopoly.org/
http://www.techdirt.com/

The top story right now is a speech by a biochemist and
something about T-shirts respectively.

Not exactly convincing as the place for serious Economic
science.

Most open source licenses do not actually enforce anything crucial to the
business model, so to the ability to profit from the software
development. Arguably even the GPL doesn't, but particularly most of the
non-"copyleft" ones just ask for attribution and maybe one or two other
minor things. Trademark law and laws against misrepresentation and fraud
could probably be used to cover such needs, and they can probably be done
without.

Apparently they believe otherwise, because they chose their
licenses.

Arne

 

[Arved Sandstrom]

FireFox and Thunderbird uses SQLite.

The authors of SQLite has released it as public domain.

If such a release is not valid in Norway, then all users
of FF and TB in Norway are breaking copyright.

I think a lot of IT people would be willing to testify about
"industry practice", "common sense", "the intention of the law"
etc. if that came to a trial.

Arne

You can make a reasonable argument that anyone who has put a statement
into their documentation disavowing copyright and explicitly releasing
it into the public domain (or so they believe) has in fact retained
copyright but effectively told users that they can do with the software
whatever they wish. This latter approach is one of the recommendations,
either that or using one of the more permissive licenses, for
accomplishing the same ends as placing something into the public domain
(which latter is difficult or impossible to do, and may be
counter-productive).

I'm no software lawyer but my reading on this subject tells me that
trying to disclaim copyright and put software into the public domain is
problematic. I recommend perusing
http://linuxmafia.com/faq/Licensing_and_Law/public-domain.html for
staters, but there is a lot of easily located discussion on this
subject...none of which purports to say that placing software into the
public domain is easy to do.

My own opinion on Unlicenses and the SQLite public domain declaration:
you can put the statements in to your heart's content. It just probably
won't have legal effect. Even the SQLite team recognizes that when they
include this as a reason for obtaining an SQLite license: "You are using
SQLite in a jurisdiction that does not recognize the right of an author
to dedicate their work to the public domain."

To sum up, my reading is that if an author or authors have stated that
they wish to disclaim copyright and place software into the public
domain, odds are very good that they haven't legally done so. But I
expect a reasonable legal case could be made that in such cases, while
the copyright is still held by the author(s), that they have in effect
granted users the permission to do whatever they like with the software.

AHS

 

[Bent C Dalager]

FireFox and Thunderbird uses SQLite.

The authors of SQLite has released it as public domain.

If such a release is not valid in Norway, then all users
of FF and TB in Norway are breaking copyright.

Even though the copyright holders have claimed the work to be in the
public domain I doubt it actually /is/ because there is no provision
in the law for that particular option. Of course, if SQLite is itself
from a nation other than Norway then things can become complicated and
it will be difficult to tell.

In practical terms I expect that this will be considered by the courts
to be a maximally permissive license, which still leaves a few aspects
of copyright in place - and is different from public domain.

Cheers,
Bent D.

 

[Bent C Dalager]

There are slightly different rules for copyright around the world,
but the fact is that public domain software is almost universal
de facto recognized.

I don't see that this is the case. What I do see is that there exist a
number of people who would /like/ public domain software to be
achievable, and there are some who act as if it is. You might say that
these people, at least, have de facto recognized public domain softare
but what is the benefit of this? Copyright is governed by copyright
law and I cannot see that the law recognizes the public domain in any
useful fashion when it comes to software (primarily because there is
no significant body of software from before 1930-ish).

They could have chose public domain.

Well, that is basically the point under debate: It's not clear that
one "can" choose public domain in this context. The only thing one can
choose is how to license one's copyrighted work. If you, as an author,
"choose" public domain then I believe that the best case outcome is
that the judge gives you some leeway for not having any legal
education and maps your nonsensical (in the legal sense) choise over
into a very permissive license governed by copyright.

Because the open source license gives some rights.

What rights are these, and who benefits from them?

Would it be a material change to protect closed source by
trade secret instead of copyright?

The material change would be that any source code you could get your
hands on would be yours to use as you wish, unless of course you
obtained it through industrial espionage.

The problem is that even though a lot of software is only used
inhouse, then most companies do not want to reject the possibility
to at some later point in time sell the software to external
customers.

To the extent that this possibility appears as an asset at all, if it
is given a realistic value it must be very very small indeed. From
what I have seen, in-house software projects tend to be completely
unsuitable for distribution to any other than the primary client and
would take gargantuan effort to change into a sellable product.

I expect the cost savings from being able to use any and all other
software without having to worry about licensing costs and liabilities
would by far dwarf whatever miniscule economic opportunity exists in
one's own in-house projects.

Otherwise corporations would rarely have a problem with GPL!

I expect the GPL resistance has very much to do with some fundamental
misunderstandings about how the GPL actually works. Surprisingly many
people seem to think that if they develop a GPL application then they
are automatically required to distribute it to the world and provide
support for it to anyone who asks. An accountant who actually believes
this will quickly point out that it effectively amounts to unlimited
liability and advise strongly against going anywhere near such a
license.

Cheers,
Bent D

 

[Lew]

In practical terms I expect that this will be considered by the courts
to be a maximally permissive license, which still leaves a few aspects
of copyright in place - and is different from public domain.

From the user's perspective, a difference that makes no difference is no
difference. They can safely use the product and leave the hair-splitting to
the lawyers.

 

[Bent C Dalager]

From the user's perspective, a difference that makes no difference is no
difference. They can safely use the product and leave the hair-splitting to
the lawyers.

Users cannot safely use the product if they end up violating any of
the inalienable rights, wrongly thinking that the work is in the
public domain.

Authors also cannot safely just write "public domain" and think it
will do what they think it will.

As an example of the latter, an author might be interested in maximum
exposure of his work and so desire to put it into the public domain in
order to make its wide dissemination legally uncomplicated. This would
seem an obvious strategy for an unknown author trying to make a name
for himself. As it is, however, the work retains its copyright and so
any institution that might not think twice about distributing public
domain works would need to evaluate the legal implications of doing so
with the copyrighted one. If the work simply says "this work is in the
public domain" that is likely to complicate matters even further since
it is unclear what the legal standing of this is while an actual
copyright license would be considerably more clear. In sum this would
tend to restrict the distribution of the work, which directly
contravenes the author's intentions in this case.

Cheers,
Bent D

 

[Jeff Higgins]

In dealing with Java,

Throughout this rambling discussion there has been no citation of case,
regulatory, or statutory law from any jurisdiction.

 

[Arne Vajhøj]

You can make a reasonable argument that anyone who has put a statement
into their documentation disavowing copyright and explicitly releasing
it into the public domain (or so they believe) has in fact retained
copyright but effectively told users that they can do with the software
whatever they wish. This latter approach is one of the recommendations,
either that or using one of the more permissive licenses, for
accomplishing the same ends as placing something into the public domain
(which latter is difficult or impossible to do, and may be
counter-productive).

I'm no software lawyer but my reading on this subject tells me that
trying to disclaim copyright and put software into the public domain is
problematic. I recommend perusing
http://linuxmafia.com/faq/Licensing_and_Law/public-domain.html for
staters, but there is a lot of easily located discussion on this
subject...none of which purports to say that placing software into the
public domain is easy to do.

They lawyers may find it very problematic.

But there does not seem to be any problems with software
where the authors put code (or try to put code) in public
domain.

My own opinion on Unlicenses and the SQLite public domain declaration:
you can put the statements in to your heart's content. It just probably
won't have legal effect. Even the SQLite team recognizes that when they
include this as a reason for obtaining an SQLite license: "You are using
SQLite in a jurisdiction that does not recognize the right of an author
to dedicate their work to the public domain."

To sum up, my reading is that if an author or authors have stated that
they wish to disclaim copyright and place software into the public
domain, odds are very good that they haven't legally done so. But I
expect a reasonable legal case could be made that in such cases, while
the copyright is still held by the author(s), that they have in effect
granted users the permission to do whatever they like with the software.

Could very well be that would be the judges approach.

If there are no legal possibilities for public domain in that
jurisdiction, then it seems as what best fits what both the author
and the users want.

Arne

 

[Arne Vajhøj]

Even though the copyright holders have claimed the work to be in the
public domain I doubt it actually /is/ because there is no provision
in the law for that particular option. Of course, if SQLite is itself
from a nation other than Norway then things can become complicated and
it will be difficult to tell.

In practical terms I expect that this will be considered by the courts
to be a maximally permissive license, which still leaves a few aspects
of copyright in place - and is different from public domain.

As I replied to Arved:

Could very well be that would be the judges approach.

If there are no legal possibilities for public domain in that
jurisdiction, then it seems as what best fits what both the author
and the users want.

Arne

 

[Arne Vajhøj]

Users cannot safely use the product if they end up violating any of
the inalienable rights, wrongly thinking that the work is in the
public domain.

Authors also cannot safely just write "public domain" and think it
will do what they think it will.

As an example of the latter, an author might be interested in maximum
exposure of his work and so desire to put it into the public domain in
order to make its wide dissemination legally uncomplicated. This would
seem an obvious strategy for an unknown author trying to make a name
for himself. As it is, however, the work retains its copyright and so
any institution that might not think twice about distributing public
domain works would need to evaluate the legal implications of doing so
with the copyrighted one. If the work simply says "this work is in the
public domain" that is likely to complicate matters even further since
it is unclear what the legal standing of this is while an actual
copyright license would be considerably more clear. In sum this would
tend to restrict the distribution of the work, which directly
contravenes the author's intentions in this case.

That is a theory.

But does SQLite distribution seem to have been restricted
due to being declared public domain?

Arne

 

[Arne Vajhøj]

I don't see that this is the case. What I do see is that there exist a
number of people who would /like/ public domain software to be
achievable, and there are some who act as if it is. You might say that
these people, at least, have de facto recognized public domain softare
but what is the benefit of this? Copyright is governed by copyright
law and I cannot see that the law recognizes the public domain in any
useful fashion when it comes to software (primarily because there is
no significant body of software from before 1930-ish).

Public domain software is very recognized.

Software by US government employees have always been public domain
and recognized as such.

Well, that is basically the point under debate: It's not clear that
one "can" choose public domain in this context.

Some people are choosing public domain.

These people could certainly have done the same.

The only thing one can
choose is how to license one's copyrighted work. If you, as an author,
"choose" public domain then I believe that the best case outcome is
that the judge gives you some leeway for not having any legal
education and maps your nonsensical (in the legal sense) choise over
into a very permissive license governed by copyright.

Most of the legal issues are centered about the fact that the
law in many countries does not mention such a concept.

But unless the law explicitly prohibits disclaiming copyright
(which is also does in some countries), then I would expect
it to be possible - the fact that something is not explicit
approved in the law should not mean that it is prohibited.

What rights are these, and who benefits from them?

It varies between licenses.

But the general purpose is to give the users certain rights.

(in a few cases also certain obligations)

The material change would be that any source code you could get your
hands on would be yours to use as you wish, unless of course you
obtained it through industrial espionage.

If it is sufficient that you did not do the industrial espionage
yourself to be allowed to use the code, then I think industrial
espionage would be a very profitable business.

If it need to be clean all the way, then it would be very
challenging to use software, because without any license
information how do you know where the code came from.

To the extent that this possibility appears as an asset at all, if it
is given a realistic value it must be very very small indeed. From
what I have seen, in-house software projects tend to be completely
unsuitable for distribution to any other than the primary client and
would take gargantuan effort to change into a sellable product.

Often it is.

But the companies does not think so.

It is like driving - some survey showed that 90% of people believe that
they drive better than average.

I expect the GPL resistance has very much to do with some fundamental
misunderstandings about how the GPL actually works. Surprisingly many
people seem to think that if they develop a GPL application then they
are automatically required to distribute it to the world and provide
support for it to anyone who asks. An accountant who actually believes
this will quickly point out that it effectively amounts to unlimited
liability and advise strongly against going anywhere near such a
license.

I think you are underestimating the corporate lawyers.

Arne

 

[Bent C Dalager]

That is a theory.

But does SQLite distribution seem to have been restricted
due to being declared public domain?

This is impossible to say of course.

In an attempt to unravel the latest series of posts: my point is that
open source projects that choose a copyright license rather than
"public domain" may very well have done so because they are familiar
with, and care about, the legal situation surrounding those choices.
They may not particularly /want/ to own a copyright but they realize
that as a point of legal reality they are forced to do so, and they
make the best out of a bad situation by choosing some very permissive
license. Therefore, the decision to have a license is not in itself an
indication that a project team feels it needs or even desires
copyright protection of their work.

Others may choose to pretend that the public domain is in fact a valid
choice for a work newer than 75 years (or so) but an examination of
the legal climate surrounding the issue would seem to indicate that
this is basically wishful thinking.

Cheers,
Bent D

 

[Andreas Leitgeb]

They lawyers may find it very problematic.

IANAL, but sometimes playing devil's advocate:

A company could write some useful code, and donate it to
public domain. Later, when others have started using it,
the company could be taken over, and the new owners may
point out, that that previous act of pd-izing was illegal
from the beginning and furtheron demand license-fees ...

 

[Bent C Dalager]

Public domain software is very recognized.

If you mean that the existence of the public domain is recognized,
then certainly. What does not appear to be recognized is the power of
an author to place his own works into the public domain, other than by
dying then waiting 75 years (or whatever).

Software by US government employees have always been public domain
and recognized as such.

Yes, I believe there is specific provision for this in US law. Can you
point to similar provisions for the public domain for privately
developed software?

Some people are choosing public domain.

These people could certainly have done the same.

By all appearances, choosing the public domain makes about as much
sense as choosing to put your software in blue. That is, it makes no
sense at all. Which, of course, doesn't actually in itself /prevent/
people from saying "I place my software in blue".

Most of the legal issues are centered about the fact that the
law in many countries does not mention such a concept.

But unless the law explicitly prohibits disclaiming copyright
(which is also does in some countries), then I would expect
it to be possible - the fact that something is not explicit
approved in the law should not mean that it is prohibited.

A central point in modern copyright law is that all works are
copyrighted by default. In order to opt out of such a scheme there
would need to be provision in the law for an author to do so. There
wouldn't appear to be any.

It varies between licenses.

But the general purpose is to give the users certain rights.

Can you give an example of one such right that would not also
automatically be available were the software to be in the public
domain?

If it is sufficient that you did not do the industrial espionage
yourself to be allowed to use the code, then I think industrial
espionage would be a very profitable business.

I do believe that it is, yes.

Trade secret laws vary widely between jurisdictions so the extent to
which a third party not involved in the actual espionage is affected
will vary.

If it need to be clean all the way, then it would be very
challenging to use software, because without any license
information how do you know where the code came from.

This is little different from the current situation: if code has no
license or copyright information you do not know its legal status and
so you cannot use it.

Chances are that under a pure trade secrets regime you would have an
easier time of it since unless you yourself were party to espionage
you will be seen to have acted in good faith. Conversely, under the
current default copyright regime you should know that all code is
always copyrighted except under rather special circumstances and so it
is much harder to claim good faith.

Often it is.

But the companies does not think so.

If this is the case, then removing copyright would be doing them a
favour.

I think you are underestimating the corporate lawyers.

In my experience lawyers often do not get involved in matters of
specific open source licenses: The idea gets killed long before a
lawyer is involved.

In the cases were lawyers actually /are/ involved on a per license
basis I would expect GPL to see much higher adoption as many of the
popular myths are likely to be dispelled by an educated reading of the
license.

One question it appears many corporate lawyers are asked is along the
lines of "can we use open source licensed software" to which a lawyer
is basically forced to answer "no" because in general you cannot: you
need to examine each individual open source license on its own merits
and cannot make a blanket statement about the set of all possible such
licenses except to reject the idea. Hence, perhaps, the "no open
source" policy in many companies.

Cheers,
Bent D

 

[Lew]

IANAL, but sometimes playing devil's advocate:

A company could write some useful code, and donate it to
public domain. Later, when others have started using it,
the company could be taken over, and the new owners may
point out, that that previous act of pd-izing was illegal
from the beginning and furtheron demand license-fees ...

They would lose. Adverse possession.

 

[Andreas Leitgeb]

They would lose. Adverse possession.

Your word in ${user.preferred.deity}'s ear.

 

[John B. Matthews]

Your word in ${user.preferred.deity}'s ear.

Just in: "Terrible Court Says It's Okay To Remove Content From The
Public Domain And Put It Back Under Copyright."

<http://www.techdirt.com/articles/20100621/2320049908.shtml>

Discussion: "Court Takes Away Some of the Public Domain."

<http://yro.slashdot.org/story/10/06/22/1724254/>

 

[ClassCastException]

Bullshit.

So, climate change deniers, creationists, etc. are not pseudoscientific
quacks and religious nuts?

That's news to me.

All the serious economic science expect there to be some effect but
consider it very difficult to estimate it.

There have been studies, by e.g. Boldrin and Levine. The measured effects
are negative, i.e. patents actually make things worse.

No. But it is by far the largest for most of the wellknown open source
companies.

Define "wellknown open source companies". If it's only companies like Red
Hat whose sole product is software, you're probably right; widen it to
include companies like IBM that are involved heavily with open source but
do other things as well ...

You can probably sell GPL software for the same price as a bottle of
bottled water.

But that will not make the software industry prosper.

Define "prosper"? Thinner margins doesn't equate to losses. So if by
"prosper" you mean companies like Microsoft laughing all the way to the
bank and pocketing billions, making people like Bill Gates the richest on
the planet, then no, probably not. If by "prosper" you mean "positive
growth, however small", then it's another story.

In practice, though, I would expect selling support to become more
lucrative than selling copies.

 

[ClassCastException]

They seem to think otherwise, because they chose their licenses over
public domain.

As Bent pointed out elsethread, it's not apparently a sure bet to
dedicate a work to the public domain. Legally astute open source
developers may choose super-permissive licenses as a legally-sure-bet
alternative that carries most of the same benefits.

Unlikely.

Apache is getting a lot of money from IBM.

IBM making less money => less resources to Apache, Eclipse etc..

Why would IBM be making less money? Selling copies of Apache and Eclipse
is not IBM's primary revenue stream. The last time I checked, selling
hardware was, or was one of them.

The top story right now is a speech by a biochemist and something about
T-shirts respectively.

Wait a few hours.

Apparently they believe otherwise, because they chose their licenses.

As Bent pointed out elsethread, it's not apparently a sure bet to
dedicate a work to the public domain. Legally astute open source
developers may choose super-permissive licenses as a legally-sure-bet
alternative that carries most of the same benefits.