Re: Alternative trigger condition.

["Ross N. Williams" <ross@rocksoft.com>]

Marc,

At 1:33 PM +0100 11/1/2000, Marc van Leeuwen wrote:

Thanks for your thoughtful comments on my licence. I think you have some
good points, and I think I'm going to have to go off and redraft this
clause and try and get it right this time.

But I also think you're mistaken on a few points, so I'll gratuitously
attack those now so I feel better about being mauled.


>Dear Ross N. Williams <ross@rocksoft.com>,
>you wrote
>
> > How's this for a new trigger condition for The Free World Licence:
> > 
> >      2.1 CONTRACT: This Licence is a legal contract between you
> >      and the Original Licensor (and possibly between you and
> >      other contributing Licensors as well). Licensor unilaterally
> >      grants you a copyright licence to receive and retain a copy
> >      of a compressed archive file containing the Module. However,
> >      by extracting the Module from such an archive for the first
> >      time, or by modifying the archive, or by otherwise
> >      installing, browsing, executing, modifying or distributing
> >      the Module, you indicate your acceptance of this Licence and
> >      you automatically enter into a contract with Licensor, as
> >      defined by this document.
>
>I'm sorry to say, but in my opinion there is no way this is going to work.
>
>For instance, since when do I need a copyright licence to receive and retain a
>copy of whatever? If somebody gives me a copy of some work then maybe he has
>violated some copyright in producing the copy, but as long as I refrain from
>making further copies this is not something I have to worry about. Look, I
>possess hundreds of books, for most of which I have no right (nor desire) to
>copy them, and some of which may contain material that was illegally copied
>(by their authors or publishers, how would I know?); however since I merely
>bought these books, there is nothing I have to worry about. In the electronic
>domain, this argument is essential. Every time one downloads a file by ftp,
>accesses a website, or receives an email, one may acquire copyrighted material
>that might have been illegally copied; if this would expose the recipient to
>legal persecution, there would be absolutely no way to safely operate a
>networked computer. There is only one sensible way that I can view this: the
>sender in the ftp/http/email transfer is making a copy, and should make sure
>no copyright is being violated, but the recipient need not have any worries as
>long as he does not make further copies.

I think you are partly correct, but I think there is a burden on the
recipient to take "due care" or "due diligence" (or some other legal
term) to ensure that the material they are receiving is not copyrighted.

For example, if you read a newspaper that said that a bookseller
was selling illegal copies of a book, then if you then knowingly
bought the book from the bookseller, I believe that you'd be breaching
copyright.

Because copyright is the DEFAULT state of a work, when one receives
a work through a channel, there is, by default, a burden on the recipient
to establish that they have recieved a legal copy. For example, if you
happen across a web page with a photo on it, and the web page contains
no copyright notice, and you make a copy of the photo and use it for
your own personal use, then the fact that there was no copyright notice
does not mean that you are not in breach. By default one is in breach,
and must seek some reason for believing that one is not in breach.
This is, as you point out separately, the basis of the strength of
the GNU GPL.

The normal way of procuring assurance that a work being received is
not in breach is to rely on the signals of the channel. If you purchase
a book from a bookshop, it is reasonable for one to assume that that
the book is not in breach. Similarly, if you download software from
the software's vendor's website, paying money to do so, then you have
a strong basis for believing the copy is not in breach. You're in the
clear!

We believe that books are not in breach because books carry the
name of their publisher in the front and a copyright notice etc.
The book makes a statement about its legal statement that is implicitly
endorsed by the seller.

However, if you just pluck something off from some random page on the
web where the delivery channel does not provide any contextual
assurance or guarantee of the legal status of the material, then
by default, one must assume that copyright applies and that no copyright
licence has been granted.

I am not a lawyer either, but I would guess that one could invoke
arguments analogous to the laws covering the passing of stolen goods.
If someone offers you a Persian carpet and you buy it, you are innocent
of any crime, but if you buy it with suspicions of, or in the face of
demonstrable proof that, it is stolen, then you are guilty of receiving
stolen goods.

So my answer is that by default, you can't pluck something off the
net and by that act absolve oneself from the responsibility to honour
copyright law. By default, all material is copyrighted and by default
you can't use it. Therefore there is a burden for those obtaining
material to ensure that the material is legal.

A case: Supposing net bandwidth were faster than it is now and I spammed
a copy of Microsoft Windows to everyone on the planet. Clearly both
you and I would agree that I am in breach. However, according to you,
the recipients would be legally able to install and run the copy of
Microsoft Windows emailed them, and no one could stop them.


>(As a side note, some have put
>forward that while holding information inside a computer, copies may be
>automatically made for all kinds of purposes (buffering, caching, display,
>(de)compression, extraction, execution, you name it). One must assume however
>that such copies, which cannot lead to further distribution, do not require
>permission from the copyright holder, or again it would become all but
>impossible to safely operate a computer.)

My understanding is that copyright law allows these as a practical
interpretation of the law. Not relevant.


>Another point: what makes you believe you can unilaterally declare that "by
>extracting ... you automatically enter into a contract ...". Suppose I opened
>one the abovementioned books and found on page 163: "by opening this book you
>automatically enter into a contract which obliges you to send a cheque of $100
>every month to the author of this book and say a prayer for Bill Gates every
>Sunday". I am not a legal expert, but my limited experience with entering into
>legal contracts strongly suggests that this requires at least
>
>(1) That all parties entering into the contract be aware of the fact that they
>    do so, and of the existence and the identity of the other parties in the
>    contract, and of which text constitutes the body of the contract.
>
>(2) That all parties actively agree to the conditions of the contract, for
>    proof of which they place signatures on copies of the text of the
>    contract, so that each party obtains at least one copy that is signed by
>    all parties.
>
>(the measure employed to ensure these points are often quite elaborate,
>involving for instance marking by hand every single page of the contract; I
>must assume there are good legal reasons for such precautions.)
>I imagine that parties that wish to regularly enter into contracts with each
>other may agree to an alleviated procedure; however such contracts would be
>binding only because of this former agreement, which itself is presumably
>closed as a legal contract by the usual procedures.

I think you are dead wrong that a contract can't be created unless it is
signed. The law allows contracts of all kinds to be made in all kinds
of ways. A very good example is when you enter a restaurant. If you go in
and look at the menu and order a meal and eat it, then you have implicitly
formed a contract of sale of food/service with the owner of the restaurant.

Similarly, commercial contracts may be formed by a handshake and I think
there can even be a verbal contract, as, for example, when one bids
at an auction.

Formal paper contracts are used, and signed and sealed so as to increase
the reliability of proof of a contract and are considered PRACTICALLY
mandatory at certain levels of business such as $100M deals. But
my understanding is that you can still do a formal $100M deal with
a handshake. You'll just have trouble proving it in court.

However, I think that you have a very strong point in regard to awareness
of the existence of a contract being made. Your example of opening a book
to discover that one has made a contract is a good one, and I think I
need to modify the clause so there is some kind of opt-out option. I
should have thought this through more thoroughly. Part of the problem
is that I've been chewing over these issues with very little feedback
from anyone for six months, and I'm at the stage of just wanting to
ship the licence, so I'm getting a little sloppy.



>The GPL states: "You are not required to accept this License, since you have
>not signed it. However, nothing else grants you permission to modify or
>distribute the Program or its derivative works."

This is, perhaps, an attempt by the GPL to assert that a signature is
required to form a contract, something that I believe is incorrect.


>Similarly, nobody can be
>required to accept the accept the Free World Licence unless they signed it, or
>they perform some act that is normally forbidden, but for which permission was
>given under condition of acceptance of the FWL.

I do not believe that the act of performing a normally-forbidden act
invokes a contract by default. I believe that it only puts the actor
in the copyright-poo. To accept the contract, the actor must RESORT to
it to get themselves out of the copyright-poo, or performs some
action stated in the contract that the contract, of which they should be
aware, states is an act that indicates acceptance of the contract.


>Copying and distributing works
>copyrighted by others is normally forbidden by copyright law, so the GPL is
>justified in making its acceptance a precondition to granting permission for
>these acts (I have some doubts about modifying without copying or distributing
>though).

The contract part of the GPL is not invoked if you modify without
copying or distribution.


>By contrast, the FWL clearly wants to reach beyond the authority
>provided by copyright law, which is why it wants to invoke a contract in the
>first place; and it _must_ do so because it wants to control ordinary users of
>the software (to prevent them from executing it on certain platforms), who

Correct.

>unlike the distributors do not need to consider copyright issues. But merely
>extracting, installing, browsing or executing a legally obtained copy of data
>is not forbidden in any way.

The person providing the copy must conform to copyright law, but there
is a burden on the person receiving it to have good grounds to believe
that it is legal, otherwise they are doing something analogous to
"receiving stolen goods".


>If I bought a cookbook, I am certainly free to
>(without explicit permission) cook and eat any of its recipes, or variations
>of them, or to do whatever else pleases me with the book (except copying and
>distributing the recipes); I may even legally eat the recipes themselves if I
>like. Yes, I know that some well known software companies like to make their
>customers believe that they enter into a contract by certain actions, but
>unless those customers signed an agreement to this effect prior to obtaining
>the software, these are just scare tactics with no legal basis. I hope it was
>not your wish to join these companies in their dubious practices when
>formulating the FWL.

The most recent test case legally validated shrink/click wrap. Don't have the
reference handy, but can dig it up.


>In short, no action I perform on data legally obtained from a public server on
>the Internet (except redistribution) will legally bind me in any way.

I disagree. If you receive something, there is a burden for you to
establish some grounds for believing that it is legal. Hence there
is a burden for you to look for a licence, notice, distributor's
trademark etc. If you find a licence and it says "by performing action
X, you agree to this contract", and action X is something that's not
compulsory (like downloading the stuff or breathing :-) then I believe
that by performing the action, you invoke the contract.



>This means that what you want is fundamentally incompatible with the Internet
>model of distribution. If you want to ensure that non-paying users do not
>compile and execute code on non-free platforms, you can require them to sign a
>contract with you before giving them access to the code; you could detail
>whatever you like in the contract. Probably this would make the "free"
>distribution more limited than you intend, though. Theoretically you could
>allow redistribution of you software by third parties provided they ensure
>that every recipient has signed such a contract (with you) prior to obtaining
>the code (because distributors need your permission by copyright law); but
>then of course no distributor could accept such conditions.

See arguments above.


> > I have changed the note to:
> > 
> > Note      : Because this licence permits use on free
> >             platforms only, it does not satisfy the
> >             requirements for use of the US certification
> >             mark "open source" as defined by the mark's
> >             registrants in the web site www.opensource.org.
> >             Consequently, this licence should not be
> >             referred to as an "open source" licence and
> >             software released under this licence should not
> >             be referred to as "open source". However, you
> >             can refer to such software as "free software".
> > 
> > which I hope is less offensive. I'm assuming it was the
> > "you are free to..." part that was offensive. If it's the
> > idea of referring to it as free software that's offensive then
> > let me know.
>
>Yes, I must admit finding the idea of calling software supplied with such a
>booby-trap scare tactics Licence "free" slightly offensive.

I'm going to rethink the trigger condition. I don't want it to be
like a booby trap.


>But as has been
>said, the word "free" is not reserved in any way. If it were called the
>Free Prison Licence (referring to its freedom within strict confines), I think
>I would be less offended though. But then you might be offended more...
>How about the Free Gates Bill?

I believe that my Free World Licence is free enough to be called free.

If all software were free, then my licence would be roughly functionally
equivalent to the GPL.



><disclaimer> I am not a lawyer; I am just trying to apply common sense, which
>I realise may be dangerous in legal matters </disclaimer>

Well, I'm not a lawyer either, so I'm going running back to mine now.

Back soon with a rewrite!

Thanks, 


Ross.

Dr Ross N. Williams (ross@rocksoft.com), +61 8 8232-6262 (fax-6264).
Director, Rocksoft Pty Ltd, Adelaide, Australia: http://www.rocksoft.com/ 
Protect your files with Veracity data integrity: http://www.veracity.com/