[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Alternative trigger condition.

Ross N. Williams" <ross@rocksoft.com> wrote:
> At 1:33 PM +0100 11/1/2000, Marc van Leeuwen wrote:
> >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.

I suppose you mean "not copied in violation of copyright" here. There is
nothing wrong as such with receiving or possessing copyrighted material (think
of the books example).

> 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.

You'd be in violation of something, yes. Taking undue advantage of an illegal
act perpetrated by somebody else, or some such. Not breaching copyright
yourself, I'd say, unless you are distributing further copies you made.

> 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.

Being the default state of a work would mean IMHO that one may not deduce the
absence of copyright from the absence of a copyright notice. I hope it does
not imply the burden you suggest, or buying a newspaper or browsing the WWW
would be very perilous and/or laborious undertakings indeed!

> 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.

I don't like that "By default one is in breach". One is not "guilty until
proving oneself innocent". By default one may not copy and distribute anything
one strays across without certifying that this does not breach any copyright,
which means one should be careful about such things; abiding to that however,
one is not thrown to the wrong side of the law with the obligation to find
ones way back to the right side.

As for the photo example; since the mechanics of the web has already made a
copy of the photo when it appears in my browser, I do believe I would be
entitled to store that image for private use; in fact I sometimes do such
things. I may not place that photo on my own web pages though (linking to the
original would be OK, except maybe if I have grounds to believe it was placed
there illegally).

The example becomes essentially different if the photo is a physical one that
I stray across on your desk. If I make a copy and replace the photo, I may
assume to be violating copyright; if I merely take the photo away, I would not
(but I would be in illegal possession of the photo).

> 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.

Actually, I think I believe my books are not in breach because they look like
they have been produced by a professional printer, who is not likely to
operate illegally. If a book looks like it has been produced on a photocopy
machine, I have good reason to doubt the legality of the copy, regardless of
the presence of copyright notices (which could have been faithfully

> 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.

To me, that is. It is not unreasonable to assume that the author of the web
page is in the clear (barring obvious evidence of the contrary), either by
owning the copyright himself or by having permission of the copyright owner.
In the latter case I may not of course assume that the permission extends to
me as well.

> 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.

In any case I would not IMHO be violating Microsoft's copyright. I guess it
would fall under "taking undue advantage of an illegal act". Besides it would
be pretty stupid because the sender (not you personally of course) would more
than likely have placed viruses all over the code (some even consider Windows
without modifications to _be_ a virus in the guise of an OS:-).

> >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.

Good points (I hadn't realised an ordinary sale constitutes a contract, but I
suppose you are right). Thinking of it a bit though, it seems that there are
globally two cases: public ones (like the restaurant) and private ones (like
the auction). In the public case, the conditions that determine what makes a
contract come into existence are fixed (maybe even by law) and supposed to be
known by everybody; the restaurant owner for instance cannot stipulate that
merely making certain gestures constitutes making a bid for some item (as it
could in an auction), not even by if this were clearly indicated on the menu
card. By contrast I suppose attendance to an auction will require some form of
admission by the organiser of the auction, and that admission will be given
formally under condition of acceptance of the "rules of the house"; those
rules could control almost anything of the organisers liking, including any
definition of what constitutes a legally binding bid.

> 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 fundamental problem with the Free World Licence seems to me to be the
timing. To modify the book example, suppose the book says instead, prominently
on page 1: You do not possess this copy of this book, rather the author has
given you a licence to keep this copy indefinitely on your bookshelf. If
however you choose to open the book to any of the pages 2 to the end, you
thereby agree to accept the terms of the further licence that now follows: ...
If I acquired this book by an ordinary purchase, or maybe as a free copy,
there is no reason I should believe that notice truthfully describes my
rights. It is not that I object to the idea that some of the books on my shelf
are not my legal property; indeed some of them are not. However, in all cases
this is a consequence of the manner in which I acquired them (e.g., by
borrowing or lending them), not of their contents. Similarly, if I download
files from a Bona Fide public distributor on the Internet (who I may assume
has verified having the necessary copyright permissions) then I am entitled to
assume that I own those copies of the data, and can do with them to my liking
(barring redistribution, as always). No message contained within those files
can alter this state of affairs (messages that would make my possession of the
files illegal are excluded by my hypotheses).

If you want to subject my handling of your files to the conditions of some
licence, the you may very well do that, but you must make sure that I accept
the conditions _prior_ to acquiring your files. You may for instance
distribute the files from a password-protected web site, and make sure to hand
out the password only after my acceptance. In that cases I may clearly not
consider copies of the files so obtained to be my property. The only thing you
could then include in a public distribution are instructions how to proceed,
and maybe documentation not subject to the licence.

> >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.

Yes, but replacing "signed" by the more general "indicated acceptance" would
not make the sentence very clear. It is not accurate anyway, because not being
a contract, it is not really an option to sign the GPL in the usual sense.

> >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.

That is not what I wanted to say. But if an act is normally forbidden, and
some person has to legal power to grant permission (as a copyright owner has),
then that person may attach any preconditions of his liking to granting that
permission. Which is what the GPL does.

> 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.

I'm confused. It's useless for a contract to provide conditions for its own
acceptance, since it is not valid until it has been accepted, and then there
is nothing left to condition. Concerning the copyright-poo I disagree, see
"guilty until proving oneself innocent" above.

> ... 
> >If I bought a cookbook, I am certainly free to
> >(without explicit permission) cook and eat any of its recipes ...
> >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.

Hmmm, I recall just having read the contrary. That was on the WWW though
(http://cr.yp.to/softwarelaw.html). I would be surprised if your statement
holds for software acquired by ordinary purchase in a store with no prior
conditions being agreed upon (for reasons given above).

> >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. 

I cannot believe this. Nobody does this. Do you realise how many copyright
notices are included in even a modest Debian GNU/Linux installation? And the
Debian people, noted for their legal scrupulousness, don't even attempt to
flash those notices on the screen while installing (which admittedly would be
pretty useless anyway, but at least might give you the feeling you missed
something that you might want to catch up on).

> 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.

I certainly hope not to find such notices in my Debian GNU/Linux installation
(it would at the least lower my opinion of Debian), but if I, do I would
probably ignore it, or possibly remove the package if I do not care enough
about it anyway. In the case of Netscape, which I do choose to use for lack of
a sufficiently convenient free alternative, I arrange that my left mouse
button gets depressed while the cursor is over the word "Accept" while I am
looking the other way (this does make me feel a little bit guilty though). If
new releases of Netscape remain sufficiently frequent I'll figure out what
makes Netscape believe it has shown the message already. For the record, if
similar messages show up while I'm looking for stuff on the Internet (not
already on my hard drive) I do read them, or (more often) decide to abandon.

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

Despite all the care you have taken to make the conditions as reasonable as
possible, it tries to control the actions of users, not just of distributors.
And while it only tries to prevent me from doing something I have no desire to
do in the first place (run under Windows or similar) this would irritate me.

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

But it is still a licence whose functioning is made dependent on other
software and licences, which is its greatest practical drawback. The licence
enigmas discussed in Debian-legal almost invariably have to do with such

Marc van Leeuwen

Reply to: