Questions on GPL for billwg

I laid out an easy business model for profiting from the HL2 source
code leak, *if* he's correct about copyright. He hasn't responded at
all:

slrnd7hoa3....@dmc22317.local
"Now, game engines, particularly high-profile ones like the Doom3 and
Source engines, license for at least six figures. (The precise details
are only offered under NDA, so I can't give you a better figure.) You
can take the source code from Valve, reorder the functions and do a
search-and-replace on the variables... you can offer to sell it (not
just license it, but *sell* the rights) for some five-figure sum, and
indemnify your customers from suits from Valve."

(He can also look at slrnd5tc6m....@dmc22317.local and
slrnd7etk0....@dmc22317.local.)

Now, he's also claimed that the GPL may not be enforcible in part or
in whole. I'd like to know why.

slrnd7ff9o....@dmc22317.local:
"Please outline the legal theory you use to reach the remarkable
conclusion that, to directly quote you, 'section 2 of the GPL can
be considered non-binding'."

(Also, any of these: slrnd7cfcu....@dmc22317.local,
slrnd7hgg0....@dmc22317.local, and
slrnd6i95d....@dmc22317.local)

To view these articles, you can (among other things) go to the
"Advanced Groups Search" at groups.google.com, and enter the
above message IDs into the "Lookup the message with message ID"
field.

--
Sincerely,

Ray Ingles (313) 227-2317

"The growing and dangerous intrusion of this new technology [threatens
our] economic vitality and future security. [It] is to the American
film producer... as the Boston Strangler is to the woman alone."
- Jack Valenti, in 1982, regarding the advent of the VCR.

I had only said that I didn't see that as a good business opportunity.
Now I was coming at it from the notion of using it to create a game and
my direct experience has been that artwork and storyline are much more
important than software technology in that regard. Now I see where your
premise is that I could take the copied material and sell it to other
would be game makers for a price that would be so attractive that they
would beat a path to my door, eager to save the money that would be
otherwise given to a higher price vendor.

Googling a bit shows the availability of a plethora of game engines,
many similar to that of HL2 and some of them OSS. Microsoft is even in
the space with their DirectX APIs that seem to offer many of the same
functions and effects. Companies already making a game have already
selected and acquired engine software and are not in the market for
another, given the apparent huge cost of changing engines.

I don't think that it is a likely place to invest the substantial time
and energy necessary to make a better decision.

You might try searching for your name (billwg). This way you can see
not only the articles you've authored, but also the replies.

It's nice to see that you are getting some value of of some OSS
software.

Keep in mind that AT&T used to license the source code for the UNIX
kernel for about $75,000. BSD licensed their source code for $25,000.
This was the original BSD 4.2 and System V kernels. This represents
less than 1% of the source code included with a typical Linux
distribution.

And yet there are many businesses that use Linux as a strategic element
of their business - creating business opportunities. There are Linux
distributors, including a number of very high profile distributors,
three of which (Red Hat, Novell/SuSE, and Mandrake), are publicly held.
There is also Canopy Group, which was the parent company for Caldera,
TrollTech, and a few other commercial enterprises.

Games are a very small niche of the overall economy. Ironically, games
and antivirus/anti-spam/anti-spyware are about the only profitable
segments of the software industry. Games are popular, not because of
the great software technology, but because there is a good deal of
creative effort. Most modern video games are virtual Hollywood
productions. Many even leverage heavily advertized movies or
television shows.

Talk about a niche market. Selling a game engine to a game developer?
How many customers really have the money to pay anything substantial?
Perhaps, if the game engine vendor was willing to accept a percentage
of the revenues, probably only about 1/10th of one percent, 1/2 a
percent at best. Even so, developing a good video game requires
authors, technical production, musical composition, musical
performances, graphic artists, then you have to finance the flooring -
essentially you have to pay for the software, ship it to dealers, rent
the shelf space, and provide a guarantee, essentially a minimum
percentage of the inventory to be sold, often as much as 80% of the
initial shipment. The cost to get the software on the shelves can
exceed $5 million dollars. Then you have to advertize the product, on
television to show the animation and video quality. All of this before
you've even sold your first copy of the actual game.

Most vendors, such as Electronic Arts, have a game engine which is
independent of the platform on which it runs. Most make calls to
OpenGL or DirectX and there Linux version:

<blockquote>
> Dear List,
>
> is there a similar library for graphics programming like DirectDraw
or
> DirectX on windows-systems for Linux (not OpenGL)?

yes, there is SDL: simple direct-media layer. try www.lokigames.com (or
similar), or search for SDL (maybe it's listed on freshmeat ?).
</blockquote>

Normally, the game makers have their own animation engines which are
engines on top of the 3D APIs. Silicon Graphics, who has been a
pioneer in the ultimate Video Game technology - military simulations,
developed OpenGL, and VRML, there are several good Open Source tools
such as Glade. The big studios, such as dreamworks and IML have their
own engines, many of which rely on clusters of Linux machines as
rendering engines.

Generally game software for PCs is less detailed than the studios'
renderings.
There are so many pieces. In some cases, characters will be "sprites",
animations superimposed over a background animation. This reduces the
work of the 3D rendering.

All of this for a profit of only $1-2 per box by the time you are
finished with all of the expenses and production.

Not only that, but most companies have their own advanced engines.
Even the best rendering engines would be of little value without the
artistic talent to create the plot, costumes, scenery, music, dialogue,
and effects that transform some interesting 3D rendering capability
into an imaginary world that never existed, where characters do
fascinating things.

Actually, this is the best illustration of what makes Open Source so
attractive. There are some very generic genres such as Word Processor
or Spreadsheet, which are general enough to be profitable as
stand-alone products.

About 95% of all of the software used worldwide, including servers, is
a combination of some core toolkits, and lots of custom developed code.
Even advanced applications such as CRM, ERP, and Workflow packages are
only components which need to be part of a much bigger overall
architecture.

Open Source technologies can be very useful or providing the "glue" to
pull many different components together. It can also be used to
function as components where appropriate. For example, one could use
WebSphere where financial transactions are needed, but can also use
TomCat,Jakarta, Struts, and JBOSS for those sections which are more
informational and less "mission critical".

--
Sincerely,

Ray Ingles sorce...@hotmail.com

"These modern kids don't know the simple joy of saving four bytes of
page-0 memory on a 6502 box." - isomeme

It's a small market, but it can be a lucrative one. For example, iD
software makes a lot of money from licensing the engines they create.
Virtually every engine they've made has been used for other games (e.g.
the orignal DOOM, all three Quakes, and now Doom 3). Indeed, some people
feel that the actual games are more tech demos for the engine.

Now, Valve has *definitely* been grooming their Source engine (the one
behind Half-Life 2) for similar licensing. A bit of digging at various
game sites will show this. The two engines considered the most advanced
are currently iD's Doom3 engine and Valve's Source.

Indeed, but none of them get anything on the screen without the engine.
And yes, engines are licenced all the time - getting someting that can
render pretty graphics at real-time framerates is a major technical
challenge and it takes very very smart people with experience to do it
well. It can be (and usually is) much cheaper to just licence an existing
engine and tweak it rather than hire the kind of people it takes (and
spend the time it takes) to make a new one from scratch.

Even Valve's previous king of the hill, Half Life, was based on a
Quake engine, though heavily modified.

Being able to undersell a competitor by an order of magnitude would
be a huge advantage, and the Source engine is one of two the most
highly-regarded engines in the PC game area. Someone could make a
serious pile of money with the leaked source code for Source... if
there weren't that pesky copyright thing in the way.

But billwg says it's no problem. But still, he doesn't jump on this
opportunity. Puzzling...

> Googling a bit shows the availability of a plethora of game engines,

http://pc.ign.com/articles/540/540725p1.html
"The Source Engine matters to gamers everywhere, even if you don't plan
on buying Half-Life 2... chances are there's a game coming down the pike
that'll be built on its technology."

http://www.hlfallout.net/viewnews.php/7456/
"VALVe announcend a Source engine licensing partnership with Smiling
Gator Productions recently. Already in development, Twilight War:
After the Fall is a Source-based Massive Multiplayer Roleplaying Game
(MMRPG) for the PC."

http://www.extremetech.com/article2/0,1558,1640262,00.asp
"an extremely advanced game engine that has already been licensed by
several game developers. In fact, the first third-party game to use
the Source engine is Vampire: Bloodlines, which should be released
"this fall.

(Oh, and BTW, Direct3D and OpenGL are, essentially, interfaces to the
3D hardware, *not* game engines. Game engines *use* these interfaces.
DirectX as a whole is designed to give simple access to the keyboard,
mouse, sound, etc. as well. Linux has the SDL to do the same.)

<sarcasm>And no more games are planned, ever again, so you've missed
the window to sell anything.</sarcasm>

Do just a bit more research. The two top engines today are the Doom3
engine from iD software and Valve's Source. Don't just take my word
for it, ask around. Ask any snotnosed gamer punk you know.

But, of course, you won't do so because you *know* you would never
get away with it. Copyright law doesn't work the way you claim it
does.

It is not the copyright laws stopping the sales, it is a reluctance to
purchase that is stopping the buying. To sell one of these things you
have to have a prominence and a history, else the customer is going to
suspect, rightfully so, that you are a quick hit artist and will be gone
with the sunrise.

"Game engines" are wrappers on the APIs or else graphics functions
proprietary to the engine maker. They are for game creators who want to
write scripts rather than directly coding to the APIs.

Well, Ray, I don't totally remember what this was all about, but the
gist of it was a fair question as to whether or not an idea expressed in
a computer language that is essentially just a mechanism for executing
the idea can ever rise to the "unique expression" required for
protection under the copyright laws. If you examine any single method
or subroutine or call or whatever unit you want to select, and then you
define the functionality of the piece as the idea expressed by it, you
are left with the source code being in the category of obvious
expression rather than unique expression.

You sneer and make up extreme cases to support your sneer, but you
refuse to answer a fair question. The ideas are very important to
creating effective software and the organization of the increments of
ideas is the architecture that makes the final form more efficient than
some other arrangement, but in the end it is still just the idea and the
idea can be expressed in other ways or else the idea can be expressed in
the same way if the language is sufficiently strict to allow no other
practical construction.

If the idea is to iterate over a collection of objects and test them for
some condition and take action based on the tested condition, for
example, you can use several different loop methods, but none are really
unique and many, many others will certainly have used an equivalent
expression to do what the idea requires. I think that you can do that
repeatedly on each element of a program until the totality of the
program is covered and you have re-stated the ideas with a modified or
even essentially the same expression. Code is not poetry and is not
even prose. You cannot copyright a math expression and computer code is
pretty much the same thing, just a lot more of it.

Ok, fine, here's what you said in the post with message ID
JSzde.25312$716....@tornado.tampabay.rr.com:

>>One theory is that the GPL is a separable contract, i.e. it is multiple
>>contracts that allow you to take the one and reject the others. For
>>example it says:

(You go on to say that you don't have a problem with Section 1 of the GPL.)

>> My interest is in whether or not section 2 of the GPL can be considered
>>non-binding. I think that it can be under some circumstances. There
>>has never been a court test or even a settlement AFAIK where that clause
>>was being asserted.

I don't understand this. You didn't specify *what* circumstances would
render Section 2 of the GPL non-binding. I also wonder - assuming that
Section 2 of the GPL *didn't* apply to someone, would they somehow have
the right to, to quote the GPL, "modify your copy or copies of the
Program or any portion of it, thus forming a work based on the Program,
and copy and distribute such modifications or work..."? What would be
the legal justification for evading normal copyright restrictions?

Even *if* the GPL *were* a separable contract, and you didn't accept
Section 2, what else would give you the right to engage in the behavior
permitted under Section 2?

The notion that it is an illegal contract that fails on the issue of
consideration. The argument would be that my extension of the program's
functionality has a value greatly in excess of the value to the original
author in being allowed to use my extension in the future. I could even
extend to the original author specific license to use my extension, but
not to anyone else. Since the issue is so imbalanced, I ask the court
to find it non-binding.

The phrase "thus forming a work based on the Program" is a leap by the
author in making an unwarranted assumption that the work is derivative.
In my scenario the work is an extension that adds to the original as
an aggregation and leaves the original unmodified in any manner of
interpretation of it as an expression, unique or not.

Well that is patently simple. I have the right to redistribute the
original code per section 1 because I provide the license and source
with it. I can't be violating the author's copyright with my own work,
which would be the part that I distribute without source or further GPL.

Both Congress and the courts have already decided that it does. There
are some special considerations when dealing with computer code, but
no one (except you and, allegedly, one golf buddy of yours) thinks that
computer code isn't covered by copyright. You haven't produced any cites
to justify that.

We've already been over this. For example, here:
http://groups-beta.google.com/group/comp.os.linux.advocacy/msg/99e6b645112e732b

You proposed a simple problem - generating the standard deviation. In
addition to variations like data structures (array? linked list? vector?)
I pointed out that there are at least four reasonable algorithms to choose
from. Even in a simple case like a mathematically defined function, there
are variations in the idea expressed that would be unlikely to be duplicated
by, say, clean-room reimplementation.

Now, let me point out some other variations - how about comments? They are
by definition *not* code but are important in the practice of software
engineering. Are you arguing that *those* can be duplicated freely?

Except when I *have* answered it.

And your example of that, the standard deviation, has been shown to
admit a wide variety of reasonable implementations. Try again. You claim
to have been reviewing the Thunderbird codebase. Point out some code that
you feel would not be protected by copyright. Put up, if this is such a
common occurence it should be no problem to find an example or two in
such a large application.

This is one reason that clean-room design is used. In some areas (e.g. a
BIOS where the functions are relatively simple and the specifications
precise and the space considerations are at a premium) there may actually
be only one reasonable way to implement some parts. That is a hard case
to make, though, especially when you've had access to the original. So
companies go to the expense of clean-room design; if there really is only
one way to do it given the specifiaction, they will hit upon it *without*
copying from the original.

Say, there's a question for you! If clean-room design *isn't* really
needed, why are all these companies going to the considerable expense of
time and effort to do it? Are all lawyers but your golf buddy incompetent
and giving bad advice?

How is this different from clean-room design? Please specify how this
is different from the following idea:
http://groups-beta.google.com/group/comp.os.linux.advocacy/msg/6a2ce076ea90ec99

"A single note cannot be copyrighted. A song is just a collection of
notes, so a large collection of such notes, even when arranged identically
to the original song, can be copied just as any individual snippet."

You *can* copyright a math textbook. You *can* copyright a math paper.
Some individual portions of it may not be copyrightable, but the setting
they are in most definitely is. Again, let's get down to brass tacks.
Start pointing out code from Thunderbird that isn't covered by copyright.

You are using the code - the code itself is the consideration. You're
getting value from it, else why use it? If you don't like that, the
consideration is the waiving of claim to copyright infringment. I've
produced cites that both of these types of things have been held to be
valid considerations.

Can you find any cites of this being tried?

How about analagous cases from other fields? Surely your golf buddies
can provide you with a couple?

I can't find too much support for this kind of case. See, for example,
here:

http://en.wikipedia.org/wiki/Consideration

"For instance, courts will not inquire as to the adequacy of consideration.
If someone honestly dislikes their car and wants to sell it for fifty dollars,
the law will not consider this an invalid deal. However, the court will reject
"consideration" that was not truly bargained for. Occasionally the court may
refer to "adequate" or "valuable" consideration, but in reality the court is
not examining the adequacy of consideration, but whether or not it was
bargained for. Another term for this sort of non-bargained-for payment is
nominal consideration. The traditional notion that courts won't look into the
adequacy of consideration, an ancient notion in the English common law, doesn't
square with the benefit-detriment theory (in which courts are implicitly
analyzing if the parties are receiving a sufficient benefit) but does square
with the bargain theory (in which only the subjective intentions of the parties
are considered).

Y'know, I've been digging through the previous posts, and I can't find
anywhere you've said that.

In this case, it is *possible* that you might be right. There is some
dispute about whether dynamic linking to GPL'd code does, in fact,
represent derivation. At worst, though, this would make the GPL
equivalent to the LGPL, it would not *invalidate* the GPL. Modifying
the actual code covered by the license and distributing it would still
land you firmly in Section 2 territory.

If you had stated this before, we might have saved some time. Of course,
you've been also claiming that source code can't be covered by copyright,
so I have a hard time believing that you've indended this meaning all
along.

The idea content of the source is explicitly not protected whereas the
"expression" would be if it is deemed sufficiently unique. I think that
it can be argued that the source code is essentially nothing more than
the statement of the idea of the program and may be used liberally, if
not verbatim.

The typical engineer is not going to be satisfied with anything but an
improved version anyway, so the source code for some program, if it can
be obtained in some legal way, can certainly be used as model to
understand the detailed idea behind the program. That has been upheld
by the courts in the very cite you provided.

Where the source code is functioning as if it were a mathematical
description of the idea, such as when implementing an algorithm or
parsing a file format, I don't think that there is any protection of the
exact source statements provided by the copyright laws.

The copyright law is for unique expression. That applies readily to
prose and poetry where there is an element of artistry in the rendering.
It doesn't apply to telephone books and it doesn't apply to
mathematics. It doesn't apply to the ordering of subroutines, either,
based on your CA cite.

Comments are expressions of the idea. To the extent that they uniquely
express the idea, they are probably copyrightable in themselves, but if
they are in the source you are using legitimately, you have the right to
read them. Any compiled code resulting from the commented source would
not have comments, eh? So a copyrighted comment that is not copied is
immaterial.

But are any of them "unique expressions" within the meaning of the law?

I'll do that.

I don't think that "clean room" is a legal requirement or even
definition. There are reasons outside of copyright to perhaps be able
to prove that some "tainted" individual, such as someone under an NDA or
trade secrets condition, be kept from offering information regarding a
competitor's product, but where the code is legitimately available, such
as OSS, it can be used as a guide to determine the idea content of the
program which, given the general directness of a computer language, is
tantamount to the idea itself and so is not protected.

All what companies? I have been working for almost 10 years now for a
multi billion dollar software company and I have never seen anything
remotely resembling a "clean room". Who has one outside of the claims
of the CA case?

The song isn't copyrighted, Ray, the recording of the song, fixed in a
media, is copyrighted. Even the sheet music, printed on paper, is
copyrighted because it is a unique expression, fixed in a media. If I
sing my rendition of Hoyt Axton's "Jerimiah Was A Bullfrog" in the
shower and record it, can I sell copies to the outre music fans who
admire my work without paying any royalty to anyone? Some say yes and
some say no. Jib-Jab won a similar suit, IIRC, brought by Arlo Guthrie
for their rip of "This land is my land" during the last election.

You can probably copyright the whole book, but I don't think you can
copyright the set of equations that might describe the proof of some
theorem.

I'll get some code from Thunderbird and discuss it, Ray, but there is no
court decision that says that any or all or none of it is copyright
protected, so you can't say yes or no either.

I haven't seen any instances of non-release ever being challenged by the
FSF. If you read the cases that have been worked over recently, there
is no apparent element of a GPL user's non-contribution of a derivative
development being challenged. Red Hat, for example, locks up their
binaries and claims to lock their source by including various
copyrightable elements with the GPL elements. You can get the clean
source, but then you have to create your own binaries. So just by
putting a trademark on a GPL program lets Red Hat claim a proprietary
product. Is that not a derivative using your definition?

One thing that comes to mind is the rescinding of obligations to pay
large sums of money for swampland sold to unsuspecting folk in the North
by Florida landowners in the 50s and 60s. The land was not
misrepresented, just overpriced by orders of magnitude.

I don't concede that, but what about where some GPL item, say
Ghostscript's viewer code, is added to some larger non-GPL program for
the sole purpose of viewing PDF report files produced by the non-GPL
program? If the GPL code is used, with attribution, does the vendor of
the larger work have to disclose the source code for the larger work if
the GPL code is compiled into the same executable as the rest? What if
the GPL code is compiled into something external, such as an OCX, that
is called by the larger work? These are the viral things warned against
by Gates and Ballmer.

These are separable arguments, Ray, and not dependent on one another.
The applicability of the GPL to copyright is pretty useless if there is
no copyright that applies, certainly, but the copyright of source could
be valid and the GPL deficient in some scenario.

http://www.co.marin.ca.us/depts/DA/main/dist/ConsumerPages/
contractformation.cfm

"- Illegal consideration: Consideration that is void for illegality
or any other reason.
- Past consideration: Acts or forbearances previously performed cannot
be consideration for a new promise.
- Existing legal duty: A promise to perform an existing legal duty.
- Legally-required duty: A promise to perform a legally required duty,
where the duty is imposed by a law of crimes or torts:
- A compromise of an invalid claim: If a claim is wholly invalid,
neither forbearance to sue nor a compromise can be valid consideration.
- A moral obligation: Performance of a moral obligation may not be
adequate consideration."

I fail to see how *any* of these would invalidate the consideration
offered by the GPL (use of the code, and the shield from copyright
infringement claims).

There are also "Unconscionable Contracts". Again, I don't see any
way to justify such a defense against the GPL.

That's, basically, an example of an "unconscionable contract" - with
terms that are "unreasonably, unacceptably, or unfairly harsh, and so
one-sided as to shock the conscience."

So far, when the GPL has come up in court, it has not been found to
be such. No one's even come close to trying that argument. Feel free
to try that argument sometime, I'd like to see it. Here's a case where
it might even have been considered:

http://www.groklaw.net/article.php?story=20050225223848129

It was settled before it got past discovery, but the SAE caved before
it got to trial. Pretty well par for the course with the GPL.

I would be very impressed if you can find *one* lawyer who would argue
that that does not constitute a derived work.

Umm, isn't that the case I just discussed above?

--
Sincerely,

Ray Ingles (313) 227-2317

"...brute-force attacks against 256-bit keys will be infeasable
until computers are built from something other than matter and
occupy something other than space." - Bruce Schnier

I don't see where the law on this had anything to do with the GPL much
less with section 2.

Well certainly the GPL work is not derived from and the GPL work is
unchanged insofar as its expression. The expression of the total work
does not derive from the GPL program expression either.

Well, given the way that programs execute, there is no real difference,
but you seem to have two opinions that are 180 degrees apart in
something that is a functionally equivalent circumstance. Seems artificial.

So copyright infringement is justifiable as long as you can make a
living off it.....

Nice to see your keeping the bottom of the barrel warm.

You've said this, but you have produced no one, except (allegedly) your
golf buddy, who agrees with you. No one.

Yes, but you can't just use the code verbatim. You have to rewrite it.
I have said this before and you ignored it:
slrnd6i95d....@dmc22317.local

...iff there is only one way to express the idea. There have been
cases where that's been true, but you are asserting that that's
*always* the case and yet you've never given any support for that and
have ignored concrete examples to the contrary. See below.

...until you *distribute copies of them*, which is part of the scenario
*you* proposed. Even *if* the source wasn't copyrightable (something I
don't concede for a moment) you would have to strip all the comments out
before distribution. Are you going to retroactively modify your scenario
to include this?

Heck, the DeCSS code was ruled to be speech protected by the First
Amendment.

I won't hold my breath.

Okay, let's test this. You propose a simple programming task - akin
to your standard deviation test, and a computer language to implement
it in. I prefer straight C, if that's okay with you. We'll each code it
up. If you're right, and there's only one way to express it, our files
should be identical modulo whitespace, comments, variable names, and
function ordering. To ensure that neither of us is cheating, we can
post the MD5sums first, then the actual files.

http://abbeyrd.best.vwh.net/mysweet.htm

George Harrison was sued, and lost, for using a similar though not
identical sequence of notes in a different song. You willfully
misrepresent the facts.

Funny, restaurants avoid singing "Happy Birthday" because they don't
want to pay royalties on it...

The song was ruled to have fallen into the public domain. You don't RC.

I think we've definitely gotten into the 'put up or shut up' area.
I don't think I'll be discussing this any more with you until you
either (a) point out some Thunderbird code for discussion or (b) accept
my coding challenge above.

--
Sincerely,

Ray Ingles (313) 227-2317

"If one is going to use standard units, one should do it
right and use Kelvins. That way nobody is happy."
- Linux Weekly News, on Fahrenheit vs. Celsius

Oh, Phil, try to pay closer attention to the thread! Copyright isn't an
issue here, just a balance of harms in a contract dispute.

If you *can't* make a profit without it, then it's *not* a small
contribution, eh? QED.

What will you say when the judge asks you, "Why did you not simply
buy, or license, or write code to do this functionality instead of
violating the license on this work?"

Try to use the same reasoning when you sell a collage made from
copyrighted photos you clipped out of a magazine.

Static linking and dynamic linking are different things. But somehow
I'm not surprised that you claim not to understand the difference.

--

Sincerely,

Ray Ingles (313) 227-2317

Trying to be happy is like trying to build a machine for which
the only specification is that it should run noiselessly.
- Anonymous

But the laws were written to take into account the macro effect of
things. Thomas Jefferson was even more ignorant of the differences
between static and dynamic linking than you say I am. Surely that isn't
an element of any legal analysis.

Using verbatim pieces of a work as part of a new work makes the result
a derived work. This applies to photographs, essays, poems, songs,
stories, screenplays, movies... and source code. There are a few
exceptions (de minimus, fair use, parody), none of which apply to the
scenarios you've laid out.

Thomas Jefferson lived before recording devices and yet we have laws
that differentiate between live performances and mechanical
reproductions. Shocking, I know, but technical reality *does*
occasionally intrude onto the law's turf.

--
Sincerely,

Ray Ingles (313) 227-2317

"When your [profit] margins are more sensitive to Bill Gates'
pricing whims than they are the price of oil, that's an
untenable position for a large company to be in."
-- John Chapman Sr., BP Amoco Technology Executive

Lower price *is* an advantage. I never said it was the only factor,
but don't you pretend it's irrelevant. An order of magnitude lower
price is a *major* advantage. According to some theories, it's the
only reason Linux survives... :->

Ah, but you can lunch off of the work Valve's already done in those
areas! Valve's already publicized their engine well, and people know
they have a good reputation. They *also* know that the Source engine
code was leaked, so they'll know you have what you say you have.

Now, everyone else on the planet (according to you) is under delusion
about how copyright relates to software. So, you can offer to indemnify
your customers against suits by Valve. You believe you'll win handily,
so that's no skin off your nose. That's *your* key advantage.

You can't beat Valve on support, of course, but you can point out
that with the money they save on license costs, they can hire a
whole team of developers to do any support and customizing they need.
(They'll do that anyway - most engines are modified to at least some
degree for any particular game.) And you won't be asking for royalties,
or new license fees for later games!

As I said, ask around. Source is one of the two most advanced engines
in the business right now and you can offer the real deal on excellent
terms, and fully legally (according to your minority legal theory)!

Actually, I've presumed that you are explicitly *saying* that's what
you are, but (a) you have a legitimately valuable product to offer, and
(b) you are *selling* and not merely licensing the code, without
restrictions.

--
Sincerely,

Ray Ingles (313) 227-2317

"Windows is easy to learn, but hard to use. Linux is
hard to learn, but easy to use." - Anonymous

Go ahead and apply your "common sense" and mock the idea, Ray, but
you're missing out on a world of education outside the C-language and
unix APIs.

Obviously you cannot market something as "hot merchandise" in
conventional channels, Ray. People who are inclined to buy that kind of
thing probably already have a copy of Valve's code. If you cannot offer
any assistance, why get it from you? They can't get Valve to help, so
where are they going to go?

I think that is a participative advantage rather than a comparative one,
Ray. Everyone who hopes to be in the software business is going to
offer indemnification. You can sell less without it, but you cannot
sell more with it.

You misinterpret a lot of things. Either I am very bad at explaining or
you are being deliberately misleading. I know what I think.

The US Code says:

"A “derivative work” is a work based upon one or more preexisting works,
such as a translation, musical arrangement, dramatization,
fictionalization, motion picture version, sound recording, art
reproduction, abridgment, condensation, or any other form in which a
work may be recast, transformed, or adapted. A work consisting of
editorial revisions, annotations, elaborations, or other modifications
which, as a whole, represent an original work of authorship, is a
“derivative work”. "
>

But incorporating another work into one's own as a sort of aggregation
is not listed as a derivative of the aggregated work. Certainly the
part aggregated can be considered as copied, but the GPL is insistent on
the category of "derivative work" in requiring a source disclosure of
the deriver's code.

I've got a sit down meeting with my friend next week. I've got him
doing some paid work for me on another matter. I'll see if I can get
some specific references to the whole thing.

If I have an application based on using the VBA facilities for MS Office
and I buy a copy of MS Office and incorporate the whole enchilada into
my product offering and pay MS their license fee, I am confident that I
do not have to worry about any "derivative work" claim by MS. I think
the same should hold true for anything that I obtain a valid license to
use and it's just real cheap to do that with OSS stuff since the license
to use it comes for nothing.

This is, to me, just using the original stuff for what it does and
adding my stuff to it for what it does. I can construct ways that this
is completely additive and I don't think that the courts care one way or
another about how the watch works, just whether or not it tells time.
That might matter in a patent case, but not for copyright where the
copyright fee has been paid. As I understand, the decisions hinge on
the financial impact at the end of the day.

>

What we have here is the taking of some source and using it to create a
new binary media form, not a copy of the original binary, but a new
binary entirely. There is some feeling that copyright protection might
apply, but how?

Well so what? You haven't produced much of anything either. I think
that it is necessary to protect source code as a trade secret rather
than relying on a copyright.

Does it make a difference if there are only two ways? Only three? Only
ten? Does it make a difference if I pick a different way than the one
the original program picked? Why? What is the "unique expression" here?

I was not going to distribute the source at all, remember? I was going
to keep it secret.

That was an idea, not an expression, Ray.

>>

I am in the process of deciphering a section of the Thunderbird. I have
a section I picked which, strangely enough, establishes a connection to
a COM object that is used for parsing message strings and then runs an
input string through the COM method to get a converted output. This
call, "static nsresult ConvertBufToPlainText(nsString &aConBuf)" is
built with layers of macros that unravel to rather plain jane code. The
macros only serve to obfuscate the process, IMO, so my "conversion" is
to simply de-macro the subroutine to expose the conventional API
connections and use.

The code itself runs to a lot of lines, though. Where would you like it
sent?

What, you mean compiling source? Copyright covers translations, and
I have a hard time believing you could convince a judge that a compiled
binary *wasn't* a translation of the source code...

I've produced plenty of cites, you haven't produced any. You are
going against what appears to be the entirety of the legal profession
here. This link, for example, took less than 10 seconds of Googling:
http://www.ipr-helpdesk.org/docs/docs.EN/
softwareCopyrightAndTheComputerProgrammer.html

In the case of copyright, *yes* it makes a difference if there is more
than one way to express something. The doctrine of "merger" requires,
very explicitly, that there be *only one way* of expressing the idea.
If there is *more* than one way, then merger doesn't apply.

Now, a restricted number of ways *might* have a practical effect in a
courtroom, I will grant. In the absence of other evidence of infringement,
the fact that you picked one of only two ways to express a programming
idea might not be considered to be a preponderance of the evidence. (I
will note that, should someone bring a copyright claim against you,
Bill Weisgerber, specificially, some of your posts here might be usable
against you... :-> )

Well, there are at least two distinct scenarios here. One involves
distributing GPL code unmodified but linking to it in ways not condoned
by the GPL. (See, e.g., NVidia drivers for ways around this.) Another
involved stealing GPL code outright.

As noted above, 'translations' are still covered by the copyright on the
original work...

How many lines? Too many to post here?

--

Sincerely,

Ray Ingles (313) 227-2317

Microsoft Windows - Simplicity made complex.

Note the words "such as". This is not an exhaustive list. Additionally,
what about the words "recast, transformed, or adapted"? Ask your golf
buddy if collages are considered derivative works or not. I doubt if
even he would say they weren't.

I'm not 100% sure there... MS has clamped down on people trying to sell
their copies of MS software, even unopened in the box. You might have to
negotiate with them specifically for the right to redistribute. (Remember,
you don't *own* MS products... you merely *license* them.)

*Not* nothing. You don't have to pay money, but you *do* have to follow
the terms, which include conditions on redistribution.

As I have said before, dynamic linking *might* be legal. That's the
only part of the GPL I can see as even slightly questionable. Otherwise,
you have to abide by the restrictions of the license to use the code.

--
Sincerely,

Ray Ingles (313) 227-2317

"The idea that an arbitrary naive human should be able to
properly use a given tool without training or understanding
is even more wrong for computing than it is for other tools
(e.g. automobiles, airplanes, guns, power saws)." - Doug Gwyn

Some versions of the license, where it is offered at a deeply discounted
price in exchange for a limited use, that is true. But a retail version in
the box is freely re-sold.

Heh, even on ZDNet he LOL!'s a fair bit....

What if the color blue were actually mozarella cheese?

Your question is ill-posed and nonsensical. Carburetors are not subject
to copyright. A blueprint *describing* a carburetor might be, of course.
In that case, yes, excerpting the description of a carburetor you pinched
from a Ford blueprint would make the blueprints for the engine for your
"Weisgerber Autos" vehilcle derivative works.

The cars made from those blueprints would not be derivative works,
though they might infringe patents. That's a separate issue.

--
Sincerely,

Ray Ingles (313) 227-2317

"Economies don't like step functions." - Dr. Leonard Bieman

Obviously the notion is that I am using a carburetor as a component of a
larger work, just as I might use some software product as a component in a
system. That does not make the system a derivative work of the component
and you would look pretty silly trying to claim that. The derivative work
must somehow substitute for the original and what I was hypothethizing
clearly does not do that. Your sneer only makes you look like you cannot
grasp the obvious.

Source code is a set of directions to a mechanism that creates object code
in a very deterministic way. The government says:

""WHAT IS NOT PROTECTED BY COPYRIGHT?
Several categories of material are generally not eligible for federal
copyright protection. These include among others:

a.. Works ...

b.. Titles...

c.. Ideas, procedures, methods, systems, processes, concepts, principles,
discoveries, or devices, as distinguished from a description, explanation,
or illustration"

which makes the source code somewhat of an idea or procedure or method for
creating object code. In a sense it is the idea that the object code
expresses. Perhaps it is not protected.

What might the legal definition of a translation be, Ray? If I translate
it back, I don't get the original source at all. Whereas if I translate a
book to another language and translate it back to the original language, I
will get something very much like the original. The reverse engineered
source from a binary can only be the idea behind the binary and would not
contain the original expression.

This cite is very low on any case law or statutory proof too, Ray. It is
a summary of the popular opinions, but not an example of a decision in a
relevant case. I have been unable to find anything where the issue was
squarely framed and a judgement made. Neither have you.

Well, that is a question, certainly. The resultant "translated" code
seems to simply call CoCreateInstance() with the objects GUID and then use
the objects methods to strip formatting from the text. That doesn't look
very unique in terms of expression when all is said and done. The object
itself doesn't seem like any innovative thing either. It is a
straighforward string replacement exercise.

Remind me again of the difference between you and a troll?

--
Sincerely,

Ray Ingles (313) 227-2317

"We're seeing crazy uptime numbers now, like three months, six months.
I fully expect we'll see a year of uptime when Windows .NET Server
2003 is finished," - Jeff Stucky, Microsoft.com operations team
(Unix systems regularly see multi-year uptimes.)

About >||< much.

--
With GPL the only thing Microsoft
gets for free is nightmares.
-- Jean Francois Martinez --

http://digital-law-online.info/cases/23PQ2D1241.htm

"While computer programs are not specifically listed as part of the above
statutory definition, the legislative history leaves no doubt that Congress
intended them to be considered literary works... The syllogism that follows
from the foregoing premises is a powerful one: if the non-literal structures
of literary works are protected by copyright; and if computer programs are
literary works, as we are told by the legislature; then the non-literal
structures of computer programs are protected by copyright... We have no
reservation in joining the company of those courts that have already
ascribed to this logic."

The decision then goes on to examine the *scope* of protection afforded
by copyright to programs, but I wonder exactly how you get 'perhaps it is
not protected' from the above clear endorsement that it is? Note that the
case was very specifically about copied *source code*.

You've seen this before, in the post with message ID
"slrnd69urv....@dmc22317.local". You replied to that post, though
you didn't address this quote.

Bounce a string of text back and forth through Babelfish a few times,
e.g. from English to German and back. For extra credit, do more than one
cycle. (Google for "fun with babelfish" if you're too lazy to try your
own example.) Similar results can obtain with human translators if you
use a different one for each direction. I used to work for a company
with both American and Japanese departments that faxed things back and
forth; I *assure* you I've seen examples.

Look up the game "Katamari Damacy" for the Playstation 2 sometime; the
invertability of the translation is irrelevant. Both source and output
can be copyrightable.

Say, since the MP3 codec is lossy, the resultant MP3 file can't be
reconstructed to the original waveform. By your logic, one of them
must be uncopyrightable. Which one, Bill?

The translation from a blueprint to a building is not necessarily
reversible, either. Nevertheless the blueprint is copyrightable and
some cities have claimed that, e.g., their skylines are copyrightable.

In languages like Java or Forth, your example isn't even technically
accurate.

Except when I have, as above. You haven't produced *any* cite agreeing
with your claims. Not one.

--
Sincerely,

Ray Ingles (313) 227-2317

Microsoft complaining about the source license used by Linux is like
the event horizon calling the kettle black. - Adam Barr

Buy the book 'On Bullshit'

billwg isn't lying, he just doesn't care, what matters is making people
believe him, the content is irrelevant.

We now know what Phil watches on Sunday night TV!

You deleted the last line of the cited paragraph that said "However, that
conclusion does not end our analysis. We must determine the scope of
copyright protection that extends to a computer program's non-literal
structure." The reasoning and dicta that followed for a number of pages
generally described the overall program structure as unprotected ideas in
the main.

So you say, Ray, but you don't do any more than say that. If the
de-translated words become totally different than the original and only
serve to convey the ideas and not the original expression, how is it that
the result is copyrighted? I don't know that it is and neither do you.
Nothing of the sort has ever been tried in court.

The blue print is copyright protected in that I cannot copy it and
distribute copies. I doubt that any skyline is copyrighted. Who owns the
copyright if it were? Is someone going to build a clone of New York in
Montana?

I'm not trying to prove anything, Ray, I'm merely saying that there are
issues that are not so black and white. You have nothing to show that they
are.

The rest of your post was equally nonsensical:

You then snipped the part where I answered your silly question with:

I can see where you may want to ignore having your premise so totally
dismantled. It's called denial.

Actually billwg, I saw him on The Daily Show, and bought the book then,
although I did see the 60 minutes excerpt, with the media so afraid of
the FCC they blurred the word shit repeatedly, and beeped it at every
utterance!

'Who's name we dare not say' <-- their own words....

--
"When the people are afraid of the government, that's tyranny. But when
the government is afraid of the people, that's liberty."
-- Thomas Jefferson

Funny you should say that, considering you had to resort to using an
argument about duplicating a PHYSICAL product, and try to blur the fact
that IT is not protected by copyright, and somehow, this 'proves' your
premise that software code [a literary work] is not protected?

Apples and Oranges indeed.

> Apples and Oranges indeed.

Might just be bullshit and whipped cream, Phil! Read your book.

Why not? Are you a derivative of your mother and father, or just your mother?

--
Reverend Paul Colquhoun, ULC. http://andor.dropbear.id.au/~paulcol
Asking for technical help in newsgroups? Read this first:
http://catb.org/~esr/faqs/smart-questions.html#intro

Why not? If you are correct, it should be easy to produce a cite that
says so. Hint hint.

On the other hand, *I* can support the position that such a work would
be derivative of both. See, e.g., 17 USC 101, section 101:

'A "derivative work" is a work based upon one or more preexisting
works...'

See the words "or more"? QED. I consider this decisively closed unless
you can cough up some kind of support for your, um, fanciful position.

--
Sincerely,

Ray Ingles (313) 227-2317

"...while several federal agencies acknowledge that they
contribute names to the [secret blacklist of people not
allowed to fly on commercial flights], none of them, when
contacted by The Chronicle, could or would say which agency
is responsible for managing the list." - S.F. Chronicle

I redundantly reiterate and repeat: "The decision then goes on to

programs..."

Sort of. "We emphasize that, like all copyright infringement cases,
those that involve computer programs are highly fact specific. The
amount of protection due structural elements, in any given case, will
vary according to the protectable expression found to exist within
the program at issue."

http://digital-law-online.info/lpdi1.0/treatise22.html

"Clearly, the highest level of abstraction - the purpose of the
program - is not protectable and would be filtered out. At the lowest
levels - the source and object code - total or even substantial
copying is likely to be an infringement. It is at the mid-levels
where the filtering is generally performed."

(Note that this directly contradicts your claim that copying of
source code in general is legal...)

>> ...the invertability of the translation is irrelevant. Both source

Simple syllogism:

If I own the copyright to a particular work, and a translation is made
of it to another language, then I own the copyright to that translation.
(This is totally supported by copyright law.)

Now, I own the copyright on the translation, right? So, if *that* work
is translated, then *by the very same principle as above*, then I own
the copyright in that second translation.

(Any chance you might answer this one, Bill?)

I distinctly remember reading about a case of a photographer sued by
a city council for this, but I can't find a cite on Google.

All right, fine. This is pointless. All you ever say is, effectively,
"is not". As I told DFS, I can talk to my two-year-old if I want someone
to say that to me over and over.

--
Sincerely,

Ray Ingles (313) 227-2317

"Look, you can't 'show' someone why the fish slapping dance is funny.
A person either appreciates it, or not all, on a totally fundamental
level and you can't change someone in that kind of way." - Nermal

But software is subject to copyright, unlike physical artifacts. And,
yes, when you include copyrighted work in another work, you create a
derivative work. To paraphrase Babbage, "I am not able to rightly
apprehend the kind of confusion of ideas that could provoke such a
[statement]."

> The derivative work must somehow substitute for the original...

What? A derivative work *incorporates* material from another work,
and some original work. What the heck do you mean by 'substitute for
the original'? It's just dumbfounding to be confronted with such
determined obtuseness.

--
Sincerely,

Ray Ingles (313) 227-2317

"I never said, 'Thou shalt not think.'" - God

*replonk*

(He is one of the most tedious posters I have ever encountered.)

--
When all you have is a hammer, everything looks like a nail.