The US Code in XML

From this
groklaw post
, I learned that Cornell University has transformed the
entire US Code to XML
. Now, I’m not well versed in US Law, but
this is basically all federal laws currently in effect in USA, right?
This is a very cool project. I’ve started to look around, but
seemingly as I only barely can navigate around in swedish law texts,
there’s a lot of exploring to be done. If you want to just read
the laws, this
seems to be a good place to start.

(By the way, could any US law scholar tell me whether the US
constitution is a part of USC, or if it’s a separate collection of
documents? I cannot seem to find them, only references to it.)

As I’ve been working
on a project with similar goals for the
last few months, I’m particularly interested in the XML format
looks. I’ve only started to scratch the surface, but initially it
seems to be somewhat more presentation-oriented (and way more complex)
than the simple structures I use.

Maybe this format could be used anyway, or at least a XSLT
transformation to transfer from (not yet documented) to uscfrag.

Terry Fisher: ”Promises to keep”

Lessig comments on
and recommends
‘s book ”Promises
to keep”
, with the subtitle ”Technology, Law and the Future of
Entertainment”. Coming from a technology background and moving into
law, these discussions always interests me. The introduction and final
chapter is available
, and from what I’ve read so far it’s a very readable
discussion on the historic reasons for intellectual property rights
(including patents, trade secrets and copyrights), how digital
distribution changes the prerequisites for the existing laws, and how
new ways of looking at intellectual property can bring legislation
that encourages the creation of cultural content and innovations,
without restricting user’s rights.

It seems that Fisher’s advocating a sort of content
solution, which I’ve previously written of as
unrealistic, but through a brief discussion around ”public goods”, he
argues his case very strongly. I will probably have more to say once
I’ve read the entire chapter.

Two more things on copyright and related laws

In regards to the previously mentioned controversy around The
Pirate Bay: The signature ‘Judas’, acting as legal counsel for TPB, argues that
providing a link (in this case a .torrent file) to copyrighted
material is not copyright infringement under Swedish law. He cites the
verdict NJA 1996 s. 79 (which isn’t available on the web, so I haven’t
read it) in his support.

However, from the context he gives, it seems
to my non-legally-trained eyes that the verdict NJA
2000 s. 292
concerns a case which is much more similar to the TPB
case. The verdict is about a case where a person provided links to MP3
files that were not hosted on his site (i.e. potential infringement of
the second degree), and it ended in partial victory for the
plaintiff. Any laywers well versed in Swedish copyright law reading
this, please feel free to weigh in 🙂

On a different, but sort-of-related note, I was browsing around Nicklas Lundbergs blog some
more, and found this
interesting paper
on the legality of search engines. Two examples
in particular that interested me:

First, the quote [my
translation]: ”It’s not impossible – with regard to the things
previously said – that [search engines] would never have come into
existance if somebody had first consulted a legal expert”
— food
for thought.

Secondly, the (un-tried) argumentation that
governement agencies should not try to restrict search engine access
to their websites (through robots.txt or similar means),
since it would violate the Swedish principle
of free access to public records
. It’s particularly interesting
since Rixlex, the main
repository of swedish law texts, does just that.

Law is not code

I have not yet read Lawrence Lessig’s ”Code and other laws of
”, but I like its catchphrase, ”Code is Law”. Sadly,
the reverse do not hold, particularly not in the real world.

As a programmer, it’s natural for me to view any legal text, be it a
law, an agreement or something else, as a set of definitions,
conditions and other instructions, similar to a computer program. Most
of the time, these instructions are badly written; they do not strive
for simplicity, they do not cover corner cases, and for many inputs
the outcome is impossible to determine without resorting to case law.
In computer terms, way too much checking is done at runtime (when laws
are applied) instead of compile time (when they are written).

There’s an interesting
up on Kuro5hin
which argues along the same lines, and how the awful quality of most
laws makes for a very opaque society, where only legally trained
people with a lot of time on their hands can fully understand the

The author draws paralells between the way laws are written and
computer programs are constructed, and suggests that the law making
process should be ”open sourced”; I think the process is already about
as open as it can be. What’s lacking is good engineering: the will to
strive for clear and simple designs.

Computer programmers realized the need for a structured approach to
programming almost as soon as programming was invented. This was
needed to be able to construct and maintain large programs. There is,
no doubt, a lot of structured thinking in law as well, with its 3000+
year history, but somewhere along the way it seems that clear and
unambigious writing fell off the list of things to prioritize.

Of course, my reasoning is a classic example of someone, new to a
field, thinking they can apply all lessons learned in another field. I
might have more insightful commentary after a few years of law school.

The GNU GPL, modifications and swedish copyright law.

Yesterday, I was involved in a discussion that started with the Affero General Public
, a modification of GNU GPL, indented to close the ASP
in GNU GPL v2. The ”ASP loophole” referst to the fact
that a Application service provider (ASP) can make the functionality
of a GPL’ed program available, without distributing the actual
program, through a Web UI or something similar.

The Affero GPL (and upcoming
GPL v3
) intends to close this by demanding (in 2 d) that, if the
software has functionality to allow users to download the source code
of that program, you may not remove this functionality.

Now, the GPL (both Affero and GNU v2) is not a binding license. It
even says so itself in section 5: ”You are not required to accept this
License, since you have not signed it”
. It’s power derives from
the fact that, unless you accpt it, normal copyright law applies,
which forbids you to redistribute software for which you do not hold
the copyright.

However, to exploit the Affero/GNU v3 GPL, you don’t have to
redistribute the software, just remove the feature that allows users
to download the source. And if you don’t accept the license, who’s
going to stop you? Section 5 of the Affero GPL goes on to say:
”However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License.”

So, the whole power of the Affero GPL hinges on the fact that,
under normal law, you do not have the right to modify a program for
which you do not hold the copyright. But is that really correct? I’ve
read and re-read the Swedish copyright law, and I cannot
find where it says that personal modifications are not allowed. In
fact, it explicitly allows the modification of programs and copying
(but not redistributing) the result in 26 g §, in certain
cases. The swedish copyright law is all about making copies and
the redistribution of copyrighted material, not modification-making.

So, under Swedish law, are you allowed to modify software for which
you do not have the copyright? If so (and I think that’s the case),
the raison d’etre for Affero GPL is null and void in
Sweden. Presumably, this is different under US copyright law.

If you know (or think that you know) that I’m wrong in the above
assumption, I’d love to hear about it, preferably with a reference to
a law and section, or reference to a precedential case. Or, you know,
just with a logical reasoning about why I’m wrong. Still no talkback
system, but I check my referrer logs, so if you blog about it, I’ll
read it. Or just email me at

The one thing I can think of that would make modifications of a
software program illegal, according to Swedish law, is 12 §, which states that
one may not create copies of a computer program, not even for personal
use (as opposed to most other works such as books or music, where
personal copying is allowed). Maybe it can be argued that modifying a
program is, in effect, making a copy of it?