 Wednesday, 16. July 2003
§ 86 TKG
Shortly after we published Factual and Legal Risks regarding wireless Computer NetworksI got aware of § 86 TKG ("Telekommunication Code"). § 86 TKG states:
A radio installation may not be used to receive messages not indendet for it. The content of such messages and the fact thet they have been received may not be communicated to others, even if the recieiving was unintential [...].
I never got arround reviewing legal literature on § 86 TKG, but it is ovious that this paragraph is so broard, it has to be interpreted restrictivly to be compatible with the constitution.
Now I steped on two court cases which indicate that the courts are very reluctant to apply § 86 TKG. [disLEXia 3000 blog]
08:32 |
permanent link |
mail this
Sunday, 09. February 2003
Programm im Sinne des (Straf-) Gesetzgebers
"Programm" ist die Arbeitsanweisung an einen Computer, die aus einer Folge von
Einzelbefehlen (sog. Programmablaufschritten) besteht. (BT-Drs. 10/318 S. 20).
"A Programm is a job instruction to a computer which consists of a series of commands (called programm steps)." (from the Parliament Documents on creatiing german cyber crime law,, ca. 1985)
11:24 |
permanent link |
mail this
12.2.: University of Bonn: Lecture on Internet Crime
PD Dr. Friedrich Toepel invited me to give a lecture on "Internetcrime" as part of his course in economic crime. So if you are near, just drop by at 16:00h in "Hörsaal N", Adenaueralee 24-42. I'll focus on why "Internetcrime" is something special and present some "interesting" cases. Lecture Notes will be online after Wednesday.
See the anouncement:
Betr. Wirtschaftsstrafrecht am 12.02.2003, Internetkriminalität:
Am 12.02.2003 wird Herr Dornseif, der selbst auf diesem Gebiet seine Doktorarbeit
schreibt und ein hervorragender Kenner der Materie ist, einen Einblick in die
aktuellen Probleme der Internetkriminalität geben.
11:07 |
permanent link |
mail this
Saturday, 08. February 2003
Program vs. Data - more notes to myself
Followup to http://md.hudora.de/blog/guids/80/44/6531103823136208.html
If we accept 'compound objects' of program and data how to handle that? IS is a program or is it data? How to decide? By percentage of bytes?
Are there Programs not created by man? Does creating a program really just can be writing some descritions in a "Programming language"? Can't a program also be painted?
Is recording a macro an act of programming? If I insist that it isn't - "you create a macro by clicking on buttons in your word processor, you don´t program it yourself - you simply tell the computer what kind of macro it shall program" (Olaf). I think this is a to limited view. Many big software systems are created nowadays with help of CASE tools which are not that different from MS Office's macro wizards.
If I create a C Programm - don't I just instruct the Preprocessor (or codegenerator) to write the program for me?
Isn't most 'logic programming' (right name? Traffic lights etc.) done by 'drawing' the logic?
It might be hard to decide if data structures are 'program' or 'raw data'.
I still insists Postscript files written by freehand/xfig/whatever are manmade programs. They express the authors intent to make a computer act in a certain way.
Perhaps creators of the law intended programs to be the stuff you don't change further and data the digital documends you further edit. This might fit the worldview of a 1985 lawyer.
Think about it again:
Perhaps when they created the law they didn't mean computer programs but
"computer data of any value." In fact this would make sense. Remind you: in the
80ies the only commercial available digital asserts where databases and programs.
Databases where not protected and later protected by special rules. Why did
programs need special rules? Because they are so easy to copy. Doesn't this apply
for all digital "stuff"? So perhaps everything is a program in legal sense.
12:11 |
permanent link |
mail this
Friday, 07. February 2003
Program vs. Data - notes to myself
Links on the subject:
Code never meant to be executed: Donald Knuth said once. "Remember I have only prroven that this programs are correct - I have never actually run them."
Could we reduce programs to binarys? No, see scripting languages, (Java|Python|Lisp|whatever) Bytecode, Microcode, CPU emulation etc.
Does a program need to be written in a programming language? This would not help us with code like /* this */
A classical 'binary' is not directly executable by the processor. The binary loader has to do a considerable amount of work to get it in a runable state. Think of dynamic linking, relocating, initialisation of segments etc.
Lambda the Ultimate reminds us that this discussion comes back to fundamental problems of computer science. See also me being cursed with the lack of a german term for 'computer science' - we use 'Informatik' instead but using a clumsy question I get an interesting answer regarding entropy as a clue to distinguish program and data.
Olaf: The work must be fixed in a tangible medium = This is, within our discussion, no problem. Nice talk I heard last summer on a "hacker conference". The speaker tried to keep data permanently "in flight" on the internet, e.g. by encapsulating it in a ICMP echo request (ping packet). Fascinating line of thought.
Olaf: ((4)) And, of course: If it shall be protected as a 'computer program' it must be a computer program. Doesn't the german law also state: "Computer Programs are programs of any kind"?
Perhaps when they created the law they didn't mean computer programs but "computer data of any value." In fact this would make sense. Remind you: in the 80ies the only commercial available digital asserts where databases and programs. Databases where not protected and later protected by special rules. Why did programs need special rules? Because they are so easy to copy. Doesn't this apply for all digital "stuff"? So perhaps everything is a program in legal sense. No "fair use" sorry.
We might not like that outcome but perhaps we should change it not by tinkering whith the definition of 'program' but in anoter way. Which Way?
11:53 |
permanent link |
mail this
Monday, 27. January 2003
Legal Books by Lenz
Karl-Friedrich Lenz is publishing several legal books (in german) under a Creative Commons License. More power to him!
13:41 |
permanent link |
mail this
How to distinguish software (programs) from data?
Olaf and I looking for ways to distinguish Software and (other) data. So what is a Programm?
In "Der Datenbegriff im Recht" we (Kay Schumann and I) suggest that distinguishing between programms and data is impractical.
There is an ugly catch. The Law assumes they can be destinguished. Especally interlectual property law provides special rules for software.
So how to distinguish software (programs) from data?
12:30 |
permanent link |
mail this
Sorcecode, Binaries and NDAs
If you consider binarys as equivalent to sourcecode how could you give out a programm (e.g. a hardware driver) and still comply to non-dislosure agreements you might have signed regarding on the information manifested in the programm.
E.g. when you want do develop a deriver for special hardware the hardware manufacturer will give you the technical specifications only if you sign a NDA keeping you from giving away the sourcecode of the driver you are developing. This hurts many Freie Software Projects. But if 'Open Code = Closed Code' you even can't give away the binary driver.
11:04 |
permanent link |
mail this
What is sourcecode? - The GPLs definition.
Lat summer we had some discussion about the meaning of source code and object code see some of it at:
Now Olaf Koglin points out that Richard Stallmann has thought about this a long time ago and found a very nice definition for the GPL:
The source code for a work means the preferred form of the work for making
modifications to it. For an executable work, complete source code means all the
source code for all modules it contains, plus any associated interface definition files,
plus the scripts used to control compilation and installation of the executable.
But it still has some problems. E.G. I have started using Unicode Terminals, an Unicode-Aware Editor (XEmacs), and converrted most of my plain-text files from latin1 to unicode. Some of my newer Python source code is unicode, too. Since python does not grok unicode sourcecode as it should a shell script does a recode utf-8..latin1 on the fly before executing. With the above definition the latin1 version of the source code would not being considered source code. I'm not sure if this is a desirable result.
We could say, that the results of lossless transformations on the source code still could be considered source code - but isn't encryption a lossless transformation, too?
11:02 |
permanent link |
mail this
Sunday, 26. January 2003
Techlaw Column
Mark Grossman on Techlaw - "Florida lawyer Mark Grossman's "TechLaw" column for the Miami Herald doesn't have a lot of depth, but provides useful alerts and ideas. Archives and the current issue are available." [via Internet Tools for Lawyers] [Ernie the Attorney]
23:14 |
permanent link |
mail this
Thursday, 23. January 2003
Use the right language for the right effect.
Doc Searls writes about language used by the different sides in the Eldred Trial:
Catch the language. While the one side talks about licenses with verbs like copy,
distribute, play, share and perform, the other side talks about rights with verbs like
own, protect, safeguard, protect, secure, authorize, buy, sell, infringe, pirate,
infringe, and steal.
This observation is a problem I have thinking of for a long time. First you can't steal immaterial goods. At least not in Germany. Thivery in German law consists of taking away a thing from the owner and usurp it for yourself or somebody else. software and music are no thing. And you can't take them away. You might steal the medium where it is embodied on, but that's a different thing. In the usual form of infringement you even the owner does not even mention that you have "something" more, since the gread thing about digital goods is the fact that they can be copied without degeneration in source and copy.
While you shouldn't allow this fuzzy terms to lead you to fuzzy thinking you might try to use it against your opponents. Try phrases like:
00:52 |
permanent link |
mail this
Tuesday, 10. December 2002
Clever Anti-DMCA Argument by 321 Studios
PC World runs an interesting story on 321 Studios and their new software package, DVD X Copy, which permit users to copy DVD movies from DVD to a recordable DVD (New Tool Makes DVD Copying Easy). The article also talks about the ongoing DMCA-based lawsuit pre-emptively launched by 321 Studios against the MPAA and movie studios. [...]
What the article doesn't mention is the current status of the case, which is trying to survive the same legal question that torpedoed Ed Felten's case - whether a case or controversy exists. The movie studios claim that they have not threatened to sue 321 Studios and that, in any case, the legal issues raised by the case will be answered in the Elcomsoft case and thus do not need to be answered in the 321 lawsuit. The studios' case on this issue is fairly strong. I hope that 321 does it make it to the next stage of the lawsuit, but it is unclear whether that will happen. The next hearing in the 321 case will be held in January.
The article does have a particularly interesting quote, however:
What's more, [Robert Moore, president and founder of 321 Studios] says, DVD X Copy doesn't actually break the CSS on commercial DVDs.
Instead, 321 Studios intercepts the video and audio stream after a DVD player has decrypted the CSS code. Moore argues that all DVD players decrypt the CSS code when they play a protected DVD. Because it intercepts the signal after decryption but before the video is rendered, the product does not run afoul of the DMCA, he says.
This is actually a very interesting legal point. The anti-circumvention clause of the DMCA distinguishes between "access" control (17 USC 1201(a)) and control over other rights of the copyright holder (17 USC 1201(b)) (essentially, copy control). Unfortunately, the courts have failed to make this distinction, conflating access with use. After all, you have to "access" a work to use it, right? Thus, all technical protection devices become "access" control devices. This is not the meaning Congress intended and has the effect of making the DMCA even worse than it is. You see, it is legal to circumvent for purposes of making fair use if the device only protects other rights of the copyright holder (17 USC 1201(b)). It is illegal to circumvent an "access control device" for fair use purposes since, presumably, you don't have a right to access the work in the first place (17 USC 1201(a)). In both cases, it is illegal to distribute the tools to circumvent (even if using the tools is legal under 17 USC 1201(b)).
This claim by 321 Studios emphasizes the distinction between the two types of devices. After all, they couldn't capture the decrypted signal unless there was legitimate "access" to the signal in the first place. A clever and proper argument against the "access" control argument, I believe. However, if the claim is that 321 Studios does not violate 1201(b) as well, then it goes too far for a judge to accept (not that the more limited claim will be a difficult win as well). [via LawMeme via The Shifted Librarian]
10:02 |
permanent link |
mail this
Friday, 29. November 2002
Tech Provisions in Homeland Security Bill
Orin Kerr, over at the Volokh Conspiracy, summarizes some tech-related provisions in the new Homeland Security bill. [Freedom To Tinker]
11:11 |
permanent link |
mail this
disLEXia, a research project by Maximillian Dornseif
|
