Southperry.net
Code is not physical property - Printable Version

+- Southperry.net (https://www.southperry.net)
+-- Forum: Social (https://www.southperry.net/forumdisplay.php?fid=14)
+--- Forum: Current Events (https://www.southperry.net/forumdisplay.php?fid=55)
+--- Thread: Code is not physical property (/showthread.php?tid=53400)



Code is not physical property - Fiel - 2012-04-13

http://www.wired.com/threatlevel/2012/04/code-not-physical-property

Wired.com Wrote:Former Goldman Sachs programmer Sergey Aleynikov, who downloaded source code for the investment firm’s high-speed trading system from the company’s computers, was wrongly charged with theft of property because the code did not qualify as a physical object under a federal theft statute, according to a court opinion published Wednesday.

“Because Aleynikov did not ‘assume physical control’ over anything when he took the source code, and because he did not thereby ‘deprive [Goldman] of its use,’ Aleynikov did not violate the [National Stolen Property Act],” the 2nd Circuit Court of Appeals wrote in its opinion (.pdf).

The three-judge panel in New York also ruled that Aleynikov was wrongly charged with espionage, since the code was not a product designed for interstate or foreign commerce, a requirement under the Economic Espionage Act with which he was charged and convicted. The court found that Goldman’s system was neither “produced for” nor “placed in” interstate or foreign commerce, nor did the company have any intention of selling its system or licensing it to anyone.

The opinion finally provides explanation for why the judges delivered a surprise ruling last February that reversed Aleynikov’s conviction and sprung him from prison a year after he had begun to serve an eight-year sentence.

The ruling also deals a blow to the government’s ability to prosecute others for similar thefts of trade secrets under the EEA.

Aleynikov, 42, was convicted in 2010 under the Economic Espionage Act of 1996 (EEA), in a high-profile case that was held up by federal prosecutors as an example of the Justice Department’s serious intent to prosecute the theft of intellectual property and trade secrets.

Aleynikov had acknowledged that he violated the bank’s confidentiality policy in downloading the source code from the company’s computers, but he’d asserted that what he’d done was not a criminal act under the EEA because the code wasn’t used in interstate commerce.

In February, the federal appeals court, led by Chief Justice Dennis G. Jacobs, agreed and reversed the conviction. The judges announced their ruling without explanation just hours after hearing Aleynikov’s attorney argue the appeal. Chief Justice Jacobs said at the time that the court’s written opinion explaining the ruling would follow “in due course.”

In discussing whether the code qualified as a product of commerce under the EEA, the judges wrote in their opinion this week that contrary to Goldman Sachs’ source code being something the company distributed in the commercial realm, the company “went to great lengths to maintain the secrecy of its system.

“The enormous profits the system yielded for Goldman depended on no one else having it,” the judges wrote. “Because the HFT system was not designed to enter or pass in commerce, or to make something that does, Aleynikov’s theft of source code relating to that system was not an offense under the EEA.”

The judges also addressed the government’s argument that the code was physical property under a 1988 amendment to the NSPA. Prosecutors had asserted that the amendment reflected an intent by Congress to include the transfer and transmission of non-physical forms of stolen property in the law, thereby covering the theft of source code. But the judges rejected this claim, stating that the 20-year-old amendment clearly had been meant to cover the transfer and transmission of money, not the theft of source code in the computer age.

“We decline to stretch or update statutory words of plain and ordinary meaning in order to better accommodate the digital age,” the judges wrote.

They added that although Aleynikov should have known that his actions were in breach of his confidentiality agreement with Goldman Sachs, and were dishonest in ways that would subject him to sanctions, he “could not have known that it would offend this criminal law or this particular sovereign.”

Circuit Judge Guido Calabresi agreed with the majority opinion and the way the judges had reached it, but at the same time suggested in a concurring opinion that Aleynikov would not have gotten off if the Electronic Espionage Act had been better written.

“[I]t is hard for me to conclude that Congress, in this law, actually meant to exempt the kind of behavior in which Aleynikov engaged” he wrote. He also expressed “hope that Congress will return to the issue and state, in appropriate language, what I believe they meant to make criminal in the EEA.”

The Russian-born Aleynikov worked for Goldman Sachs until June 2009, when authorities said he siphoned source code for the company’s valuable software on his way out the door to take a new job with another company.

Aleynikov, a naturalized U.S. citizen who emigrated from the disintegrating Soviet Union in 1991, earned nearly $400,000 a year as a vice president with Goldman Sachs. He was arrested in July 2009 at the Newark Airport in New Jersey as he returned from a trip to Chicago to meet his new employers.

Authorities said he stole “hundreds of thousands of lines” of source code from Goldman Sachs in the days before he left the company, downloading various software from the Goldman Sachs network and transferring it to a storage website hosted in Germany, before trying to erase his tracks from Goldman Sachs’ network.

The software is used to make sophisticated, high-speed, high-volume stock and commodities trades and had earned the company “many millions of dollars in profits” each year, according to prosecutors.

Prosecutors said Aleynikov made several copies of the code and had it on his laptop when he met his new employers at Teza Technologies in Chicago, although a later search of Teza computers uncovered no copies of Goldman Sachs’ source code.

Goldman Sachs only uncovered the theft after it began monitoring HTTPS transfers and saw a large volume of data leaving its network, according to court documents. The company initiated the monitoring after noticing suspicious activity on the network.

At the time of his arrest, Aleynikov acknowledged taking the code, but told FBI agents he only intended to collect open source software files on which he had worked, and that his collection of proprietary files on his last day of work had been inadvertent.

This means:

- The BMS source code downloaded from an open server was legal.
- Any person that leaves their server open with critical files needs to have their heads checked.


Code is not physical property - Locked - 2012-04-14

I thought this would be about Oracle vs Google.


Code is not physical property - Exidous - 2012-04-14

Stupid back button ate my reply. Read the court's decision if you're interested. It's actually a lot more narrow than you might think. Under one specific (old) law that wasn't written to involve code or intellectual property, the court is saying the accused was not guilty of theft. I don't even think it would be double jeopardy for him to be indicted on charges from a more relevant law, since he'd be tried for a different crime.

Although regarding the internal reasoning of this case, I think the espionage charge should've stuck. Wickard v filburn craps all over the idea that something being used internally isn't interstate commerce. I don't agree with it, but that's the precedent.

The reason why I wanted to post was to say how important this realm of legal jurisprudence is and will be. A lot of the major problems faced by the internet stem from weird or outright bad precedent/legislation regarding intellectual property. If we as people/courts/legislators can sort that out, things like patent trolling and file sharing lawsuits could be a thing of the past.

Pursuant to this case, should code be patented or copyrighted? I don't know the answer, but I can tell you preserving code as a trade secret contributed to Goldman effectively losing this case.


Code is not physical property - Flonne - 2012-04-14

I saw this but decided against posting it, since I didn't really do any research on it beforehand. Nevertheless, this is a pretty important precedent, I just hope people don't take advantage of it...I can see self-entitled pirates flooding the courts because they think they have a real chance of winning (and maybe even do, now).


Code is not physical property - FenixR - 2012-04-14

Flonne Wrote:I saw this but decided against posting it, since I didn't really do any research on it beforehand. Nevertheless, this is a pretty important precedent, I just hope people don't take advantage of it...I can see self-entitled pirates flooding the courts because they think they have a real chance of winning (and maybe even do, now).

What pirates have to do with OPEN source code? None.


Code is not physical property - Flonne - 2012-04-14

Thus the "because they think" part.

And I added the "and maybe they do" because, you never know, our justice system is pretty fucked up sometimes, and judges are usually old men so they have no clue how to technology.


Code is not physical property - Shidoshi - 2012-04-16

Exidous Wrote:Pursuant to this case, should code be patented or copyrighted? I don't know the answer, but I can tell you preserving code as a trade secret contributed to Goldman effectively losing this case.

The problem with patenting or copyrighting code is that copy-pasting already optimized pieces of code is common practice for building your program. Having to do everything from scratch each time while having to make small changes so it doesn't step into a patent would be annoying and a hindrance to the work of programmers.


Code is not physical property - Exidous - 2012-04-16

Shidoshi Wrote:The problem with patenting or copyrighting code is that copy-pasting already optimized pieces of code is common practice for building your program. Having to do everything from scratch each time while having to make small changes so it doesn't step into a patent would be annoying and a hindrance to the work of programmers.
Sure, but making it possible or convenient to copyright code need not involve presuming all code is copyrighted. That's actually a relatively recent feature of copyrighting media. It didn't use to be that if you wrote a book you had an assumed copyright to it, you had to register for one.

And at least in this case, the code in question wasn't freely available to be copied. My comment was more about comparative systems of intellectual property rather than protected versus not.

So to rephrase, if you want intellectual property over code, what is the best way for society to enable that?


Code is not physical property - Eos - 2012-04-16

Exidous Wrote:That's actually a relatively recent feature of copyrighting media. It didn't use to be that if you wrote a book you had an assumed copyright to it, you had to register for one.


This statement is somwhat inaccurate depending on how you scale 'recently'.
The Berne Convention established automatic copyright in 1886.
The US has been part of the convention since 1988. From 1790 to 1988 the US would've been under the Copyright Act of 1790, which did require registration, but that was over 20 years ago and not recent by the standards of most people, only recent in comparison to the ~200 preceeding years. More accurate to say 'for the majority of copyright history in the U.S.' than 'recent'.

And of course it's patently untrue for the people who've been in the Berne Convention since day 1 who've had 'assumed copyright' as a de facto standard for over a century and a quarter.


Code is not physical property - Exidous - 2012-04-17

Eos Wrote:This statement is somwhat inaccurate depending on how you scale 'recently'.
I scale it 'relatively'. Over the course of copyright law in the US, it is a relatively recent development that copyright is automatically assumed over written works (of various kinds) rather than explicitly registered for. It's part of the reason why we've been so slovenly to address the impact on machine code - for most of the history of intellectual property over code, it has been this "new" copyright regime. With utterly inadequate analysis as to whether it's at all appropriate.