Hacker News Reader

Comment by downandout | Hacker News Reader

It's very clear that Goldman is trying to punish him for leaving, while sending a message to their other programmers that if they leave, their lives will be ruined. Even if this guy ultimately wins the state trial, he will never get back the years he has spent fighting it. Goldman has already won, and in this case that is appalling.

View on HN

No, it's very clear what he did was wrong. He works in an industry that does not allow you to email code home or copy it off of company computers. Goldman might be a little different, but where I've worked, this was definitely a no, no, and you could easily be fired for doing it.

Fired is one thing, but malicious prosecution is completely out of proportion to a rules violation.

According to the original article, Goldman was actually violating the license terms of the open source software it had modified by not releasing it. Further, while you and I may have our opinions, a judge found Goldman's actions to be so egregious that he ordered them to pay the defense costs for someone accused of stealing from them. That is exceedingly rare, and wouldn't have happened if there were any doubts about either Goldman's conduct or the intentions of the programmer.

I don't know details of this case but it's very unlikely GS was violating any FOSS license for just modifying code (assuming internal use only).

I think I see your point, but I wonder... do GS distribute code to their employees? To partner organizations? If so, what are the terms of that distribution? If they are anything other than the original license requires, that's a violation.

Oh noes, now we can't have nice things anymore...

I suggest you read the GPL faq which should help clear up some of the issues for you:

http://www.gnu.org/licenses/gpl-faq.html#GPLRequireSourcePos...

You can use open source code without giving up your modifications as long as you don't distribute it. Even if a GPL program is modified by GS and runs on their mainframes, GS is not required to release the source code.

You only need to release the source if you're distributing the modified code, which GS obviously would never do...

Goldman was actually violating the license terms of the open source software it had modified by not releasing it.

I down-voted you because people keep saying that and it is false.

The Vanity Fair article uses the qualifier "possibly" when talking about GS violating the licenses and that's because none of the major license families (BSD, GPL, MPL, etc) have a requirement to release modifications if the whole is not released outside the organization that made them. The Vanity Fair article does not say that the code was ever distributed out of Goldman.

http://www.vanityfair.com/business/2013/09/michael-lewis-gol...

And none of them have that requirement because distribution inside of an organization does not count as a distribution for the purpose of copyright law. (Though if you let people carry it out of the organization, that's a different story.)

That was a point in the original Vanity Fair article that really caught my attention because it seemed very wrong the way it was stated. None of us know all the details but--based on what was written in that article--it sure doesn't sound like a license violation.

> a judge found Goldman's actions to be so egregious that he ordered them to pay the defense costs for someone accused of stealing from them. That is exceedingly rare,

No, GS will have to pay because the defendant was an officer at GS.

> Further, while you and I may have our opinions, a judge found Goldman's actions to be so egregious that he ordered them to pay the defense costs for someone accused of stealing from them. That is exceedingly rare, and wouldn't have happened if there were any doubts about either Goldman's conduct or the intentions of the programmer.

I don't think this is at all what happened. I think a judge ruled that he was a director at the company and therefor the company had to cover any legal costs he incured due to being a director of the company.

As far as I can tell, Goldman's actions had nothing to do with them paying his legal bills.

From the Vanity Fair article, it sounds like Goldman is in fact violating the terms of the license, but not in the way you're describing. (I.e., not by refusing to release changes.)

According to the article Goldman would routinely strip off the license/copyright from off of the open source code file and replace it with their own copyright. And that is a violation of many open source licenses.

I could be wrong, but wouldn't it only be a violation to release the code after stripping off the license/copyright?

Whoever modified the copyright headers then redistributes the files internally to other programmers and systems. You cannot modify a copyright to your liking and redistribute the work.

Any Russian programmer consider any administrativelimitation stupid, amd has the most of the code carved in his wetware. GS and other bozos can try to sue you for keeping the code in your brain, since you obviously taking our brain home. Or do American programmistas leave their brain in the office to comply?

Oh please. If you're Russian, you and I both know that bureaucracy and administrative detail (especially in terms of printed contracts) is well and alive in Russia. Stop being obtuse.

You're conflating the ideas of right and wrong, of law, and of company policy. If something is against policy, that doesn't make it illegal. If something is illegal, that doesn't make it wrong.

>No, it's very clear what he did was wrong.

That might be clear to rubes who believe everything that they see on television.

What's clear to me is that Goldman Sachs wants to ruin this guy's life. Which, to some people, makes him a stand-up guy.

Wasn't he doing this for the entire time he was employed at Goldman? They only checked _after_ he decided to leave?

...you could easily be fired for doing it.

Except he wasn't fired; he was sent to federal prison. Those two results are not equivalent.

Yes he did something wrong and deserved to be fired (if he was still working there), but he definitely doesn't deserve to be facing these ludicrous trumped up charges.

Goldman Sachs have come out of this as the bad guy, throwing their massive legal resources against a single individual.

Well its a bit of both. Goldman claimed at trial that the code uploaded could be used to manipulate the markets. If a python erlang bridge can be used to do that it is certainly a new one on me. So what he did was wrong. And Goldman was trying to send a message to people leaving. At the time people were leaving their HFT group and they wanted to do anything possible to stem that.

At best this should have been a civil lawsuit. The public almost certainly has no interest in what happened.

  He works in an industry that does not allow you to email 
  code home or copy it off of company computers.

If you read through some of the subpoena petitions in the docket[1] such as [2] you'll find that the defense was asking for documentation of conversations indicating that it was a normal thing for him to take work home with him, and his supervisors were aware of it. Since most of the evidence is under seal (as some of it contains trade-secret information), it's hard to say if he found the evidence that he needed. But it does look like it he's at least claiming that it wasn't unusual for him to take his work home with him.

For example:

  The defense will demonstrate at trial that Aleynikov
  regularly transferred files offsite and to a local 
  directory on hosts at Goldman’s development network so 
  that he would be able to continue working productively in 
  case of a network outage at Goldman to servers hosting 
  repositories with platform source code. The records 
  requested in Item 21 will demonstrate that shortly before 
  Aleynikov left Goldman, there was an outage that made 
  access to source code unavailable at least for several 
  hours that impacted developers’ productivity. This
  information will aid the defense in demonstrating that 
  Aleynikov did not have the specific intent necessary to
  commit the crimes with which he was been charged.

[1]: https://ia600209.us.archive.org/9/items/gov.uscourts.nysd.35... [2]: https://ia600209.us.archive.org/9/items/gov.uscourts.nysd.35...

How is that clear? I haven't followed the case, but it seems as though he copied important intellectual property (HFT code) when he left the firm. I'm sure if a developer at Apple or Microsoft downloaded the Windows or OS X kernel code when they left that they would face charges of some kind as well.

I read the original (very detailed) article. According to it, everything he copied was non-proprietary. Almost all of it was open source stuff.

Changes to open source stuff may or may not be proprietary, depending...

Not according to Goldman Sachs: "While some of those files included open source software, the Court determined that 'a substantially greater number of the uploaded files contained proprietary code.'"

Sounds like two different definitions of the word "proprietary" are involved here:

(1) Mechanical changes specific to the company -- this could be as simple as hardcoded hostnames that match the deployment environment.

(2) Implementations of algorithms. In this case it isn't the code that is proprietary so much as the algorithm that the code is an expression of.

I suspect that downandout is talking about defintion #1 and the court is talking about definition #2. Hard to say without seeing the code itself.

Presumably 1 subsumes 2.

According to it, everything he copied was non-proprietary. Almost all of it was open source stuff.

That doesn't make any sense. If the code truly were FOSS, why on earth would he copy it from his employer when he could eliminate legal risk by downloading it off the open internet? We know he's not stupid.

The story I read about it claimed the following:

He had made changes/improvements to various FOSS projects he was using at Goldman that he wanted to submit as patches upstream but his bosses wouldn't let him. He figured that since he was leaving he might as well take the FOSS code he had modified with him so he could submit patches later. The legality of this maneuver is tenuous at best, as others in this thread have noted, but in my opinion probably doesn't merit years of hounding prosecution and criminal charges.

Okay. That's interesting. I guess it could be true. Or it could also be a very creative and barely plausible explanation that happens to fit the provable facts of the case. It certainly requires us to accept an astonishing degree of naïveté in an otherwise very smart guy.

> It certainly requires us to accept an astonishing degree of naïveté in an otherwise very smart guy.

You haven't met many programmers, have you?

(I'm speaking of myself as well as friends and associates, here.)

There was a really well-written feature on the case. As I recall, he used and modified open source code in his system. He tried to push his changes upstream, but ran into political pressure from inside of Goldman, so he tried to make a personal copy of the libraries he modified.

You can certainly make the case that Goldman paid for that code, and if they didn't want it pushed upstream that's their prerogative, but it's a very different thing to sell trade secrets to a competitor (how the mass media has reported the case) and to make a backup of an open-source library you modified at work.

The article you reference is linked to within the posted article.

If all the changes were only his idea, this may be a reasonable assertion, but at large firms like goldman, the modifications he was making to code very, very likely came from insight from other team members.

If he incorporated other's insight in to his open source code by adding more code, I would view his actions in an extremely negative light.

Of course. At an investment bank such as GS, no code or algorithm is written that isn't checked by traders and managers...

Very dry humor!

Not humour, fact. You think billions of dollars a day in transactions are going to be entrusted to a single person writing code? Every single line, algorithm, and bracket is going to be examined by a dozen people over and over.

When a single bug can do a billion dollars in damage or bankrupt you, nothing is left to chance...

I think his reference was to the debacle at Knight, where what that level of rigor did NOT happen. Or, it happened, but was insufficient.

Kind of off-topic but the Mac OS X kernel code is actually (mostly?) open source.

http://www.opensource.apple.com/release/mac-os-x-1084/

That is essentially correct. However this fight is about more than just him. Essentially he took open source that he modified while at Goldman (Which Goldman prohibited him from releasing *fixed). So while his life is ruined (Although he is still better of then 99.99% of people out there) there is a mighty precedent in the works here: employees can't go to jail for legally following license.

"while license explicitly said if you modify it for commercial product you must make your contributions public"

I hadn't heard this part of it - what was the license in question?

I'm also curious about this. Not even AGPL would have such a requirement.

Not unless they were using it to offer some service electronically - which is conceivable, but doesn't seem hugely likely.

I originally read it in some in depth article that was talking about this. I can't find this now, so I removed that part. Now that I think about article might have been talking about "spirit" of open source license - if it is useful to you, contribute back, rather then actual license requirement.

> (Which Goldman illegally prohibited him from releasing, while license explicitly said if you modify it for commercial product you must make your contributions public)

I don't know the exact details of the case, but just to clarify: under the GPL, you don't have to release your changes if you aren't distributing that commercial product.

e.g., If he modified it for in-house use, there is, AFAIK, no legal obligation to release those modifications (under the GPL).

If he modified it for his own use, then that's clearly the case. If he modified it for the use of others within Goldman, they might have conceivably had a claim to a copy of the source under the GPL; he would not. I'm not entirely sure of that, or of whether employment contracts &c could limit that. For that matter, I'm not entirely sure what the answer should be there.

I think an internal user could easily be persuaded not to demand the source code.

In the typical case, probably, but if you have a large organization you could easily wind up with one exception. What rights these people do and don't have still seems important.

One exception who invited legal trouble for the organization as a matter of some kind of principle would likely find himself no longer a member of the organization. In the best-case scenario, this person gets to see the source code for an internal tool. Who exactly is going to fund this legal battle?

What about one exception who was already on the way out? Or what about one exception who was engaged in industrial espionage?

I'm just saying if you're just relying on employees not exercising rights they have - after which something you've been trying to prevent distribution of can be freely distributed - your situation is somewhat fragile.

Apparently the company itself if he or she is an officer of the company (as in this case). :P

If he gets convicted of a felony it is pretty certain he won't be able to work on high frequency trading systems that compete with GS, which is likely all that they want.

I think they're more concerned with their reputation and the perception of their internal practices and controls, and most importantly the value of their "technology." Just read the second and third sentence of Goldman's reply.

Things always appear clear to conspiracy theorists.

Do you have facts to back this up? People join and leave jobs in the financial industry every day. Is Goldman Sachs special in this regard?

Do you have any idea how rare it is for a judge to order the alleged victim of a crime to pay the defense costs of the alleged perpetrator? If I'm a conspiracy theorist, the judge that made this ruling is as well.

Ok can we please stop with the "conspiracy theorist" label/ad hominem attacks? After the NSA revelations have shown things to be a lot worse than even the more paranoid conspiracy theorists thought, folks with an eye for corruption have earned a little wiggle room to be taken seriously.

If you begin a statement with "It's very clear" then I expect you to back that statement up with facts so that the rest of us can see why it's so clear. Does it seem reasonable that Goldman Sachs want to send a message to their own programmers? Is that going to work well for them in recruiting?

If anything, the NSA revelations have given conspiracy theorists carte blanche to go full steam on their theories. You'll hardly find a thread related to Google on HN without the obligatory "Well, now that we know that Google has given NSA has unfettered access to all data..."

Another important aspect is whether or not the belief in these theories is justified, regardless of whether or not they turn out to be true. This kind of relates to the Gettier problem in epistemology, where one may "know" something completely by accident, and we wouldn't really consider that "knowledge."

Were conspiracy theorists justified in believing their theories prior to the NSA leaks? Maybe, maybe not. But we should probably consider that independently of whether or not their theories turned out to be right. After all, we can never be 100% sure of our knowledge; the best we can do is seek the most reliable methods of inquiry. If a conspiracy theorist makes 50 predictions and 45 of them turn out to be correct, then we might have a reason to adopt their thought patterns; if someone has one big suspicion and that suspicion turns out to be true, I'm not sure how much we can conclude from that.

(Sorry, I know this discussion is getting off-topic... I just have a mild fascination with conspiracy theories and epistemology.)

Yep. This is the same question as people who always believe there is a bubble. Really great comment.

Nothing the NSA was doing was news. Or indeed even hidden, much of it having been previously reported on HN and even the mainstream news.

Few people connected the NSA dots because few people connect any dots. There are dozens of other important stories snowed under by the same blizzard of laziness and stupidity.

After the NSA revelations have shown things to be a lot worse than even the more paranoid conspiracy theorists thought

Really? Even the more paranoid conspiracy theorists didn't think that the Government could read the address and subject of your emails? The same people that think the Government did 9/11 and faked the moon landing?

Don't even get me started on chemtrails...

No, we can't really. Being right once doesn't mean you get to throw around wild accusations without proof and avoid being called on it.

Those people wearing the tinfoil hats were right.

Reply on news.ycombinator.com