Did Goldman Overstep in Criminally Charging Its Ex-Programmer?

I think vested stock options not expiring a few months after leaving the company is far more valuable, but no one does that. It would allow you to put no money down and only exercise when the options are actually liquid without being forced to purchase the stock at some point due to circumstances not under your control.

it is when you're arrested. Until arrested - my understanding that one don't have to answer questions. Though personally i found it is really impossible to insist on your rights, i basically just chicken under stress : https://news.ycombinator.com/item?id=6142434

Their job is to put away criminals. In the case where there is not really a crime, their training, which is meant to get confessions from actual criminals, ends up pulling seeming confessions to nonexistent crimes from innocent people. After that happens, you are screwed.

Simply refuse to comment.

.. man this sounds like some mafia.

Every company has its own policy, but I would bet GS wasn't cool with putting their code on personal machines.

http://rationallyspeaking.blogspot.com/2012/11/odds-again-ba...

EDIT: Actually, this is the one I should have linked to: http://www.avclub.com/articles/did-12-angry-men-get-it-wrong...

The defendant was a gang member and a participant in an armed robbery, and when things got ugly (the victims fought back) one of the gang members shot one of the victims.

It was not proved to us that the defendant was the shooter, but -- as I was shocked to learn -- California law put him on the hook for the murder.

Possibly whoever actually shot the victim is still at large. The shooting took place at night, outdoors, the eyewitnesses were all drunk, and their testimony made it clear that, under the circumstances, they could barely tell one black man from another.

I could see theft of property, possibly, but the furthest I could see it going is breaking an employment agreement, which is a contract between two private parties and doesn't bring criminal charges for violations.

By both the letter of what you agree to and industry standard it's not really ethical to dump your employer's code into Dropbox or whatever. They go a long way to block it, blocking sharing sites, USB ports, etc. I'm not saying it doesn't happen all the time, just like brokers taking their client lists. People justify it by saying 'really, I'm supposed to look these people up in the phone book?' or 'really, I'm supposed to re-implement this simple function?'

How unethical sort of depends on how much of an edge you take, it's one thing if it's stuff you did yourself that would be inconvenient to recreate, the equivalent of your spiral notebook, something else if it's what other people did that would have been impossible to recreate.

But I think Serge demonstrates ignorance and naivete by thinking that was normal practice, and both Serge and Lewis demonstrate an ethical blind spot.

Did he deserve to be made an example of? I have no idea if he did or why that happened. Maybe there was stuff in there that was viewed as very proprietary. Maybe it was clever, or they just had an inflated sense of its value. Maybe what seemed trivial to Serge seemed extremely valuable and proprietary to GS. Or maybe he just pissed off the wrong guy. Or maybe they were getting poached a lot and decided to take a stand. I don't think we'll ever know. Certainly not based on this article.

This is a huge grey area, well maybe not so grey right now. Often you can't seperate quants from their excel sheet( or R models) and they will keep these models as they move to new firms.

As far as programmers keeping the code they wrote, This is the first time I've ever heard of a programmer, and I'm one, leaving a shop and bringing their code with them.

Let's just say, there is no ambiguity what-so-ever about how employers feel about this in the finance industry.

Secondly: Being fired is not what happened to this guy. He was prosecuted and sent to prison. Not even remotely the same thing.

Maybe it should be more incumbent on politicians and law enforcement to learn more about technology and incorporate more private citizens as opposed to just big corporations in the dialogue.

It's frustrating, I think everyone realizes that technology is going to become a greater and greater part of every day life in government and yet there hasn't been much of a push for elected officials to give a shit to do anything but fundraise with it.

Life isn't that black or white. Companies run on soft games, the world, your product succeeds not because of "cold, hard, objective" amazing technical design but due to soft irrational people. I'd argue that there are enough people in tech well versed in the soft game for money to be made.

Especially if you are thinking (later in the article, quoting the programmer), 'they wouldn't like it', 'it was like speeding, in a stolen car'.

> The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

That makes it sound like ripping the license out and replacing it with a proprietary one, even for purely internal use, is probably a bad idea. I don't know who'd be able to sue them for that, though, if the original copyright "owners" aren't supposed to have seen it in the first place.

Article II of the GPL has a specific exception for transferring GPL-ed software to employees to work on without letting them spread it beyond that, but I don't know how much of this that would cover.

He deleted his bash history, which many programmers do every day, indeed, that's what the justice department used as evidence that he "deleted the audit trail"

Once you look at the details of this case--in fact, even just the details as recited in the public documents filed by the Justice Department--a completely different story emerges. A few facts just for context:

the person arrested and later set free by a Federal Appeals Court, Sergei Alienikov, is an accomplished programmer with a long history of open source contribution, for instance, he is the author of the (original?) ZeroMQ Erlang bindings. (http://zeromq.github.io/erlzmq/)

The code that Goldman claims that Alienikov "stole" was code that he had written while employed by them; indeed, according to Goldman SA was hired to build their HFT rig, which he did. According to the Justice Dept documents, he only took a small portion of the total codebase (~ 2%) and he took that on the last day, six weeks after he had resigned from Goldman.

SA claimed that code comprised his revisions to an open source library he used to build Goldman's HFT system. Combined with the fact that he took that on his last day in the office (he had resigned six weeks before (at Goldman's request, he stayed an additional month) and was open about his plans to work for a competitor, (and which his employment agreement did not forbid). Clearly a smart guy, but a pretty dumb criminal, which pretty strongly suggests he's no criminal at all. If his plan was to steal this codebase (the one he had in fact spent the previous two years writing) he would have, to begin with, actually taken all of it, not just a small fraction, and second he would not wait until his last day. These circumstances were relevant in that case because SA was on trial for committing a felony, and to prove guilt requires proving some criminal state of mind ("intent to steal", the complaint calls it.)

What's more, Goldman had absolutely no evidence that SA used any of this code or gave it to anyone. The government's complaint is astonishingly frank about that. My understanding is that trade secret theft requires some unlawful use to actually constitute an IP theft.

     set +o history
    steal secrets
    launch missiles
     set -o history

The closest English precedent for that was not the "jury of one's peers" but the "benefit of clergy".

"Most were surprised by how little he had taken in relation to the whole: eight megabytes in a platform that consisted of an estimated one gigabyte of code. "

Really? One gigabyte of code? Or one gigabyte of third party binaries, precompiled libraries, documentation etc.? All you need are the source files, and those language keywords compress nicely. Exactly how much plain text do you think they have?

' "But that’s the secret sauce, if there is one,” said the juror. “If you’re going to take something, take the strats.” '

Rubbish. The strats are the easy (or at least easier) bit in the high frequency space. That's like taking a dump of the html from facebook/twitter and thinking you've stolen their site. It's the scalable, distributed infrastructure that drives large websites, and it's the fastest possible infrastructure that drives HFT. One thing Lewis got right is the Coke/Pepsi analogy, and it really is that simple. A straight up race.

" “It’s way easier to start from scratch.” "

It is, and it isn't. In even the worst codebase there will be nuggets of usefulness that will accelerate the process. Isolated parts of the system that you've perfected. Exchange message decoders for example (ITCH, FIX/FAST, etc.). This is especially true when you, the genius who will be starting from scratch, has worked on it for a few years. I assume he was doing something useful in his time at Goldman.

I'm not buying the innocent act.

I agree; this was my point. Its frustrating to trot out a trite middlebrow dismissal for when we don't like something and to pretend it doesn't exist otherwise.