Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John Carmack on expert witnesses and 'non literal' copying

facebook.com

Post reply on HN
Print view View original

11–20 of 322 posts

Re: John Carmack on expert witnesses and 'non literal' copying

#11

Post by gravypod »

trendia wrote: post #5

This struck out at me: > There are objective measures of code similarity that can be quoted, like the edit distance between abstract syntax trees ... If this became the primary legal metric, then programmers who stole code would change the code so that functions achieved the same output with practically no AST similarity. That is, they could maximize functional similarity while minimzing code similarity. This would b…

Funnily enough I actually implemented this [0] two semesters ago while working as a Teacher's Assistant for the into CS class.

It worked fairly well. I basically took the levenshtein algorithm and mapped it directly to ASTs and used that to find the % of similar elements in the code. It worked a few times and I caught 1 or 2 groups of cheaters.

....sadly in that class you were allowed to "work together" you just had to include a note saying you did so I've got no "kill count" anymore and I'm not planning on TAing any time soon (at least not for python).

Also that code was hacked together in a weekend and that anti-cheet was done in 2-3 hr. It's not production ready by a longshot.

[0] - https://github.com/gravypod/GradeO/blob/master/libs/cheating...

Re: John Carmack on expert witnesses and 'non literal' copying

#12

Post by pyb »

"The expert witness circuit is surely tempting for many academics, since a distinguished expert can get paid $600+ an hour to prepare a weighty report that supports a lawyer’s case. I don’t have any issue with that, but testifying in court as an expert should be as much a part of your permanent public record as the journal papers you publish. In many cases, the consequences are significant. There should be a danger to your reputation if you are imprudent."

So basically he's kind of threatening the expert witness now?

Re: John Carmack on expert witnesses and 'non literal' copying

#13

Post by gravypod »

rocky1138 wrote: post #6
trendia wrote: post #5

This struck out at me: > There are objective measures of code similarity that can be quoted, like the edit distance between abstract syntax trees ... If this became the primary legal metric, then programmers who stole code would change the code so that functions achieved the same output with practically no AST similarity. That is, they could maximize functional similarity while minimzing code similarity. This would b…

Someone should write a script which takes any code and mucks the AST.

You should be able to just compile it with -O3 and then decompile it and rename all your variables back. You'll get a very wonky source output but it will be extremely different.

Re: John Carmack on expert witnesses and 'non literal' copying

#14

Post by whatever_dude »

squizzel wrote: post #7

I often wonder if John was involved with the story telling aspect of his games, not just the code.

I think it has a recurring theme that any time he got involved, things didn't go well so he decided to let go of it.

That's not to say he didn't have influence; many times the games' stories were derived FROM the tech, not the other way around. Doom was originally supposed to be much more story- and character-driven, but that took a backseat given the focus on the development side was on the gameplay mechanics.

Also, this quote:

> Story in a game is like a story in a porn movie. It's expected to be there, but it's not that important

From "Masters of Doom".

Re: John Carmack on expert witnesses and 'non literal' copying

#15

Post by lmm »

addisonj wrote: post #10

I don't have the evidence so I can't make a judgement about whether Zenimax or Carmack is in the wrong here, but this does point out something strange in our court system and that is where a "jury of peers" which translates to normal people with little to no knowledge of the underlying subject matter. As our society becomes more specialized it seems a bit absurd to have people make judgements mostly based on how well…

How is the court supposed to judge familiarity? What if jury members think they're technical experts, but aren't?

There are always going to be complex technical matters that are relevant to lawsuits - this was the case before software even existed - and the way the system is supposed to work is that rather than the jury bringing their own set of preconceptions on the technical aspects, the expert witnesses answer the technical questions in the court (where they are subject to cross-examination etc.) and the jury makes the factual judgements that have always been their job.

Re: John Carmack on expert witnesses and 'non literal' copying

#16

Post by smacktoward »

addisonj wrote: post #10

I don't have the evidence so I can't make a judgement about whether Zenimax or Carmack is in the wrong here, but this does point out something strange in our court system and that is where a "jury of peers" which translates to normal people with little to no knowledge of the underlying subject matter. As our society becomes more specialized it seems a bit absurd to have people make judgements mostly based on how well…

The more familiar you are with the industry, the less likely you are to be impartial.

Re: John Carmack on expert witnesses and 'non literal' copying

#17

Post by pavel_lishin »

addisonj wrote: post #10

I don't have the evidence so I can't make a judgement about whether Zenimax or Carmack is in the wrong here, but this does point out something strange in our court system and that is where a "jury of peers" which translates to normal people with little to no knowledge of the underlying subject matter. As our society becomes more specialized it seems a bit absurd to have people make judgements mostly based on how well…

> this does point out something strange in our court system and that is where a "jury of peers" which translates to normal people with little to no knowledge of the underlying subject matter.

Nothing in the constitution says anything about peers, and the fact that most juries are comprised of people who don't know much about the subject matter comes from lawyers wanting those people on the bench, since they're less likely to have pre-conceived notions - possibly notions conflicting with their clients' goals.

If you're trying to convince someone of something, it helps if they start with a blank slate.

Re: John Carmack on expert witnesses and 'non literal' copying

#18

Post by ChuckMcM »

Interesting rant on expert witnesses. Mostly interesting because I've done some expert witness testimony and found it fascinating.

I completely resonate with John's issue with the expert's testimony and his understanding of it. John mentioned that their own expert testified differently but he didn't say whether or not their lawyers attempted to impeach their expert with cross examination. In the two cases I participated in, both sides agreed ahead of time on how to "handle" experts (in terms of cross examination). In one it was documents only (no testimony) and in one I was deposed by the opposing counsel after documents in which the lawyer clearly had some notes (presumably from their expert(s)) and was trying to get me to recant or change some of my points (I didn't need to).

In a jury trial I can imagine that having the jury understand your testimony is probably the most challenging.

And, like John, I consider the notion of 'non-literal copying' to be pretty ridiculous. In the limit it means "you read this code, understood how it worked, and wrote new code that could do the same function." You can stretch that to cover anything you have ever seen. Which is sad.

Re: John Carmack on expert witnesses and 'non literal' copying

#19

Post by sgift »

trendia wrote: post #5

This struck out at me: > There are objective measures of code similarity that can be quoted, like the edit distance between abstract syntax trees ... If this became the primary legal metric, then programmers who stole code would change the code so that functions achieved the same output with practically no AST similarity. That is, they could maximize functional similarity while minimzing code similarity. This would b…

> But, while it's possible to game any 'objective' metric like AST distance, maybe that's preferable to the subjective and incentivized claims of an expert witness, who will be biased in favor of the party paying them. And while it's easy to get lost in the paragraphs of expert witness testimony, a collection of objective metrics is easy to compare to previous cases.

That's the important part. Sure, you can game an objective metric. The moment you use any metric you risk gaming, but at least you have something which is not "well ... I thought so. And I'm certainly not biased because I get paid by one side. Never ever."

Post reply on HN