Is reverse-engineered code submittable?

[Gabriel Pulido]

I'm interested in reverse engineering a closed-source algorithm (DURILCA, which as I understand it is PPM with enwik-specific improvements) to see how it could improve performance past what has been achieved by fx-cmix. I plan to create my own C++ source with comments to emulate the design of this algorithm. However, because the actual algorithm isn't my design, would such an effort be ineligible for the hutter prize?

Thanks,

Gabriel Pulido

[James Bowery]

It is up to you to comply with whatever intellectual property rights may obtain.  The Hutter Prize judging process is on objective criteria such as the license under which the entrant releases the source code.  If you fraudulently release code under an OSI license, it is not up to the Hutter Prize judging committee to make that determination, but to the relevant jurisdictions.

[Gabriel Pulido]

Hi, thanks for your response.

So, the judging process does not attempt to determine whether the code is entirely your own intellectual property, and instead is based on objective factors like the license used and the rest of the rules for participation. As such I could potentially submit an eligible entry with code that is not strictly my own intellectual property, but may face legal consequences for doing so. Is that correct?

[James Bowery]

Subject to this catch-all proviso which is available at the "More Information" URL:

• The prize will be paid if the solution reflects the spirit of the contest. In particular decompressors (secretely) receiving any kind of "outside" information are forbidden. Also in order to verify your claim we need to be able to run your executable on our machines. Payment of the prize cannot be legally enforced. Marcus Hutter will make the final decision whether to recognize a record, award a prize, and the amount.
• Rules may change at any time without notice to meet the goals of fairness, accuracy, fostering progress and public participation, and recognizing existing practice.

For instance, I could easily imagine that if your C++ code even looks like it amounts to a decompilation program's output of a binary, (DURILCA or not) that the entry would be rejected on one or more of the above subjective criteria.  Also, if someone comes along and specifically says, "I'm going to violate someone's intellectual property to to win the Hutter Prize," you can pretty well rest assured that the entry will not be accepted. 

The purpose of the making the judging rules objective is:

1) to assure the best talents that there is no room for favoritism 
2) to making judging as automated as is practicable
3) to inspire the use of the most principled information criterion (algorithmic information approximation) for world model selection relative to a given selection of data (e.g. language data such as enwik9).

It is not to so-restrain the judging committee that people acting in bad faith can still win.

[Jan Mercl]

On Sat, Dec 30, 2023 at 9:52 PM Gabriel Pulido <ung...@ungato.moe> wrote:

> So, the judging process does not attempt to determine whether the code is entirely your own intellectual property, and instead is based on objective factors like the license used and the rest of the rules for participation. As such I could potentially submit an eligible entry with code that is not strictly my own intellectual property, but may face legal consequences for doing so. Is that correct?

I am not a lawyer. Yet I think you should not pursue the path you're
discussing. Copyright infringement is heavily punished in many parts
of the world. Doing so for profit, even for just a potential one, can
make the consequences even harsher. Moreover, I would not be surprised
at all if the sentence would also include seizure of any prize money
obtained via illegal use of copyrighted code.

tl;dr: It is probably a bad idea. Even when ignoring the moral aspect.

[Gabriel Pulido]

I see, so if the judges feel subjectively that protecting intellectual property is more valuable than advancing research and understanding of lossless compression, then no prize would be awarded.

Excuse the loaded wording, but I think we both have strong opinions on "intellectual property," and whether a violation of it would be in bad faith or not. Let's just agree to disagree on that.

Thanks,

Gabriel Pulido

[James Bowery]

It's only common sense:  If someone says they're going to steal something, it puts those to whom they say it at legal risk -- especially if it is phrased in terms of asking for some sort of permission to do so subject to incentives.  Insofar as your "loaded wording", it has more to do with the common sense legal risk than any kind of ideology.

[Michael Kass]

I am not a lawyer, but please distinguish two kinds of IP:  copyright and patent.  Copyright does not protect ideas.  It protects the specific expression of those ideas, for example in a particular piece of code.  Reverse engineering, in general, is not a violation of copyright.

If there are no patents on DURICLA, then I would expect there are no legal problems with reverse engineering it.

Everyone builds on everyone else’s ideas.  Giving proper credit to the sources of your ideas keeps you in good moral position.  Avoiding patented ideas and creating your own expression of the ideas you use keeps you in good legal position.

—Michael

--
You received this message because you are subscribed to the Google Groups "Hutter Prize" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hutter-prize...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hutter-prize/82cf7c96-9b98-47c0-a6cd-5c3553739284n%40googlegroups.com.

[James Bowery]

On Sun, Dec 31, 2023 at 7:51 AM Michael Kass <michae...@gmail.com> wrote:

... Reverse engineering, in general, is not a violation of copyright...

And that was the basis for my original response to his question.