Back in February 19th, 2021, Take-Two sent DMCA notices to the reverse engineered projects re3 and reVC, (hosted on GitHub). After the takedowns, the project leaders filed a counter-claim, effectively restoring the projects and their whole repositories.

Today, September 2nd, 2021, Take-Two has taken legal action in the state of California, USA, claiming the projects and whole repositories are infringing on the copyrights of their games, Grad Theft Auto III & Grand Theft Auto: Vice City.
Not only is Take-Two suing the main repositories, but also against forks and derivative work (and the devs behind them) that sent the counter-claim back in February, like the PS Vita & Nintendo Switch forks.

The resolution of this case might be one of the biggest legal battles of recent years in terms of gaming, homebrew, hacking and reverse engineering, as other projects that have flourished under reverse engineered premises could be affected in the event of Take-Two winning the case.

Source

 

[smf]

Basically, the entire case hinges on how to separate the "functional" parts of the program from the "expression" part, and that is why the results of these kinds of cases are not easily predictable and are unlikely to be pursued.

In terms of a larger amount of code being copied for functional/interoperability reasons, I think this is a good comparison.

https://en.wikipedia.org/wiki/Apple_Computer,_Inc._v._Franklin_Computer_Corp.

There aren't that many cases because people generally aren't brazen to copy code wholesale and then try to defend themselves in court.

Companies abusing open source licenses is kind of similar, because they turn into copyright infringements if you break the license.

https://en.wikipedia.org/wiki/Open_source_license_litigation

And a lot of that software could be argued to be frameworks/functional pieces of code.

But there is a difference (legal term used: transformative). The reverse engineered stuff is not exactly the same (probably) and has been compiled for other systems. If you wanted to differentiate them further then you would have to do a side by side comparison with the original source.

The changes made either by a computer program or person when compiling source to binary and then back again are not transformative.

It would be like taking the text of a book and putting it through google translate, correcting any errors and then selling it while claiming it is your own work. It's now in a different language, but the authors copyright has still been violated.

A comparison with Sony v. Accolade might apply here. The Digital Station or whatever they called it performed roughly the same function as a real PS, yet the only thing Sony even went after them for was the BIOS (although you can't really decompile hardware, but that isn't the point here).

I'm not saying that it's illegal to create an engine that allows you to run the GTA games, which would be equivalent to your argument.

https://en.wikipedia.org/wiki/Sony_Computer_Entertainment,_Inc._v._Connectix_Corp.
While Connectix did disassemble and copy the Sony BIOS repeatedly over the course of reverse engineering, the final product of the Virtual Game Station contained no infringing material.

You're just not allowed to create your own engine by putting someone elses binary through a program which converts it to source code/pseudo source code and then distribute it, because that would contain infringing material.

The connectix court case went over the details of how the PlayStation was reverse engineered in great detail, I know one of the developers involved. If they had done what re3 showed in their live stream, then the decision would have been different.

 

[Zajumino]

In terms of a larger amount of code being copied for functional/interoperability reasons, I think this is a good comparison.

https://en.wikipedia.org/wiki/Apple_Computer,_Inc._v._Franklin_Computer_Corp.

The reason for apple's win was because there were other ways to achieve the same function without verbatim copying (it seems to have been a bit for bit copy too)

There aren't that many cases because people generally aren't brazen to copy code wholesale and then try to defend themselves in court.

I was referring to the fact that large companies do not want to risk a loss when there is not much to gain, especially when it could impact the entire industry.

And a lot of that software could be argued to be frameworks/functional pieces of code.

What is protected is the function and not the code itself. These kinds of things happen when people just copy-paste. It is obvious because everything is open source. It would be perfectly fine to look at the code and then write your own based on the techniques that you saw being used.

The changes involved when compiling source to binary and then back again are not transormative.

You may be right about that, but the end result does enable people to play GTA on other systems, which is similar to what Connectix did. The new binary provides roughly the same functionality, but did not copy any "expression" (we hope).

I'm not saying that it's illegal to create an engine that allows you to run the GTA games, which would be equivalent to your argument.

Pardon my bad analogies.
What I was trying to say was that just because you use something as a reference while you make something that is functionally similar doesn't necessary mean it is a duplicate. For example, many violins are based on the ones made by two Italian guys, and they all function the same way, but that doesn't mean were are all made using the exact same techniques or materials.
Basically, my point is that decompiling stuff and using that as a reference as to how it works is different than copy-pasting the original source code. It is definitely not an exact copy of the original, but some parts may or may not infringe on copyright. We just don't know for sure.

https://en.wikipedia.org/wiki/Sony_Computer_Entertainment,_Inc._v._Connectix_Corp.
While Connectix did disassemble and copy the Sony BIOS repeatedly over the course of reverse engineering, the final product of the Virtual Game Station contained no infringing material.

You're just not allowed to create that engine by putting the binary through a program which converts it to source code/pseudo source code and then distribute it as it would contain infringing material.

The thing is, at this point we don't even know which parts were taken from decompilation, and even then we would have to wait for the court to decide if they were "functional" or "expressive." Or whether decompilation was appropriate in the first place.

So do these GTA projects contain infringing material? The answer to this question lies in how the disassembly was used. For Connectix, it was okay. For this situation, all we can do is speculate until more facts are gathered or a court decision happens. Also notice how a lot of these cases get appealed.

 

[smf]

The reason for apple's win was because there were other ways to achieve the same function without verbatim copying (it seems to have been a bit for bit copy too)

It was identical, but changing some bits wouldn't have helped. They could have made a clean room clone, like Phoenix did of the IBM bios, but chose not to.

They won the original case because they managed to argue that a binary wasn't copyright, only a printed copy of the source code was. However the court found on appeal that this was an incorrect application of the law. The output from going from source to binary is still covered by the original copyright, so it's safe to assume that the translation from binary back to source is still covered.

Pardon my bad analogies.
What I was trying to say was that just because you use something as a reference while you make something that is functionally similar doesn't necessary mean it is a duplicate. For example, many violins are based on the ones made by two Italian guys, and they all function the same way, but that doesn't mean were are all made using the exact same techniques or materials.

Sure, you can learn how to make a watch by taking a rolex apart. But you can't then make a watch that looks like a rolex.

Pardon my bad analogies.
Basically, my point is that decompiling stuff and using that as a reference as to how it works is different than copy-pasting the original source code. It is definitely not an exact copy of the original, but some parts may or may not infringe on copyright. We just don't know for sure.

The live stream that aap did, shows him copy and pasting from the output of ida and that is what is in github.

Pardon my bad analogies.
The thing is, at this point we don't even know which parts were taken from decompilation

From what re3 have said publicly, it is 100% of it from decompliation. There will be a huge hurdle for them to overcome to argue that it was not.

Pardon my bad analogies., and even then we would have to wait for the court to decide if they were "functional" or "expressive."

It's impossible, after copying 100% of the code, to not have picked up something expressive.
It would change a lot of understanding of copyrighting of software, if the engine didn't include something expressive.

So do these GTA projects contain infringing material? The answer to this question lies in how the disassembly was used. For Connectix, it was okay. For this situation, all we can do is speculate until more facts are gathered or a court decision happens. Also notice how a lot of these cases get appealed.

There are the possible outcomes as I see it

1. there is no defense and instead T2 get a default judgement, github will be forced to remove all copies & if they can track down the developers world wide then they could go after them in their own countries (if they care enough).
2. re3 mount a defense based on fair use
3. re3 magic up evidence that shows it was a clean room reimplementation.

From the evidence re3 have posted publicly on github/twitter/youtube over the last couple of years, outcome 3 seems unlikely as everything shows that it is just copy and pasted out of ida (including the bugs).

outcome 2 will change how copyright applies to run time environments like java, libraries, etc. It's not really important that you can't run the game without extra files. The code itself is copyright, you aren't legally allowed to photo copy half of a book for example. What happens if someone now creates alternate assets for the GT3 engine?

A script for ida/ghidra that allowed you to create the source code locally from the original binary and then compile and run that is more likely to be seen as a fair use

I think outcome 1 is the most likely.

 

[Dungeonseeker]

The wrong choice was when re3 countered the DMCA take down. That was a get out of jail free card, T2 suing has no downside for them.

On or about April 8, 2021, Defendant Morra sent to GitHub a Section 512(g)(2)(b) counter notice, stating, under penalty of perjury, that the repository available at https://github.com/td512/re3 was removed and disabled by GitHub “as a result of a mistake or misidentification of the material to be removed or disabled.”

The repository wasn't removed because of a mistake and it wasn't removed because of a misidentification.



It is also established that re3 isn't clean room reverse engineering, they did it the illegal way.

Fair enough then, I was under the impression it was done cleanly. Just watched MVGs video and its pretty obvious they did it the dirty way.

The copyright claim is cut and shut, TT will win.

 

[scoobydude51]

Wow. The amount of people in denial here is insane.

the code was 100% infringed, even if they modified it a little bit.

 

[livid_hen]

Wow. The amount of people in denial here is insane.

the code was 100% infringed, even if they modified it a little bit.

What code was there to modify?

Just watched the MVG video... I hadent realized that this was made differently than the SM64 decomp! Thats kind of worrying.

 

[GaryOderNichts]

What code was there to modify?

Just watched the MVG video... I hadent realized that this was made differently than the SM64 decomp! Thats kind of worrying.

SM64-decomp was also done in the same way. They just didn't mention the reverse engineering tools on their repo.
The only thing that's different is that sm64 didn't have debug symbols.

 

[MK73DS]

SM64-decomp was also done in the same way. They just didn't mention the reverse engineering tools on their repo.

Prove it.

 

[GaryOderNichts]

Prove it.

Join their official discord and see for yourself.
They mostly used the disassembled assembly and not directly decompiled code. So that's a fair point I guess.
If that makes a big difference for court is debatable.
I'm also not sure if the function names can make a difference in terms of copyright. That would be interesting for re3.

(Getting a 100% binary accurate decompilation without using disassembly is almost impossible.)

 

[FAST6191]

[Super Mario 64 uses the same techniques as this did which saw a lawsuit filed]Prove it.

We could go into computer forensics if you want. However on balance of probability then they made a version of the code that compiles to byte for byte the original game, warts and all. Even if they had a 1000 strong team that did it full time since the original game was released back in the 90s the chances of that happening are so minute as to not even be worth considering. I would be impressed if they could do that for windows 3.11 minesweeper with the same provisos, never mind a complicated 3d world game for the N64 that has such random functionality as is described in the following sorts of videos



Two people write the same short story, song, poem, take the same photo... there could be some doubt. You get to ask for earlier copies, whether they crossed paths, whether one could have been influenced by another (same location around the same time), whether it is based off an earlier work of some form and is an obvious change... usual copyright case stuff.
Two groups write the same thousands and thousands of lines long code for a complicated game that compiles to something exactly the same as another. Forensics at this point is basically academic**.

If said thousand strong group had started back in the 90s, limited themselves to static non code analysis, maybe some cheat type memory viewing and the like then they might have made a functional clone that plays basically the same as far as anybody cares; there are what 30 moves really, a few dozen enemy behaviours you could map from video, including non linear acceleration, you could do some kind of unit test for camera to divine its behaviours (personally I would want to make them better, https://web.archive.org/web/2016051...loads/332-aim-assist-cameraReady-v8-final.pdf https://docs.google.com/document/d/1iNSQIyNpVGHeak6isbP6AHdHD50gs8MNXF1GCf08efg/pub?embedded=true , but maybe we are deliberately cloning and not straying), quarter frame position checks is a bit outside the box but not that unreasonable when playing with floats, 3d level layout will follow one of a handful of formulas* that look obvious when you know what to look for, you could probably track or trick any 3d effects to not showing up (see various graphics based cheats in all manner of N64 games from wireframe to no textures to whatever else) and calculate differences... even with all that then a functionally identical (which is radically different to byte for byte in something this complicated, indeed it would be impressive for tic tac toe/noughts and crosses) it would still be a massively impressive project. All the things covered just now would also be the general theme of a clean room project, and if it sounds horribly long winded and tedious then yes, yes it is.

*basic 3d model, model of "track" as you want to control behaviours without having to calculate 3d collisions all the time (more prominent in racing games as a concept hackers care about; see mario kart kcl files if you want a reasonable start in such fields), powerups/interactable objects, animated objects, enemy placement/spawners... might be all in one, might be 3 major ones as seen in a lot of 2d games, might be all five or spun into even more. Each of those is likely to be one of of the two major 3d model approaches (bones and vertices, the victor tending to be picked based on the hardware the system uses that you could reasonably know) or some flavour of coordinate system.

**though actually as 99% of stuff here we ever see in court cases is "you used this open source project and we can prove it from the code", "you used this leaked code/specification and we can prove it", "you copied these lines from this as it has the same bugs*** we had that nobody else would have programmed in" and "you took this code when you left the company and we can prove it" then it might actually be somewhat interesting to ponder.

***concept going back to maps https://bigthink.com/strange-maps/670-nil-how-to-hide-an-elephant-the-1923-gold-coast-survey video on similar matters, the telephone directory cases are another popular one for this https://supreme.justia.com/cases/federal/us/499/340/ , and seen somewhat recently when it was shown google was scraping lyric websites https://www.masslawblog.com/contracts/does-genius-have-an-illegal-scraping-case-against-google/

 

[MK73DS]

Join their official discord and see for yourself.
They mostly used the disassembled assembly and not directly decompiled code. So that's a fair point I guess.
If that makes a big difference for court is debatable.
I'm also not sure if the function names can make a difference in terms of copyright. That would be interesting for re3.

(Getting a 100% binary accurate decompilation without using disassembly is almost impossible.)

We could go into computer forensics if you want. However on balance of probability then they made a version of the code that compiles to byte for byte the original game, warts and all. Even if they had a 1000 strong team that did it full time since the original game was released back in the 90s the chances of that happening are so minute as to not even be worth considering. I would be impressed if they could do that for windows 3.11 minesweeper with the same provisos, never mind a complicated 3d world game for the N64 that has such random functionality as is described in the following sorts of videos



Two people write the same short story, song, poem, take the same photo... there could be some doubt. You get to ask for earlier copies, whether they crossed paths, whether one could have been influenced by another (same location around the same time), whether it is based off an earlier work of some form and is an obvious change... usual copyright case stuff.
Two groups write the same thousands and thousands of lines long code for a complicated game that compiles to something exactly the same as another. Forensics at this point is basically academic**.

If said thousand strong group had started back in the 90s, limited themselves to static non code analysis, maybe some cheat type memory viewing and the like then they might have made a functional clone that plays basically the same as far as anybody cares; there are what 30 moves really, a few dozen enemy behaviours you could map from video, including non linear acceleration, you could do some kind of unit test for camera to divine its behaviours (personally I would want to make them better, https://web.archive.org/web/2016051...loads/332-aim-assist-cameraReady-v8-final.pdf https://docs.google.com/document/d/1iNSQIyNpVGHeak6isbP6AHdHD50gs8MNXF1GCf08efg/pub?embedded=true , but maybe we are deliberately cloning and not straying), quarter frame position checks is a bit outside the box but not that unreasonable when playing with floats, 3d level layout will follow one of a handful of formulas* that look obvious when you know what to look for, you could probably track or trick any 3d effects to not showing up (see various graphics based cheats in all manner of N64 games from wireframe to no textures to whatever else) and calculate differences... even with all that then a functionally identical (which is radically different to byte for byte in something this complicated, indeed it would be impressive for tic tac toe/noughts and crosses) it would still be a massively impressive project. All the things covered just now would also be the general theme of a clean room project, and if it sounds horribly long winded and tedious then yes, yes it is.

*basic 3d model, model of "track" as you want to control behaviours without having to calculate 3d collisions all the time (more prominent in racing games as a concept hackers care about; see mario kart kcl files if you want a reasonable start in such fields), powerups/interactable objects, animated objects, enemy placement/spawners... might be all in one, might be 3 major ones as seen in a lot of 2d games, might be all five or spun into even more. Each of those is likely to be one of of the two major 3d model approaches (bones and vertices, the victor tending to be picked based on the hardware the system uses that you could reasonably know) or some flavour of coordinate system.

**though actually as 99% of stuff here we ever see in court cases is "you used this open source project and we can prove it from the code", "you used this leaked code/specification and we can prove it", "you copied these lines from this as it has the same bugs*** we had that nobody else would have programmed in" and "you took this code when you left the company and we can prove it" then it might actually be somewhat interesting to ponder.

***concept going back to maps https://bigthink.com/strange-maps/670-nil-how-to-hide-an-elephant-the-1923-gold-coast-survey video on similar matters, the telephone directory cases are another popular one for this https://supreme.justia.com/cases/federal/us/499/340/ , and seen somewhat recently when it was shown google was scraping lyric websites https://www.masslawblog.com/contracts/does-genius-have-an-illegal-scraping-case-against-google/

I don't see why having a byte to byte compiled ROM is a proof of them using the decompiled code.
As far as I know, it is legal to decompile the ROM, document it and write code with help of this documentation. And yes, in a vacuum I agree that having a byte to byte perfect copy is miraculous. But here we aren't in a vacuum : they can compare their builds and the original ones and tweak their code until they match, without never **seeing** the decompiled code.
The poem analogy isn't that great because you can't check if your poem is the same as another one without looking at its "source code". Here you can compare your build with the official build without having access to the decompiled source code, which makes it legal as far as I know.

I can't be sure they didn't used directly decompiled source code but you can't prove they did either with this argument.

Then, everything follows : a byte to byte perfect copy obviously behaves exactly the same as the original one.

Also, I'd like to add that they do NOT have the same thousands and thousands lines of code. Different code can produce the same output after compilation. Their work is very impressive and the fact that even Nintendo can't take it down (Nintendo!!) is a strong argument towards the legality of their approach, though it isn't a proof of them never using directly the decompiled source code.

 

[FAST6191]

If this is a civil suit then you have the standard is not the beyond all reasonable doubt of criminal cases (percentages are tricky but call it 99%) and instead it is more like 51%.

Similarly
"it is legal to decompile the ROM, document it and write code with help of this documentation"
Not in any court case or sensible reading of various copyright laws (intent, previous interpretation or spirit) I have ever seen. It is also a textbook case of derived work, which is something copyright law is quite concerned with. Doing so would similarly render copyright of code basically pointless in about 5 seconds, or the 5 months it takes to train up a new crop of assembly programmers and hackers from your existing stock of programmers.

If then the only means they have to legally create something is the tedious as you like clean room method I described before then them having 1:1 code means they copied it from somewhere (leak, disassembly and manual creation, decompilation using fancy tools, decompilation using an export of the dynamic recompilation of emulators... it matters little as far as the clean hands idea in law is concerned, though I suppose technically that makes you right and I can't be sure they did not somehow have access to the Mario 64 late beta leak and use that rather than decompiled code by dint of 1:1 existence, such a thing would require comparisons of things there and an inference based on the similarities or lack thereof).

 

[livid_hen]

SM64-decomp was also done in the same way. They just didn't mention the reverse engineering tools on their repo.
The only thing that's different is that sm64 didn't have debug symbols.

I'm not saying they can't get sued for it, but the sm64 one was a clean room project. One group de-compiles the code, and writes a document describing the requirements to get the files to run, then another team builds the game from the ground up based on those listed requirements. Then the one that was built from the document gets distributed.

 

[chrisrlink]

this is why corperate takeovers (I don't even call them partnerships) 99% of the time always bad for the original IP holder because their opinion means jack sh*t think of how rockstar origionally view PC mods (even after the "hot coffee"incident even though their view soured about it they still cared very little about mods same with Gamefreak's view on Pokemon fan games) but it means jack shit to their masters they just want them to be the obedient dogs they should be, sad part is they can't break away either the should have read the contracts before signing. now I don't know about Japanese business law but if it's anything like their copyright laws well their stuck in a bad marriage without the option to divorce

 

[FAST6191]

I'm not saying they can't get sued for it, but the sm64 one was a clean room project. One group de-compiles the code, and writes a document describing the requirements to get the files to run, then another team builds the game from the ground up based on those listed requirements. Then the one that was built from the document gets distributed.

A) That is not clean room reverse engineering. Still derived directly from the code even if you have another party implement it afterwards.
B) Why are you so sure that was the method used? There was no real mention of that in any of the documentation I saw, though I was more interested in the code so guess I might have missed that (not that it means anything if it did happen that way).

 

[CoolMe]

What's Rockstar stance in this? Or they would go with whatever makes more money for them in this case i guess.. Is this legal BS just because the remasters are on the horizon, to prove some kind of point?

 

[GaryOderNichts]

I don't see why having a byte to byte compiled ROM is a proof of them using the decompiled code.
As far as I know, it is legal to decompile the ROM, document it and write code with help of this documentation. And yes, in a vacuum I agree that having a byte to byte perfect copy is miraculous. But here we aren't in a vacuum : they can compare their builds and the original ones and tweak their code until they match, without never **seeing** the decompiled code.
The poem analogy isn't that great because you can't check if your poem is the same as another one without looking at its "source code". Here you can compare your build with the official build without having access to the decompiled source code, which makes it legal as far as I know.

I can't be sure they didn't used directly decompiled source code but you can't prove they did either with this argument.

Then, everything follows : a byte to byte perfect copy obviously behaves exactly the same as the original one.

Also, I'd like to add that they do NOT have the same thousands and thousands lines of code. Different code can produce the same output after compilation. Their work is very impressive and the fact that even Nintendo can't take it down (Nintendo!!) is a strong argument towards the legality of their approach, though it isn't a proof of them never using directly the decompiled source code.

I'm not saying they can't get sued for it, but the sm64 one was a clean room project. One group de-compiles the code, and writes a document describing the requirements to get the files to run, then another team builds the game from the ground up based on those listed requirements. Then the one that was built from the document gets distributed.

I just checked the first commit on the official repository.
It still contains directly disassembled code with no labels and memory addresses for functions. This is 1:1 code from the official game.
Even containing file offsets and the original instructions as hex.
https://github.com/n64decomp/sm64/tree/89e86908578d2fdc424f9f47d91b1d82948eaa59/asm/non_matchings
These functions were later translated into C as well.

 

[Zajumino]

It was identical, but changing some bits wouldn't have helped. They could have made a clean room clone, like Phoenix did of the IBM bios, but chose not to.

They won the original case because they managed to argue that a binary wasn't copyright, only a printed copy of the source code was. However the court found on appeal that this was an incorrect application of the law. The output from going from source to binary is still covered by the original copyright, so it's safe to assume that the translation from binary back to source is still covered.

Everything here is true except for the last sentence, which I am not sure about. It might be possible that it only works one way, because the reverse engineered source code is different from the original, and when it is compiled, the binaries are also different. There may be other court cases that clarify this, but as a guess, perhaps the source and corresponding binaries are treated as a set.

Sure, you can learn how to make a watch by taking a rolex apart. But you can't then make a watch that looks like a rolex.

It is my understanding that the appearance of a rolex is protected under copyright regardless of if you take one apart, since it has no influence on the function of the watch.

The live stream that aap did, shows him copy and pasting from the output of ida and that is what is in github.

This is supposed to be a valid way of studying how the game works, but based on other things you have said, they haven't done much else. Theoretically, if they did not declare their project to be completed, maybe they could have kept their code available as an intermediate copy (assuming you are allowed to have those public). Perhaps this is why T2 waited so long to do anything.

It's impossible, after copying 100% of the code, to not have picked up something expressive.
It would change a lot of understanding of copyrighting of software, if the engine didn't include something expressive.

It really depends on the nature of the original work. In Sony v. Connnectix, the BIOS was considered to be purely functional, and the court did not find any infringing material in the clone.
In this case, the engine probably has expressive elements, but it does not matter if they do not end up in the final product. The ones that do could be passed off as some of the exceptions to copyright. If the amount that remains is small in significance it might be overlooked.

Basically, it depends on what the court finds. However, in this case, given the degree of similarity to the original, I think they would find enough to disqualify fair use.

There are the possible outcomes as I see it

1. there is no defense and instead T2 get a default judgement, github will be forced to remove all copies & if they can track down the developers world wide then they could go after them in their own countries (if they care enough).
2. re3 mount a defense based on fair use
3. re3 magic up evidence that shows it was a clean room reimplementation.

From the evidence re3 have posted publicly on github/twitter/youtube over the last couple of years, outcome 3 seems unlikely as everything shows that it is just copy and pasted out of ida (including the bugs).

I think outcome 1 is the most likely.

Because they decompiled the original, fair use has always been their only chance, but things do not look favorable for them.

The point of clean room is to be able to go to court and say it was impossible to have had copyrighted material in your work. Referencing a decompilation defeats the purpose of clean room (for those doing the recreation).

--------------------- MERGED ---------------------------

A) That is not clean room reverse engineering. Still derived directly from the code even if you have another party implement it afterwards.

The key is to have a lawyer check to see if there was no copyrighted material in the documentation.

 

[Zajumino]

Join their official discord and see for yourself.
They mostly used the disassembled assembly and not directly decompiled code. So that's a fair point I guess.
If that makes a big difference for court is debatable.
I'm also not sure if the function names can make a difference in terms of copyright. That would be interesting for re3.

(Getting a 100% binary accurate decompilation without using disassembly is almost impossible.)

Disassembled assembly likely implies that the original was only used in order to study its functional aspects, which are not covered under copyright.

Having function names of the original source code would imply that it was looked at and directly copied from (or it could be a coincidence). Otherwise, function names are automatically generated by the decompilation process or added in afterwards. With regards to copyright, function names would typically be used as evidence of infringing code, but are otherwise largely irrelevant. Even from a programmer's perspective, the only purpose of function names is to make code easy to work with.

 

[duwen]

Seems like they don't give a fuck that Bully was also RE'd.