There is a health app that is very important to sufferers of a specific disease and it is freemium and I find this detestable. I am doing a clean room rewrite of it in its entirety and plan to release it on fdroid. Is there a license that prevents some ghoul from forking it to profit from people’s misfortunes? I am trying to avoid legal issues by being clean room but will try to keep my publishing of the app as anonymous as possible regardless. Thanks
Yeah, for sure there’s a lot of complexity in evaluating the specific case at issue.
The “the only way to gain access to the ideas and functional elements” bit kindof goes back to the legal theory that copyright gives protections to “expression”, not “ideas”. (For instance, Baker v. Selden. To oversimplify, Selden came up with an original way to do bookkeeping and wrote a book about it, Baker learned the method from Selden’s book and wrote his own book about the same method, and Selden sued Baker. The courts said that since all Baker “copied” was the “ideas” and not the “expression”, there was no copyright infringement, so Selden didn’t get any damages or anything.) And “functional elements” basically count as “ideas” more than “expression”. So, they’re kindof saying that (again, oversimplifying, IANAL, seek legal council, etc) if your method of keeping people from understanding your algorithms is just the difficulty barrier involved in decompiling, then copyright won’t say that decompiling isn’t allowed. Which is still open to a lot of interpretation, don’t get me wrong.
The “legitimate reason for seeking such access” in this case was understanding the Sega Genesis scheme for preventing unlicensed games from working on their systems and the algorithms involved. Accolade did that specifically in order to defeat Sega’s “DRM” and create games that would run on the Genesis.
Of course, that was before the DMCA, which… who knows how Genesis v. Accolade would have gone had the DMCA “circumvention” thing been in the mix. It probably would have to have been considered, but who knows if it would have changed the actual ruling.
Anyway. Yeah. This stuff fascinates me.
My knowledge basically comes from reading through that whole IP caselaw textbook that I linked in my previous comment. But it’s cool to have a little insight into some of the legal theory involved in such things as copyright licenses and such.
IANAL either, but even under that interpretation (which I think it’s similar to what I said), I see it can be problematic still.
It implies that one would need to demonstrate in court that “your method of keeping people from understanding your algorithms is just the difficulty barrier involved in decompiling”. Because if it is not (lets say… you are openly offering documentation / help / manuals for your software describing the ideas, without detailing how those ideas are expressed in the software), then you might be able to successfully make the case that decompiling the protected “expression” of the ideas was not necessary to access those ideas. And so it might no longer be true that “copyright won’t say that decompiling isn’t allowed”.
Yeah, I don’t mean to imply otherwise. My point was basically just that decompilation qualifies as fair use in some cases.
Sure. I could see some counterarguments that might be persuasive, though.
(I could also see the argument, however, that the first ruling was in error because automated decompiling makes a copy in the same way as translating a novel to another human language makes a copy. So, yeah. Definitely dicey territory legally. But may main point is just that it’s definitely incorrect to blanket statement that “decompiling is copyright infringement”. And the fact that it was found to be fair use even in one specific circumstance really makes it feel like a neat legal loophole. Just like how recipes or fashion designs or car designs or board game rules aren’t covered by copyright.)
All that said, if the documentation is so comprehensive that decompiling the code isn’t necessary, then… why would one want/need to decompile it in the first place. I’d imagine anyone who wanted to make a reimplementation could just… reimplement it from the documentation.
I’m guessing that’s kinda how GNU Octave came about. It’s a mostly-compatible clone of the proprietary “Matlab” application. (Matlab is basically a programming language.) There was probably good enough documentation out there about Matlab that the people behind GNU Octave could just build a replacement without needing to decompile the existing application.