Originally posted by jacob
View Post
You would need a prototype "translation machine" to get a patent (so it's not just an abstract idea), you can use whatever methods to translate/record/play underneath. A working prototype is absolutely necessary for a patent, or it'll be thrown out in court afterwards even if granted. Like you said though, once you know the math/science of the MP3 codec, implementing it in a language is an insignificant amount of work, making such patents primarily patents of abstract ideas. It's bad for innovation in the long and short term to have such patents that are constantly being used to offensively sue and shakedown someone because the subject matter of the patent is either (copyright) math/science or trivial coding.
edit: a patent for a translation machine would need to do record/process/translate/play in that order and with purpose. This is relatively non-trivial to do right even if I gave you a microphone/rpi/speaker and you had a EE/CS degree and all you had to do was translate someone saying hello to hola. The amount of work I'm doing and the primary subject matter of the patent is real engineering, not coding or math/science. Process patents in chemical or mechanical engineering make sense in the same way. Software patents just don't.
Leave a comment: