Oracle v. Google - Silnith’s Lair
May. 15th, 2016
02:34 am - Oracle v. Google
The copyright case being waged between Oracle and Google right now is possibly the most critical court case to affect general-purpose computation that has ever happened. It cuts to the core aspect of software. Basically, can you use copyright law in order to apply patent-level prohibitions on what computations other people can perform.
The question is so fundamental that trying to explain its importance is like trying to explain the importance of clean air to breathe. Without it, everything stops. However, right now the case is being heard and tried by people who apparently do not understand what makes mammals function.
I have come up with a much better analogy to what an API is than the lame examples used in the article and apparently in the courtroom. A programming language is a language. It defines how we articulate thought and reason. An API is a lexicon, a vocabulary. A lexicon is not the language, but a language is useless without a lexicon. Expressions in a language are composed of invocations of the lexicon.
A lexicon cannot be restricted by copyright. It creates the semantic meaning that language is there to communicate. If people are restricted from using the lexicon, the language might as well not exist, its entire purpose is rendered void.
If Oracle prevails, then essentially the courts are saying that a company can prohibit others from using words from their lexicon. Note that this is not like a dictionary. A dictionary provides definitions for a lexicon, but it is not the lexicon itself. Many people can write dictionaries. These dictionaries will be different. But they will be very similar, because they are expressing the semantics of the lexicon. A dictionary can be covered by copyright. But the vocabulary itself cannot be. Any other person can come along and write their own dictionary that offers definitions of all the words in the vocabulary, and Merriam-Webster can do nothing to stop them. Anybody can come along and invoke the words from a dictionary, and there is nothing Merriam-Webster can do to stop them.
This is how fundamental the question of copyright applying to APIs really is. How insanely convoluted would communication be if languages were never allowed to inherit or appropriate words from other languages? What if every country that shared a language had to invent its own vocabulary completely different from every other country's vocabulary? The idea is laughable on its face.
Nevertheless, right now that very question is being decided in a courtroom, and it is going to be decided by people who do not understand what the word
lexicon even means. And people who do not want to know, who denigrate the very idea that they should be cognizant of the concept.
(The saddest part of this for me is that I hate what Google did with Android and the Android RunTime. I think it was a despicable, slimy thing for them to do, and if not for that then Sun might very well still exist today. Despite all that, the case Oracle brought threatens to destroy the entire industry as collateral damage, and that is not an acceptable cost. Please consider that every other industry on Earth now depends on computing in some way. Do you really want to go back to a time when banks employed thousands of people to do interest calculations instead of having a machine to do it automatically? A world where no factories are automated and every single stitch of clothing, every morsel of food, every inch of metal or fiber is fashioned by a human hand? I sure as hell do not.)
(While I personally would not shed a tear if the stock market went back to paper-only, the world financial systems would collapse without the constant churn of high-frequency trading, according to bankers and economists.)