|
Authored by: mschmitz on Friday, April 20 2012 @ 01:38 AM EDT |
I'm fully aware that inside a (digital) computer, everything is ones and zeroes.
So I should not have said 'binary code' but object code, bytecode, assembly code
or whatever is applicable to the language at hand. See how it gets a bit fuzzy
when I try to be more precise?
With the right tools everything is human readable, I don't deny it. objdump -d
has been my friend many a time when chasing Linux kernel bugs. hexedit has been
useful to 'binary patch' programs that I did not want to recompile.
The distinction 'text' vs. 'binary' files has long been used to good effect,
without getting things mixed up too badly. Probably because everyone initially
understood that even text is always represented in binary format. I don't think
we need to worry about that too much. It's a good enough metaphor.
Copyrightability on ones and zeroes - the concept of 'one' and 'zero' sure
cannot be copyrighted. A particular sequence of ones and zeroes evidently can -
as far as it represents copyrightable text, music, video, data, ...
It's a particular creative work that is copyrighted, not its various
representations. API library, yes. API definition, as in 'a list of facts or
conditions required to get the library to perform the intended function', I
don't think so.
-- mschmitz
(IANAL, in case it wasn't clear)[ Reply to This | Parent | # ]
|
|
|
|
|