(Admittably, I have not thought the international problems that this might cause)
(Admittably, I have not thought the international problems that this might cause)
...or it gets bought by some more powerful entity - capable of paying the tax - shortly before this happens. Leading to concentration of copyrights in the hands of big corporations, to an extent even larger than under current system.
The "doubling after that each year" provision of the original comment neatly takes care of that. After a relatively small number of doublings the size of the tax for the single item would be larger than the current US national debt.
The 'math' term for this is "exponential growth".
Sounds great to me.
My suggestion is that if the copyright holder fails to pay the tax, it is automatically considered to be licensed under the latest version of the GPL.
Edit: In some ways this is actually an improvement over the status quo. It solves the problem where software can't be relicensed because no one can track down all the authors. You only have to track down the authors who have been active in the last two years.
> In some ways this is actually an improvement over the status quo. It solves the problem where software can't be relicensed because no one can track down all the authors.
I wouldn't really call that an improvement. There's no longer even a point in relicensing code, given that the license only matters for code less than two years old.
They're not. The copyleft portion of the GPL prevents improvements to open source code being locked up. In the case of source code, simply putting it in the public domain is of little use if the source code isn't public.
e.g., Disney won't stich together all their movies in a year-long pseudo unit of work, because, if it did so, a two-hour clip from that work would fall under fair use.
An initial cost of 1000 USD, doubled every year, for 20 years is:
1000 * (2^20) (where ^ means exponentiation).
Or 1,048,576,000 USD - a wee bit over 1 billion dollars for the 20th year.