Page 1 of 1
Derived Licensing Issues (LGPL)
Posted: Sun Oct 21, 2007 3:52 pm
by Alboin
I'm writing program X. However, I run into a small wall. I look to library Y (Which is LGPL.) for guidance. They use a mathematical formula. (let's say x_b-(a/a)+1) I discover said formula from their code, and use it in my code. I don't actually use any of their code, just the formula. (In fact, I do it in a very different way.)
Does this count as derived work? I don't believe it does, because for example, if it did, everything that decoded a PNG would be at the mercy of the license of libpng.
Am I right in my assumptions?
Thanks.
By the by, these are the pieces of code:
This is from SDL_ttf:
Code: Select all
rdiff = fg.r - bg.r;
gdiff = fg.g - bg.g;
bdiff = fg.b - bg.b;
for( index = 0; index < NUM_GRAYS; ++index ) {
palette->colors[index].r = bg.r + (index*rdiff) / (NUM_GRAYS-1);
palette->colors[index].g = bg.g + (index*gdiff) / (NUM_GRAYS-1);
palette->colors[index].b = bg.b + (index*bdiff) / (NUM_GRAYS-1);
}
And this is mine:
Code: Select all
SetForeground(Color(background.color.red+(b.image[_h][_w]*(foreground.color.red-background.color.red)/b.grays),
background.color.green+(b.image[_h][_w]*(foreground.color.green-background.color.green)/b.grays),
background.color.blue+(b.image[_h][_w]*(foreground.color.blue-background.color.blue)/b.grays)));
(Sorry for the formatting.)
Posted: Sun Oct 21, 2007 3:59 pm
by Brynet-Inc
I think you're safe, AFAIK it's legal to "look" at someones work and re-implement your findings...
Linux teams tend to look at BSD drivers for information... and that likely goes both ways.. Just don't copy & paste..
Not sure of the legality in the USA though.. damn software patents
Posted: Sun Oct 21, 2007 4:42 pm
by Alboin
Awesome....
Hello xlib and Freetype. (Without xft.
)
Posted: Sun Oct 21, 2007 4:54 pm
by SpooK
If you reimplemented a mathematical formula and the code is significantly/obviously unique, all you should worry about is what possible patents might be associated with that formula.
Posted: Sun Oct 21, 2007 5:38 pm
by Brynet-Inc
SpooK wrote:If you reimplemented a mathematical formula and the code is significantly/obviously unique, all you should worry about is what possible patents might be associated with that formula.
Please stop dragging American illusions into other cultures
Posted: Sun Oct 21, 2007 5:50 pm
by SpooK
Brynet-Inc wrote:SpooK wrote:If you reimplemented a mathematical formula and the code is significantly/obviously unique, all you should worry about is what possible patents might be associated with that formula.
Please stop dragging American illusions into other cultures
There are no illusions, as patents are a reality and he may be interested in not violating any that are enforced within his jurisdiction.
Posted: Sun Oct 21, 2007 5:55 pm
by Alboin
SpooK wrote:Brynet-Inc wrote:SpooK wrote:If you reimplemented a mathematical formula and the code is significantly/obviously unique, all you should worry about is what possible patents might be associated with that formula.
Please stop dragging American illusions into other cultures
There are no illusions, as patents are a reality and he may be interested in not violating any that are enforced within his jurisdiction.
No, no. I don't believe it's patented. (At least, according to the guys at Freetype.) Thanks though!
Posted: Sun Oct 21, 2007 11:07 pm
by Candy
SpooK wrote:Brynet-Inc wrote:SpooK wrote:If you reimplemented a mathematical formula and the code is significantly/obviously unique, all you should worry about is what possible patents might be associated with that formula.
Please stop dragging American illusions into other cultures
There are no illusions, as patents are a reality and he may be interested in not violating any that are enforced within his jurisdiction.
Software patents are no more than an illusion in most jurisdictions.