View Full Version : LW Dev question

12-27-2006, 04:19 PM
If LW is on the verge of a developmental renessaince with 9.2 and beyond, there is something I am curious about.

I have not used many any other applications for Character Animation, but it is something I want to start doing.

If LW is playing catch up in this area, how will it be implemented? Where do patents and intellectual property collide with respect to procedures inside something like a 3D app?

Because an app like Maya does it one way, does that mean LW will have to do it differently?

If I came out with a word processor today that allowed you to select text and then press a button with a bold 'B' on it...will I get sued?

12-27-2006, 04:32 PM
Spose it might depend on how it's done.. it would be infringement if they just stole some code from Maya and passed it off as their own (which of course they won't do). If you think about it, there's numerous programs that all do the same thing - but they're all written differently and nobody gets in trouble ...or do they?

You got me thinking, though, what about stuff like Catmull-Clark subdivision algorithms.. they were thought up by persons or people and are implemented in various programs. Does that mean Catmull gets a royalty from each program that uses his work? And what about Clark?

12-27-2006, 04:47 PM
If I were Newtek I'd be looking at how everyone else does things and stealing the best ideas from all of them. I don't think it's sensible to talk about things like character animation in isolation from everything else though- what's really needed is a tightly integrated app which is as relational as possible- so all the bits talk to each other in an intelligent way- which I'm sure is exactly what NT are working toward now.

12-27-2006, 04:55 PM
I didn't mean it to be an isolated question. The CA is more of an example.

The same could be asked about modelling tools.

Martin Adams
12-27-2006, 07:31 PM
I bet the patents certainly do collide just like any other company. Remember when Adobe were suing Macromedia and visa-versa (ironic ending). For what I understand, a patent must be a novel, useful and an innovative improvement. A patent must be non-obvious as well. So for example, the concept of patenting a walk cycle in general probably wouldn't pass, as it is obvious that the feet connect to the legs, the legs connect to the hip, etc. But the algorithm (non-obvious) code that can evaluate such a result could be patented (provided it invents something new - i.e. a new type of algorithm or mathematical solution). Based on that, there is nothing stopping other companies coming up with a different algorithm for achieving the same thing. Just like the different algorithms to render a sub patch model - they do the same, but in different ways.

At least that's my understanding.