LWN.net Logo

Software is The Glass Bead Game

Software is The Glass Bead Game

Posted Feb 1, 2012 13:03 UTC (Wed) by java_developer (guest, #82469)
In reply to: Software is The Glass Bead Game by nybble41
Parent article: LCA: Addressing the failure of open source

nybble41 said:

Actually, it's more of a user-interface patent than a software patent--as you say,

nybble41 a user interface patent is almost always a software patent. I am not aware of any patents being placed on user interfaces which are not software based. If there such a thing out there, then it's an abomination also.

Either you're patenting the software code itself, in which case it's called a "copyright" and not a patent, or you're patenting what the thing does and and a series of steps involving abstract things, irrespective of the code that underlies it. This later case is what a software patent is.

Therefore user interface patents ARE software patents- they patent an end result , from the POV of programmers, regardless of how it's realized in code.

They are defined on things which are purely abstract in nature and can be "realized" in many different ways. So not only is slide to unlock covering the Apple interface, it's covering ALL interfaces.


(Log in to post comments)

Software is The Glass Bead Game

Posted Feb 1, 2012 17:42 UTC (Wed) by nybble41 (subscriber, #55106) [Link]

> nybble41 a user interface patent is almost always a software patent. I am not aware of any patents being placed on user interfaces which are not software based.

I agree that user interface patents are almost always about interfaces implemented in software. However, when I use the term "software patent", I am specifically referring to a patent on an algorithm, not the vague and likely over-broad "patent on an invention that is typically implemented with software" which was expressed elsewhere.

Something like "slide-to-unlock" can be said to contain a very simple algorithm ("if a slide gesture is recognized, unlock the device"), but the patent is really about a distinctive style (sliding vs. entering a code, or tapping a button, or any of the other means of unlocking touch-screen devices used before), not the algorithm, which ought to be rather too trivial and obvious to qualify for a patent on its own merits, even assuming patents on algorithms were generally acceptable.

> If there such a thing out there, then it's an abomination also.

Agreed, 100%.

> Either you're patenting the software code itself, in which case it's called a "copyright" and not a patent, ...

You don't "patent the software code". As you said, code is covered by copyrights, not patents. A software patent--a patent on an abstract algorithm--would cover *all possible implementations* of that algorithm in software code, including independent clean-room code written without knowledge of any other implementation or the patent itself, which is much broader than the copyright on specific code implementing the algorithm.

> ... or you're patenting what the thing does and and a series of steps involving abstract things, irrespective of the code that underlies it. This later case is what a software patent is.

I agree that this is what a software patent is, but user interface patents are something else entirely. The slide-to-unlock patent, for example, doesn't seek to monopolize the "algorithm" of sliding your finger along the screen to unlock the device (which, as an action performed directly by a human, would be rather dubious as the subject of a patent), but rather the "look and feel" of the interface presented by the device, the graphics drawn on the screen combined with the device's response to a particular input gesture, without reference to the algorithms for producing those graphics, recognizing that gesture, or actually unlocking the device.

Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds