Subject: Re: [linux-audio-dev] [Fwd: Graphical dataflow programs violate patents]
From: Jens M Andreasen (jens.andreasen_AT_chello.se)
Date: Wed Dec 15 2004 - 12:27:33 EET
On ons, 2004-12-15 at 00:11 -0500, Dave Robillard wrote:
> Om, a modular synth I'm working on, absolutely "infringes" on that
> patent as well. The code that would be considered infringing is an
> elementary graph-traversal, very similar to a DFS (depth-first search)
> algorithm you'd learn in any second-year computer science program. No
> kidding you can't execute a node if it's dependencies aren't run!
> Geeze..
Actually, the patent is not on graph-traversal as such, but on graph-
traversal in context with virtual instrumentation having some kind of
adjustable front-panel on screen ... It is a "methods" patent that only
applies when all the ingredients are in place. The prior art shown in
court was overturned in part because it did not use windows and mouse.
An interesting (longish) thread about SoftWire versus LabView (coming up
in january) is here:
http://forums.lavausergroup.org/index.php?showtopic=140&st=0
The patent itself can be found here:
http://patft.uspto.gov/netacgi/nph-Parser?Sect2=PTO1&Sect2=HITOFF&p=1&u=
%2Fnetahtml%2Fsearch-
bool.html&r=1&f=G&l=50&d=PALL&S1=4901221.WKU.&OS=PN/4901221&RS=PN/4901221
(you will have to "repair" that link, or pick it up in the thread
mentioned above)
The (smallish) changes in Simulink 6.1 to work around the patent is
here:
http://www.mathworks.com/access/helpdesk/help/toolbox/simulink/rn/simulink14_6-1.html
> I'm Canadian, so I'm not sure if America's BS applies, but I'm hardly
> worried. What's next, getting sued for infringing on a patent for "a
> method of executing sequential lines of code"? Or perhaps evaluating
> (2+3)*4 correctly?
Simulink was in court shown to have been designed with the *intent* of
stepping on the patent. At least this jury thought so ...
Although it would be *possible* to use a virtual modular synthesizer in
place of LabView, I believe you would be in the clear. Your goals are
clearly different, no?
> There's absolutely no way that someone can't find prior art for this.
> It's completely frivolous.
I get the impression that SoftWires strategy (in january) will go along
the lines of showing that
1) They own relevant patents predating and invalidating the LabView
patent.
2) Furthermore, the method is obvious to anyone skilled in the trade ...
It will take some years before we know the outcome though.
>
>
> -DR-
>
--( ) c[] IANAL // Jens M Andreasen
This archive was generated by hypermail 2b28 : Wed Dec 15 2004 - 12:35:06 EET