Re: Scheper's Catmull-Rom curves, and Spiro curves

Hi All,

	For this sort of curve to be added to SVG, Mr Levien would have to
assign royalty-free rights to any patent that may be granted in relation
to his thesis work for all implementors of SVG or related W3C specs using
the invention as claimed.

<snip/>
>Raph has a USA software idea patent on Spiro, with a GPL grant,
>explained in http://levien.com/garden/ppedit/README :

Well that grant is effectively useless right now. Since there is no
patent granted yet, there is nothing to grant. What he should be
doing is pledging to grant rights should the patent be issued
in future.

Here is the application:
http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=1&f=G&l=50&co1=AND&d=PG01&s1=%22Levien+Raphael%22.IN.&OS=IN/"Levien+Raphael"&RS=IN/"Levien+Raphael"

	So the GPL "grant" is well and good, but to form part of a W3C spec it
has to be granted to all implementors of the spec. both open and closed source.
I wonder why would anyone continue to pay the considerable expenses required
to obtain a patent if it was just to be given away for free?

	IMO, if authoring tools can use Spiro curves and generate Bezier
content for rendering then the logical place for the Spiro curves themselves
is in the authoring tool. That provides the author all the creative freedom
they need, preserves backward compatibility (by not changing anything)
and avoids the dreaded 'p' word;-)

	Interestingly, ASV used Beziers to approximate elliptic arcs,
so one could argue those are superfluous too...

	It would be nice to be able to specify a curve that smoothly
interpolates between a bunch of points as originally discussed, but
can we please consider only technologies where the IP situation
is clear. SVG.264 anyone...

Alex

Received on Wednesday, 14 July 2010 22:11:42 UTC