W3C home > Mailing lists > Public > www-patentpolicy-comment@w3.org > October 2001

Re: Apple's SVG patent

From: Glenn Randers-Pehrson <glennrp@home.com>
Date: Fri, 05 Oct 2001 09:10:21 -0400
Message-Id: <>
To: www-patentpolicy-comment@w3.org
>> The Apple patent, on the other hand describes taking a pair of color
>> images, call them A and B, and then using the *color* in B as the alpha
>> for A when compositing over some destination.

Claim 1 of the patent doesn't say anything about "color" images; it merely
describes alpha composition as we already know it.  I doubt that the
claim is valid and would like to see it overturned (or better, disclaimed
by Apple).  It's only Claim 2 that limits itself to "color" masks.

>> So to expand the equation
>> Redresult=((1-RedB)*RedA)+(RedB*Reddestination)
>> Greenresult=((1-GreenB)*GreenA)+(GreenB*Greendestination)
>> Blueresult=((1-BlueB)*BlueA)+(BlueB*Bluedestination)
>> Notice that 'alpha' does not occur anywhere in this system of equations.

Does this imply that all we need to do to avoid patent infringment
is to change "RedB, GreenB, and BlueB" to "RedMask, GreenMask, and BlueMask"
in our implementation?

>> You will not find that this system of equations is used in the SVG
>> specification and you are not required to use this system of equations
>> to implement SVG. Thus, the patent does not apply to SVG - it is not
>> "essential technology".

Claims 6 and 7 of the patent claim the idea of using text as a mask (and
say nothing about the text mask having to be 3-color).  Example mask01 in
the SVG spec seems to infringe on claims 7 and 8.  I find it hard to
imagine that prior art for this claim doesn't exist.

I am not a patent lawyer, just a very interested bystander.

Received on Friday, 5 October 2001 09:13:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:06:44 UTC