W3C home > Mailing lists > Public > www-svg@w3.org > June 2010

Re: [SVG1.1F2 LC] 15.12 Filter primitive ˇfeComposite˘ formula

From: Alex Danilo <alex@abbra.com>
Date: Tue, 29 Jun 2010 22:05:42 +1000
Message-Id: <IXYR4L.B5MFDI26OLEN@abbra.com>
To: "Dr. Olaf Hoffmann" <Dr.O.Hoffmann@gmx.de>
Cc: www-svg@w3.org
Hello Olaf & Erik,

	We included all the Porter-Duff equations in the
compositing chapter back around 2002 or so, and the current
draft of that is at:


	I agree it would be good to include the actual equations
in the spec. especially since I don't believe the original
Porter-Duff paper took alpha into account IIRC.


--Original Message--:
>Erik Dahlstrom:
>>What other operations? Where in the spec?
>>Are you suggesting adding all the porter-duff equations for 'over', 'xor'  
>>etc to the feComposite section?
>Yes, for operator = "over | in | out | atop | xor" of feComposite.
>This should be not very difficult, because at least for them the
>usual implementations seem to agree on a result.
>Interestingly, for the arithmetic operator, where a formula is
>provided, I have seen different interpretations, for example
>Batik/Squiggle differs sometimes from Opera ;o)
>But I think, this is no indication, that providing formuals is
>a disadvantage.
>Informative additions would be enough, if it is not intended
>to risk some relevant change (what is not really likely).
>To find out, what the operators mean (including opacity), 
>I had to search, read an article at wikipedia, finding a reference 
>to the online version of the article from Porter and Duff, reading
>and understanding this, deviating the meaning for SVG filters.
>I think, this cascade of actions for the reader can be shortened
>dramatically by simply providing the formulas.
>If we assume, that the online version of the article is not
>accessible forever, the necessary action can be even more
>time consuming - traveling to a specialised library, booking the
>magazine, maybe including interlending, waiting, finally reading,
>captured by Star Wars reminiscences... ;o)
Received on Tuesday, 29 June 2010 12:06:31 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:54:27 UTC