W3C home > Mailing lists > Public > www-svg@w3.org > August 2009

Radial gradients in SVG Full

From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
Date: Thu, 27 Aug 2009 17:22:40 +0200
To: www-svg@w3.org, klaus.foerster@uibk.ac.at
Message-Id: <200908271722.41566.Dr.O.Hoffmann@gmx.de>
Hello,

(just my personal observations)

SVG has already a radial gradient for a long time, 
a simplified form is available in SVG tiny 1.2 too.

I think, this canvas method creates more a cone than a
radial gradient. But I'm not sure, because as far as I have
seen, it does not define the range for 'w' (omega?) within
the formulas (looks like something to fix in the draft). 
The draft itself claims, that this creates a cone (why do
they call it createRadialGradient? This may create some
confusion with the SVG radialGradient)

To avoid conflicts with the existing radial gradient, one
could call it coneGradient ...
I think, there are other proposals to improve gradients
as well and to allow more sophisticated color distributions.
We will see, what will happen.

Do you have specific applications for such a cone gradient?
Is it of general importance to have such a specific type?
Could be interesting to gather a larger collection of different
types of specific gradients types to see how to organise this
in a useful way...
Is it useful for SVG and for authors to create for every specific 
type of gradient a new element or is there a more generic/elegant 
method to define gradient like entities with a set of  
well-considered element+attribute collection?

I have several applications for arbitrary color distributions.
Up to now it is not trivial to generate an arbitrary distribution
using the linear or the radial gradient. And I think, such a
cone gradient is just another sample for another specific
gradient type - and is typically less useful than the linear and
the radial gradient to approximate such an arbitrary distribution
with some number crunching.
Wouldn't it be more interesting to have a general and simple
mechanism to describe a color distribution without the requirement 
of heavy number crunching for authors?

Olaf
Received on Thursday, 27 August 2009 16:06:12 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:43 GMT