On 12/13/05, bulia byak <buliabyak@...400...> wrote:

Sorry I didn't notice that RFE before. Now that I see it I think I
rather disagree with it. It postulates an equivalence of two colors -
the color under the SVG/PNG image and the background color of the SVG.
These two are NOT the same, however. For example I may want to have a
50% opaque black background in PNG over a yellow page.

I'm the one who submitted that RFE. Now that you mention it, I see my mistake. Yes, the drawing surface color is different than the image background color.

As for the example given by the RFE author, I usually do this by
adding a background rectangle of the color I need to a new layer. Then
I just hide that layer before export. This takes some time but at
least it's logical.

This is what I do too, and is the real reason I submitted the RFE. All I want to do is change the color of the Inkscape drawing surface while still having a _fully transparent_ image background for export, browser rendering, etc.. There is no way to do this currently, and adding a rectangle only to hide it later is an uncomfortable hack.

Now, the ideal way to handle this would be to store TWO changeable
colors: one is the SVG background with its transparency, and the other
is the color _under_ the background, to which the background would
blend (currently fixed white). This way we'll satisfy everyone. But
before that is done, the current scheme is at least more logical and
requires less awkward explanations than the scheme proposed in that
RFE.

This is certainly the way to go. Actually, I do not think the drawing surface color should have a transparency option at all. Since there is nothing under it, transparency makes no sense. In fact, we already _have_ a drawing surface color - it's just fixed white with no way to change it.

So, I propose that you please consider reverting that change.

Agreed. Should I submit a new RFE for "adjustable drawing surface color?" One of these days I'll learn GTK and start making these changes myself.

-s_tec