[ We didn't have a top-level thread for this, so for the sake of 
  completeness, I'm starting one. ]

At 09:07 PM 5/3/01 -0400, Dom Lachowicz wrote:
>the $64,000 question is "who is going to actually code a SVG 
>rasterizer?"
>
>Now, there is some support already done here by Raph, et. al. in librsvg 
>(see gnome's cvs), but we'll need to port this elsewhere and certainly will 
>need to improve it, should we choose to go this route. This code is 
>embryonic.
>
>Now, my advocacy for PNG is really simple: SVG simply doesn't work. You go 
>off all the time on how features should just work, and work right, etc... 
>But having SVG in our Import->Graphic dialog is *really really really* bad 
>for a few reasons:
>
>1) We can't display SVGs
>2) We have trouble reloading SVGs (or docs with embedded SVG data)
>3) No one has touched this code in ages, and it's not to that happy fuzzy 
>mature state where you can just leave it alone to gracefully acquire 
>bit-rot.
>
>So I'm taking out SVG from the NON-DEBUG version of Abi. No one gets to 
>argue here without providing code to correct *all 3* of my above problems. 
>There will be no discussion here.

Absolutely.  I'm surprised that nobody's turned that off already.  

The real question is ... does anyone have a strategy for adding *real* SVG 
support for 1.0?  If so, this would be an excellent place to start filling 
us in on the details.  

We'd all love to have a cool XP implementation of the subset of SVG which 
corresponds to WMF, but our efforts to date to design and/or code one have 
been, shall we say, anemic.  ;-)

Paul


Reply via email to