- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Fri, 13 Mar 2015 13:22:46 -0700
- To: ddailey <ddailey@zoominternet.net>
- Cc: Philip Rogers <pdr@google.com>, Dirk Schulze <dschulze@adobe.com>, Thomas O Smailus <Thomas.O.Smailus@boeing.com>, Boris Zbarsky <bzbarsky@mit.edu>, www-svg <www-svg@w3.org>
On Fri, Mar 13, 2015 at 12:52 PM, <ddailey@zoominternet.net> wrote: > "Tab Atkins Jr." replied > On Fri, Mar 13, 2015 at 10:41 AM, Philip Rogers <pdr@google.com> wrote: >> Has the SVGWG considered specing the differences in <img> vs <object>? It is >> not obvious to users that there are large differences between the two. > > Yes, this is the Integration spec: <https://svgwg.org/specs/integration/>. > Several things aren't clear to me about this discussion: > > a) I was thinking that the reason Social Media and Wikipedia might not want to allow user uploads of SVG into <object>s is because they don't want to trust 3rd party script > b) it seems like the only danger associated with SVG SMIL/SVG interactive SMIL is when one listens to keystrokes. Suppose <img src="file.svg"> allowed mousedown mouseover mouseout onclick, mouseup etc. but no keypress events. Is there any danger then? The pedagogical objectives that make SVG SMIL cool are then not harmful. > > that was the reason for the request but maybe I am missing something. I get the feeling though that the people saying just use <object> if you want interactivity are missing the basic point here. Running script in <img> is out of the question, so we won't get a full document context regardless; adding in enough plumbing to handle interactive SMIL (when we're rapidly dropping it in the first place) is almost certainly not worth the engineering effort. ~TJ
Received on Friday, 13 March 2015 20:23:33 UTC