Robin Berjon wrote:
On Feb 8, 2010, at 13:51 , Marcos Caceres wrote:
As much as I dislike it, I'm also not that inclined to define what to do for
non-"Window"-based implementations (beyond what we currently have in the spec;
which, I admit, is not much).
I don't think that we could even if we wanted to. Presumably, any technology
that embeds Javascript will have to have defined a global object of some form
or other (I don't see how it works otherwise) but the exact details will either
1) be copied from HTML or 2) be something different that we can't predict. So
for (1) it'll be obvious what to do, for (2) it's obvious that we can't do
anything to help.
Agreed.
It is, non-the-less, kinda bad that we don't have a spec that defines Window
independently of HTML but I understand why that is: because Window is a foul
intertwined incestuous mess...
Actually, the reason is because we don't have an editor for it. If we did,
Window could get split out of HTML (as it used to be).
No, I don't mean split in editorial terms. I mean in the way it
interweaves its slimy windowy tentacles into everything else and how
everything relies on it being there. Like Storage firing events on it,
etc. They should have called it the BlackHole interface, as there is no
escape:)
Also, the main use cases we are dealing with all rely on Window being there (HTML and, as Robin
points out, SVG). If UAs start to emerge that are windowless (e.g., "daemon" or
"headless" widgets), then we should formally define how this binds to some top-level
context. WDYT?
Yeah, but we can cross that bridge if and when we get there.
Agreed.
