On Fri, Sep 04, 2020 at 10:59:05AM +0200, Tomas Hlavaty wrote:
> thanks for the interesting info
> 
> On Fri 04 Sep 2020 at 10:43, <tomas@tuxteam.de> wrote:
> >> i can imagine a case where svg is inside html and maybe it could have
> >> arbitrary attributes, e.g. data-myattr1
> >
> > In theory, that's what namespaces are for. In practice, though...
> 
> html does not have namespaces

No, but SVG is XML, and thus it has. You can embed hyperlinks in SVG by
using the xlink namespace -- firefox even honours that. And you embed
SVG in html5, which isn't XML (we nearly had that, btw). That's the result
of that consortium thing: they prefer to continue their "fuzzy parser
wars". Gotta live with that.

> but you are right, if treated as xml, args could contain attributes from
> other namespaces.  i forgot about that

Yep. SVG /is/ XML, for better or worse :-)

> > Nevertheless, independently of how you embed the thing syntactically
> > (SVG even allows extending the DTD!) users will throw chairs at you
> > whenever it looks differently than in Chrome/Firefox/Internet Exploder
> > Version 0.95 or something. Be sure to duck quickly :-)
> 
> extending dtd?  interesting

Yep [1]. "Interesting" is a polite way to put it. My take would rather
be "@%#$&*!", but then, I was born in Spain, and some of us are said
to have bad language ;-)

My point is that those "norms" are built to leave bigcorps enough
wiggle room to try to dominate the market by bending them. Otherwise
the "norms" wouldn't be relevant in the first place.

OTOH, no norms would be even worse. But one shouldn't be too naive
to believe what's on the box. Just look inside :-)

Cheers

[1] https://www.w3.org/TR/SVG11/extend.html#PrivateElementsAndAttribute

 - t