Posted by Neredbojias on 03/25/06 21:00

With neither quill nor qualm, Alan J. Flavell quothed:

> On Sat, 25 Mar 2006, kchayka wrote:
>
> > Jukka K. Korpela wrote:
> > >
> > > The name orange is an unfortunate choice, since it is not a
> > > defined name by CSS 1.0 and CSS 2.0 specifications,
> >
> > It is, however, one of the well-established X11 color names, which
> > are supported by many browsers, even Netscape 4.x.
>
> That's as may be, but Jukka's advice is good: a browser has every
> right to ignore it, supported by the CSS recommendations on error
> handling.
>
> > IIRC, X11 color names are included in the CSS 3 spec.
>
> It's about compatibility. Be conservative in what you offer, and
> liberal in what you accept. In CSS terms, that means CSS authors
> should not offer newly-defined stuff unless willing to cope with the
> consequences of it being ignored, whereas developers of software
> should implement newly-defined stuff as soon as possible, while still
> following the CSS mandates and guidelines about ignoring stuff that
> they don't understand (i.e DON'T try to guess what the author might
> have meant, as the consequences can be disastrous).

I was playing around with something like this the other day:

<a href="hula.mp3" type="audio/x-mpegurl">A Hip Song</a>

....in an attempt to avoid using .m3u redirection files. IE6 seemed to
try to open it as an image file in a new window! (FWIW, it worked in
Gecko and Opera offered a disappointing Open/Save dialog.)

--
Neredbojias
Contrary to popular belief, it is believable.

[Back to original message]