W3C home > Mailing lists > Public > www-style@w3.org > December 1997

RE: HSL (fwd)

From: Chris Wilson <cwilso@MICROSOFT.com>
Date: Wed, 10 Dec 1997 16:33:40 -0800
Message-ID: <C35556591D34D111BB5600805F1961B90229E723@red-msg-47.dns.microsoft.com>
To: "'Douglas Rand'" <drand@sgi.com>, MegaZone <megazone@livingston.com>
Cc: www-style@w3.org
I'm not sure what is meant by "RGB percentile values" - percentage values in
rgb functional notation ("color: rgb( 100%, 0%, 100%)") are supported by
both Internet Explorer 4.0 and Navigator 4.0.  Rgb functional notation was
not in IE 3.x.

At any rate, my objection is that a new color specification type is a
non-optional technology.  Envision loading a page that sets its text color
using HSL, and has the background set to "black" (the named color).  It
makes sense to do this - you know you want a solid, normal black background,
so you use the named color 'cause it's easy, and you want to tweak the
foreground color, so you use HSL.

Now look at the page in a browser that may support all of CSS1, but none of
CSS2.  Look at it in IE4 and Navigator 4.  It's black-on-black.  This is
much worse than not supporting the right semantics for margin values, or not
supporting font variants, or even not supporting positioning - even the
smallest embedding system UA could support enough CSS to understand "black",
and show you black-on-black because it doesn't support HSL.

HSL may be a wonderful thing - but putting it into CSS >1.0 may not be.  Put
it in the tools.  Heck, if someone sends me a Javascript/C/C++/Java
algorithm, I'll put support for it in my DHTML CSS editor for IE4.  (Public
URL pending.)

Chris Wilson

> -----Original Message-----
> From:	Douglas Rand [SMTP:drand@sgi.com]
> Sent:	Tuesday, December 09, 1997 1:04 PM
> To:	MegaZone
> Cc:	www-style@w3.org
> Subject:	Re: HSL (fwd)
> MegaZone wrote: 
> > Would HSL be something that doesn't take practice?  It seems like any
> > format would require practice to use effectively.
> More to the point,  there are a class of people who would be able
> to deal with this more effectively than RGB.
> >...
> > color so the system used really doesn't matter.  And to mean that means
> > there is no point adding yet other options.
> No,  I don't think that follows.
> > In the real world that that means is something else that will NOT be
> > supported.  Just as the RGB percentile values aren't now by the major
> > browsers.  The more possible systems, the greater the likelyhood of
> > deployment issues, which is a major pain for developers already.  The
> fewer
> > the possible systems, the more reliable support will be, and the easier
> > it is for authors to develop pages.
> While this is probably true (although I *do* support percentiles).  I
> don't particularly understand it.  Percentiles take even less effort.
> But I'm not in favor of getting into predicting the behavior of browser
> vendors.  If I were to do that I'd advise dropping alot of things.
> Doug
> -- 
> Doug Rand				drand@sgi.com
> Silicon Graphics/SSO			http://reality.sgi.com/drand
> Disclaimer: These are my views,  SGI's views are in 3D
Received on Wednesday, 10 December 1997 19:37:12 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:26:46 UTC