W3C home > Mailing lists > Public > www-style@w3.org > October 2011

RE: [css3-images] simplifying radial gradients - Spotlight example

From: Brian Manthos <brianman@microsoft.com>
Date: Sun, 9 Oct 2011 20:42:47 +0000
To: Brad Kemper <brad.kemper@gmail.com>, Chris Lilley <chris@w3.org>, "Tab Atkins Jr." <jackalmage@gmail.com>, Sylvain Galineau <sylvaing@microsoft.com>
CC: Alan Gresley <alan@css-class.com>, "www-style@w3.org" <www-style@w3.org>
Message-ID: <9710FCC2E88860489239BE0308AC5D17F01047@TK5EX14MBXC266.redmond.corp.microsoft.com>
I wasn't being sarcastic.   You changed the rules intended in the example and I called you on it.

Clearly, you're missing the point and it seems like you've put up a mental block on it so basically it's becoming worthless to spend time attempting to get you to grasp the concept.

I'll try again though:  We're defining a new <image> type, not a <background-image> type.  As such, we should make it define it fully so that in all places where <image> is (and will be in the future) used the intended use cases are supported.  Tying clearly desirable renderings to the crutch of background properties compromises the desire for a full <image> type.


I think the WG has to decide whether they want incomplete -- and incompatible with existing use cases -- radial gradient support via <image> or not.  IMO, it does a disservice to the community to take away functionality between WD and LC that has been there since before it was even drafted in a CSS specification.

Perhaps it would be useful for Dave Hyatt to speak to why the center point was included back in 2008, and why he thought it was useful to support them in border-image, list-style-image, and the content property.
http://www.webkit.org/blog/175/introducing-css-gradients/

Further, it might be productive for him and others to discuss whether they think it represents a positive or negative move for them to have to remove that functionality when making the transition from prefixed to unprefixed in Gecko support for gradients.

-Brian

> -----Original Message-----
> From: Brad Kemper [mailto:brad.kemper@gmail.com]
> Sent: Sunday, October 09, 2011 12:36 PM
> To: Brian Manthos
> Cc: Sylvain Galineau; Alan Gresley; Chris Lilley; www-style@w3.org
> Subject: Re: [css3-images] simplifying radial gradients - Spotlight
> example
> 
> My desired approach in the modified version I created was to show that
> it could be done with a simpler syntax, allowing authors to leverage
> their existing knowledge of authoring backgrounds to get that effect,
> without having to learn a new, mostly redundant syntax for getting the
> same effect inside the image itself.
> 
> You didn't say "my desired approach" before, you said "the desired
> approach". The desired approach I took is as described above. I think
> we should both try to refrain from sarcasm. It's not helpful.
> 
> On Oct 8, 2011, at 11:09 PM, Brian Manthos <brianman@microsoft.com>
> wrote:
> 
> > I made the page.  My desired approach was to change the gradient with
> background-image alone, not by fiddling with other properties.  So, no,
> it wasn't in my desired approach when writing the page.  Sorry, you
> can't read my mind as well as you might think you can.
> >
> > -Brian
> >
> >> -----Original Message-----
> >> From: Brad Kemper [mailto:brad.kemper@gmail.com]
> >> Sent: Saturday, October 08, 2011 10:53 PM
> >> To: Brian Manthos
> >> Cc: Sylvain Galineau; Alan Gresley; Chris Lilley; www-style@w3.org
> >> Subject: Re: [css3-images] simplifying radial gradients - Spotlight
> >> example
> >>
> >> Yes it is.
> >>
> >>
> >>
> >> On Oct 8, 2011, at 10:22 PM, Brian Manthos <brianman@microsoft.com>
> >> wrote:
> >>
> >>> Again, this leverages background-position which is again not part
> of
> >> the desired approach.
> >>>
> >>> -Brian
> >>>
> >>>> -----Original Message-----
> >>>> From: Brad Kemper [mailto:brad.kemper@gmail.com]
> >>>> Sent: Saturday, October 08, 2011 3:37 PM
> >>>> To: Brian Manthos
> >>>> Cc: Sylvain Galineau; Alan Gresley; Chris Lilley; www-style@w3.org
> >>>> Subject: Re: [css3-images] simplifying radial gradients -
> Spotlight
> >>>> example
> >>>>
> >>>>
> >>>> On Oct 7, 2011, at 12:19 PM, Brian Manthos wrote:
> >>>>
> >>>>> Here's a relatively simple example of the kinds of things I
> expect
> >>>> people to want to use gradients for in the next year or two.  Most
> >> of
> >>>> the complexity is due to browser differences such as rules vs.
> >> cssRules
> >>>> and the prefixing of gradients.
> >>>>>
> >>>>> Ignore the logic related to scenery.  That's just a random
> backdrop
> >> I
> >>>> cobbled together to have a visual behind the effect.
> >>>>>
> >>>>> The key function to look at it BuildSpotlight.  Notice that the
> >>>> responsiveness to mouse location only requires updating background
> >>>> image, and only to specify the center of the circle.  No color
> stops
> >>>> need to be adjusted.  No other CSS properties need to be nudged.
> >>>>>
> >>>>> Notes:
> >>>>> (1) I've only (briefly) played with this in FF7 and IE10.  Based
> on
> >>>> my other sample pages, I expect it works in Chrome and Safari (but
> I
> >>>> suspect not in Opera).
> >>>>> (2) On my machine it's a bit CPU-intensive in FF7 (even with "Use
> >>>> hardware acceleration when available").  I don't know why and it
> >> wasn't
> >>>> intentional.  If there's a simple workaround, private e-mail would
> >> be
> >>>> appreciated.
> >>>>
> >>>> Here is a modified version that is compatible with my simplified
> >>>> syntax.
> >>>
> >>>
> >
> >
Received on Sunday, 9 October 2011 20:43:17 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:45 GMT