W3C home > Mailing lists > Public > www-style@w3.org > January 2015

Re: [css-text-decor]: text-shadow should also apply to replaced content, like semi-transparent images

From: Axel Dahmen <brille1@hotmail.com>
Date: Thu, 29 Jan 2015 12:23:39 +0100
To: www-style@w3.org
Message-ID: <mad5ab$47k$1@ger.gmane.org>
> Kinda.  You brought up the cost of animating a shadow, for some
> reason.  That seems like a non-sequitur - you can animate any kind of
> shadow, and it's all about the same cost, so what's the point of
> bringing it up here?

The cost of daily work is this:

While creating and maintaining a website ...

... you have to implement shadows twice - always if inline images and inline 
text get involved.
... you have to test for shadows twice - always if inline images and inline 
text get involved. This becomes particular costly with dynamic content
... you have to test your web pages on each common browser - twice - as I 
wrote before.

So you expect to double the efforts of CSS programmers just for the slight 
difference in creating a rasterized version of fonts vectors and raster 

> Sorry, got distracted during that sentence and didn't finish it
> correctly.  It's a pure-geometry vector format.  There's no colors or
> transparency or whatever to it.  You just get some paths, and then
> tl;dr

I guess you must have have you heard about Bit blit operations yet? Then you 
must know that it is no difference between creating a b/w version from an 
image or text and that it's just a snap.
Vector fonts must get rasterized to be output. The rasterized output then 
gets "Bit blitted" and Gaussian blurred. So at the stage when shadows are 
getting added to an inline block, images are in no way different from fonts. 
I'm doing this kind of programming for decades now, so if IT didn't invent 
some pretty new stuff recently than that's what should happen within a 

Received on Thursday, 29 January 2015 11:30:29 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:50 UTC