Le 15 mars 2014 à 05:52, Tom Livingston <tom...@gmail.com> a écrit :

> You say "Font inflation exists to solve only this problem, which is a
> problem that fundamentally would make mobile Web browsers unusable.
> It doesn't happen if pages declare a <meta viewport>".
> 
> The OP's page here: http://81.4.104.136/fonts.html doesn't inflate the
> fonts for me on my iPhone. It miniaturizes them. Severely. This page
> doesn't have the viewport meta tag in the head. Without the meta tag,
> I will need to pinch and zoom, but the sizes are ridiculously small.
> I'm gonna guess that because there is no layout, the width of the page
> is very large so it is zoomed OUT a great deal. Right? Am I missing
> something?

Mobile Safari, in the absence of any viewport <meta> declaration, assumes a 
viewport 980px wide (that is equivalent to 980px wide window on a desktop 
browser). Of course the 'window' on iPad/iPhone is (much) narrower, thus the 
uniform downwards scaling of the page - this is/was done to insure that pages 
designed for the desktop would be viewable as a whole on those devices. Apple 
then invented two things to improve the usability: the double-tap on a column 
zooms/inflates that column and center it in the viewport, and then the viewport 
meta to set a specific width.
(OK, and a few more - font-inflation in some circumstances [*], and the ability 
to control it via the text-size-adjust property)

So no, you're not really missing anything.

And fwiw apple.com still has <meta name="viewport" content="width=1024" />

 [*] it always puzzles me how it exactly works, most of the time relying on it 
is like a game of Russian roulette… and gives bizarre results. On sites that 
can't really be made “responsive” for a variety of reasons I routinely set the 
text-size-adjust property to 100% (prefixed!), and use the viewport meta to set 
a width - similar to what apple.com does. Although Apple doesn't use the 
text-size-adjust property afaict; but then, they structured their content to 
avoid the font-inflation problem, I think.

Philippe
--
Philippe Wittenbergh
http://l-c-n.com




______________________________________________________________________
css-discuss [css-d@lists.css-discuss.org]
http://www.css-discuss.org/mailman/listinfo/css-d
List wiki/FAQ -- http://css-discuss.incutio.com/
List policies -- http://css-discuss.org/policies.html
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/

Reply via email to