- From: David Singer <singer@apple.com>
- Date: Thu, 7 Jan 2010 14:57:35 -0800
- To: Boris Zbarsky <bzbarsky@MIT.EDU>
- Cc: www-style@w3.org
it's a minor point, but... (and this is all personal opinion)... On Jan 7, 2010, at 10:55 , Boris Zbarsky wrote: > > What I wrote was simply a characterization of how Gecko's "page zoom" feature works. It does not claim to be a pixel-for-pixel zoom, nor is it. > > Then again, neither is Safari's (at least not desktop Safari). Nor Opera's. Nor Chrome's. I don't have IE on hand to test. > > Try it: open http://en.wikipedia.org/wiki/Main_Page in any of those browsers, then zoom in in any of the above browsers. Observe that the places where line-breaks happen change. So the engine is clearly not doing a pixel-for-pixel zoom but rather a relayout at the same device width and height using a different mapping from CSS lengths to device pixels. well, it shouldn't be a pixel-by-pixel zoom; fonts should be rendered at twice the fidelity (as if they were rendering the same 12pt text on a display that has a pixel density that is <zoom factor> greater. The wikipedia page layout changes because (at least in safari) the actual window doesn't change size; it is therefore (logically) 1/<zoom factor> the size it used to be, and so all the relatively-sized columns etc. are also 'smaller' when you zoom in. > > Maybe I misunderstood your objection, though? It's pretty minor. It's just that a designed 24pt font face is not the same as the 12pt face drawn twice as large. If the user asks for general zoom in, they should get the 12pt face drawn larger. (Smaller faces tend to have larger 'holes' in them, and be a little 'fatter', for readability). David Singer Multimedia and Software Standards, Apple Inc.
Received on Thursday, 7 January 2010 22:58:09 UTC