W3C home > Mailing lists > Public > site-comments@w3.org > December 2009

Re: Redesign Styles Hypocritical

From: Peter Moulder <Peter.Moulder@infotech.monash.edu.au>
Date: Tue, 01 Dec 2009 16:48:31 +1100
To: "www-style@w3.org" <www-style@w3.org>
Cc: "site-comments@w3.org" <site-comments@w3.org>
Message-id: <20091201054831.GA21503@bowman.infotech.monash.edu.au>
On Mon, Nov 30, 2009 at 10:19:05PM -0600, Ian Jacobs wrote:

>>> I, too, had understood that the best practice was to leave the  
>>> default
>>> font size up to user settings. However, after discussion with a  
>>> number
>>> of designers,
>>
>> Can you identify those designers with whom you had a discussion?
>
> Sure. Some of them include:
>
>  * Airbag Industries
>  * Dan Cederholm
>    http://www.darowski.com/tracesofinspiration/2006/11/02/bulletproof-notes-a-day-with-dan-cederholm/
>  * Richard Rutter (clagnut)
>    http://www.alistapart.com/articles/howtosizetextincss/

Although at least the second of these web pages advocates changing
the default font size, it still advocates doing so as a proportion
of the default font size rather than using px.

I believe Gérard intended to ask for evidence that specifying a base
font size in px is a good idea, whereas the above two pages seem to
advocate not using px for font sizes:

Searching for the string `font' in the first one, the only relevant occurrence
is in a section entitled "Guideline #2: Let go of pixel precision",
which is followed by a guideline "Try using keywords or ems for sizing text.
Allow users to adjust as needed."  (I should note that I don't fully
understand the exact recommendations in guideline #2.)


In the second linked page, the penultimate iteration has all font
sizes be relative.  The final iteration targets an issue with a
browser that had been superceded even when the article was written
(2 years ago); and even then, the final iteration attempts to use
relative sizing for all but the affected browser (though in practice
using pixel sizes for all non-IE browsers).

Even two years ago, this workaround was prefaced with:

 RR> You could decide that undersized monospace text in Safari is something
 RR> you and your readers can live with, and as Safari 3 is included in OS
 RR> X Leopard and the latest update to Tiger, it will not be long until
 RR> the problem pretty much disappears. For the nervous control freak who
 RR> can’t wait, an alternative fix is to send text sized in pixels to
 RR> Safari.

The article's Conclusion talks only of using relative measures,
without even mentioning this pixel-based trick, so I'd conclude that
even back then, the author didn't that specifying a px-sized base font
size was the best approach for most web sites.


Only the final of these links gives any evidence that px might be better than
alternatives:

> I note that Eric Meyer does not choose sides in the author/user battle:
>  http://webstandardsgroup.org/features/eric-meyer.cfm

 "Eric: No [I don't think there there's a font-size solution that works
  for all users]. I wish it were otherwise, but I don't. Every font-sizing
  approach has benefits and drawbacks. The best you can do is find an
  approach that maximizes the former while minimizing the latter, and
  furthermore, that has to be done on a design-by-design basis. One
  design's goals might be best served by pixel-sized text, while
  another's will be better fulfilled by using ems or keywords
  ('x-small', etc.). Any time a person tells you that there is one and
  only one way to size fonts for all sites, they're trying to hand you
  a philosophy, not a solution."

Unfortunately, he doesn't go into what the drawbacks of non-px
font-sizes are, but at least it's evidence that an acknowledged expert
thought that there were some drawbacks back in 2004.

(His answer to the next question might indicate that that he sees px
font-sizes as a work-around for limitations in existing
implementations, though he's far from explicit.)


Incidentally, I'm not trying to take sides in whether or not to use
px in font-size, I'm just trying to draw attention to a relevant
distinction, and to clarify the opinions expressed on the referenced
pages.

pjrm.
Received on Tuesday, 1 December 2009 05:49:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 24 October 2012 16:21:32 GMT