W3C home > Mailing lists > Public > www-style@w3.org > January 2000

RE: Best practice for font control

From: Karlsson Kent - keka <keka@im.se>
Date: Tue, 25 Jan 2000 01:23:32 +0100
Message-ID: <C110A2268F8DD111AA1A00805F85E58DA6854D@ntgbg1>
To: www-style@w3.org

> (*) Windows/other OS'es 12pt onscreen for a Windows PC is 
> 16pt for other
> OS'es, 12px is 12px everywhere. This goes for printouts too.

I haven't got the foggiest idea what "px" means in a printout.

For pica points, Did˘t points, mm, cm, even inch, one can pick up
a suitable ruler and measure the size.  Assuming one knows
the actual (rather than just nominal) zoom.  And for printouts
as well as desktop screens, the actual and nominal zoom
should be as close as practically possible.

I know that Windows has one builtin (but not properly declared
to the user) zoom, and MacOS has another builtin zoom for
desktop screens.  Now it appears that at least future browsers
will use the same buitin zoom at least for those two platforms.
What I suggested (among other things) was that the default
desktop screen zoom would be 1) (about) 130%, specified by
CSS, 2) the zoom be properly 'declared' as such to the person
in front of the (desktop) screen, and 3) that the person in front
of the screen should (if practically possible, i.e. when in 
control of other aspects of the "surfing") be able to change
this zoom for his/her reading convenience.

So, 12pt at a (total) zoom of 250% should be (pick out your
Pica ruler, or convert to mm) 30pt on the (desktop) screen
surface.  (If not, complain to your favourite browser provider.)

Now, what is 12px at a zoom of 250%?  I don't dare to
guess.  (Remember that there is also a builtin zoom 
of maybe 110%, or 130%, or 160%, where 1px should be
1 pixel high.) This is not well specified, and almost anything
might go.

And someone with bad eye-sight, will probably want to *reliably*
and gracefully enlarge *everything*: text, images, vector graphics...
Not just perhaps enlarge some of the text (if em, %, or !important
is used).

	Kind regards
	/kent k
Received on Monday, 24 January 2000 19:23:35 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:26:52 UTC