[thelist] dynamic font size

Felix Miata mrmazda at ij.net
Wed Jul 13 08:24:00 CDT 2005


=?ISO-8859-9?Q?VOLKAN_=D6Z=C7EL=DDK?= wrote:

> Felix Miata wrote:

> > Yes a pixel is a pixel regardless. The problem is that definition has no
> > meaning until you know the amount of space available for pixels to
> > occupy, and how many pixels are designated to occupy that available
> > space.
 
> But we should also note that generally users with high resolutions are
> users with large screen areas. For instance an 14inch monitor-user
> will possibly browse with
> 800*600, while an 17'' will possibly browser with 1024*768.

15" CRT display is 605 cm^2, and at 800x600 has a pixel density of 796px/cm^2.
19" CRT display is 1003 cm^2, and at 1600x1200 has a pixel density of 1914px/cm^2.
22" CRT display is 1366 cm^2, and at 2048x1536 has a pixel density of 2303px/cm^2.

So, 1600x1200 is 4 times as many pixels as 800x600, but for the
indicated monitor sizes 1600x1200 only has 2.4 times the area to pack
them into as 800x600. IOW, even though higher resolutions tend to be
used on larger monitors, the size difference generally isn't even close
to making pixels the same physical size. Note also that 1600x1200 is
just a common high resolution, not anything close to any limit. I can
run up to 2048x1536 on 19" here. Even on a 22" CRT, 2048x1536 is 6.6
times as many pixels as 800x600, but for 22" vs. 15" only has 2.3 times
the area to pack them into. http://members.ij.net/mrmazda/auth/dpi.html
 
> If I am a CAD designer and prefer to user 12048*11024 on 17'' monitor
> or higher, even 1 em will be small for me.

If you set your preference exactly according to your needs, then 1em
will ALWAYS meet your needs EXACTLY.
 
> Besides, there are problems wrt mac/win dpi differences, css-rendering
> differences (of 1em normal) between various versions of MSIE and many
> more that I cannot recall right now. I do not want to go in detail but
> the issue is a candidate to take a whole chapter in a book on web
> typography.

The DPI differences aren't Mac/Win, they're a complex combination of
various resolutions, display sizes, and system or software "DPI"
settings. If as recommended by everyone who actually understands the
issues involved you stay away from sizing using pt or cm or other
assumed physical sizes, DPI is a non-issue.
 
> If you use pixels, you will make some unhappy. If you use em's you
> make some others unhappy. There is no means of making everyone
> satisfied.

There is no reason to assume that using em will make anyone unhappy
except those who feel compelled to control the uncontrollable. By
definition, 1em is equal to the user preference. Nearly all users have
the power to change their preference to suit their needs, and for those
that don't generally someone with power has set something close enough
to be usable, e.g. public libraries.
 
> (However afaik, 1em corresponds to approximately 14px on msie-win
> 1024*768; which some designers (and some users) may think a
> larger-than necessary font size.)

What 1em corresponds to on Win depends on various system settings. On
Win IE, the default is always 12pt. At the Win nominal default of 96
DPI, that translates to 16px. But, Win provides users several larger
font options, which at up to 192 DPI can make 12pt up to 32px at
"medium", and 43px at "largest".
http://members.ij.net/mrmazda/auth/absolute-sizes-IE6.html
 
> > Preferring the wants and needs of
> > the clueless over the clueful is utterly backwards nonsense.
 
> If the clueless happens to be your client thing changes a lot :)

Ah, but your client is at your disposal to train. There is no good
reason for your client to remain clueless.

> If the client has a clue, but s/he thinks that the audience is clueless
> that's another story. You will have a hard time persuading your client.

It may not be easy, but trying your best is the right thing to do.
Pretty pages your client's visitors can't reasonably use do him little
or no good.

> To sum up, nothing is 100% right for everyone.
> The converse is also true: "nothing is 100% wrong for everyone".
 
Providing people who know what they want what they want is not a bad way
to succeed. Guessing what people who don't know what they want should
have is an efficient time waster, particularly when it impedes those who
know what they want from getting what they want.
-- 
"If you love your children, you will be prompt to discipline them."
                                                Proverbs 13:24

 Team OS/2 ** Reg. Linux User #211409

Felix Miata  ***  http://members.ij.net/mrmazda/auth/



More information about the thelist mailing list