Tim Climis wrote:
> I have a related question, because when I first took up CSS in my
> designs in 2002 or so, I used to size my fonts in points. That was
> what word processing programs did it in, so that was how I did it.
>
> I gradually learned through online reading that that was not the
> right way to do it, and stopped, but I've never been able to figure
> out why it's wrong in the first place.
It isn't "wrong", but it cripples certain browsers and makes it harder
to apply reasonable end-user preferences in all browsers.
We refer to points for font sizing as part of the "print on screen"
philosophy, and "print" doesn't work all that well on all that many
screens - it belongs on paper and similar surfaces.
> Most graphic arts programs have the ability to guess the size of a
> pixel on your monitor, presumably from your drivers or some setting
> in your OS or something, so it seems that web browsers must be able
> to do that same thing. So it stands to reason that if you want your
> fonts to be 10pt (which is normal for print media) instead of 12 or
> 16pt (which is the common default size at the most common monitor
> resolutions) why not just set the font size to 10pt? and then if you
> have a 120dpi monitor, your browser knows that's 17px, and if you
> have an old 72dpi monitor, your browser knows that's 10px.
The problem is complex but also quite simple.
We work with what we have and what we gonna get - pushed by ourselves
and those who build browsers and write standards, and, most important:
what the end-users are likely to have, or get within a reasonable
time-frame.
With way more than a billion end-user installations on line, we better
put emphasis on "reasonable time-frame", as we will have to cater for
decade-old installations alongside the latest and greatest week-old
installations. Of course: we don't have to cater for old and new
installations in the same way, as long as we don't cripple any of them.
Practically:
1: screen-resolution is low compared to print-resolution - we need
300dpi on screens before we can call it acceptable.
Coarser steps makes it harder to make points - and any other unit - hit
screen-pixels exactly, and browsers round up/down at differing points.
This results in variations, that are much larger on screens than on print.
2: most graphic arts programs - or their users - solve this "low
screen-resolution" problem by resizing the entire project - zoom it, and
the artist then moves/scrolls parts into view while working.
Most browsers can do that too - now, but end-users are not _working_ on
the project - they _interact_ with it, and most end-users would prefer
to not have to scroll both ways just to be able to read and interact.
3: browsers have only recently started to take screen-resolution into
account and apply default-resizing. Not all browsers are there yet, and
those that are are not equally good at it.
4: end-users should be able to use their software - browser - to
re-format text to suit their needs/preferences. This is an advantage we
have in our digital age, and it would be sad if we limited and/or
crippled our software to "frozen print" now that we have finally gotten
out of it.
5: some browsers - IE - can't resize text where points or pixels are
used. The solution: "ignore font sizes on web pages", doesn't solve the
designer's problem. Blame whoever you like, but the problem persists.
> And then it's no more illegible than a novel or a newspaper.
6: if a printed work has too small text, the end-user can either use a
magnifying glass or throw the entire work into the fireplace. Neither
are very practical when the work is on screens, but the end-user can of
course look for better options elsewhere on the world wide web. I can't
prove it but I think many do.
I derive from the above that it is better to conform to reality than to
apply wishful thinking (and mouse-type in points) now.
At the same time I test how much of my wishful thinking that actually
works anywhere, hoping I may be able to apply some of it tomorrow as I'd
hate being stuck in the past at every crossroad and turn of evolution.
regards
Georg
--
http://www.gunlaug.no
______________________________________________________________________
css-discuss [cs...@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/
