This is a legacy document, and retained on the site in order to avoid link rot. The content is likely no longer (a) accurate, (b) representative of the views and philosophies of current site management, or (c) up to date.

Another way to think about font-size control

Matt and Todd posted to ciwas

Where it started

in article
mmcirvin-0910002045540001@ppp0b158.std.com, Matt McIrvin at
mmcirvin@world.std.com wrote on 10/9/00 5:45 PM:

Hmm… your users are upset that their own default text size is too large? I think you're sunk, since any attempt to change this is going to bother somebody else. Basically, the users are demanding that you read their minds -- they want the text to look a certain way, but aren't willing to change a setting in order to make that happen.

Unfortunately, I think that this is a very common occurrence, possibly because people have gotten used to the model of Web content as a TV-like broadcast that they passively consume.

If it looks bad, there must be something wrong with the broadcast.
» » »

in article
B60800C6.1A3B5%fahrner@pobox.com, Todd Fahrner at
fahrner@pobox.com wrote on 10/9/00 11:09 PM:

A descriptive analogy

This is KCSS Television, how may we help you?

Could you please turn the volume down? Your show is too loud.

[ pause ]

Er, have you tried the volume control on your television set?

[ sigh ] Next you'll start with PAL and SECAM and all that mumbo-jumbo. Just turn the volume down please. Not everybody has the time to fiddle with obscure dials on television sets. I don't work in the industry, you know.

Well, sir, if we were to cut our audio transmission levels, all those people who have set the volume to their liking would have to turn the volume up.

Serves those pedantic buffoons right for messing with the factory settings in the first place, right?

Sir, the point is that not everybody wants the same volume, and there isn't really any standard setting anyway, what with different set designs, seating distances, room sizes and acoustic properties. That's why it's your job, and not ours, to adjust the volume to your liking.

O really? Look, Mr. Smarty Pants, my uncle is a Senior Production Designer for your parent network -- he works very hard -- and if he hears that KCSS is letting ordinary untrained viewers set the volume of his shows, you'll be hearing from their lawyers.

Now for the last time…

We'll get right on it, sir!

You get the idea?

Now here's the rub:

This sort of insanity is already rampant.

The designer/builders behind the large majority of page views have long since kowtowed to such absurd demands from their employers, clients, and end users. Which means that it is now not quite so meaningful for users to set their preferred font size as the default: they'll see it so rarely on mainstream pages.

What they need to do is pick something a few steps larger: then they'll see their real favorite more often. This is of course a devolutionary cycle; the best we can hope for is a cataclysm -- something like an overwhelmingly popular browser with a powerful, prominent, fun font-size or page zooming UI, or a cheap wave of ultra-high res displays -- such that the designers can assert, confident that they will be believed.

Median font size is not my job -- all I handle are intervalic proportions.

Digression type="freeform rambling beery"

I think the primeval notion that a user can have one preferred font size for all Web pages is badly out of date, recalling a time before Web designers (called "information providers" in the early days) could even dream of doing layout, specifying different fonts, colors, leading, etc.

All of these things have a huge influence on what the most appropriate font size might be. A smallish, well-designed font set in narrow columns, with plenty of leading (line-height) and tonal contrast between fore- and back-ground, will be much more readable than a much larger, less well-designed font set in very long lines with little interlinear space, blue on deep red.

You will be able to read the former faster than the latter at a constant strain.

Faster (more readable) isn't necessarily better, though. I fantasize about rolling all these factors up into some sort of constant of readability algorithm, such that authors could simply specify a scalar value appropriate to the subject matter and typical sentence structure. The formatting would then adapt to the user's needs, the available resources, and the nature of the material.

Light material, in choppy journalistic style, is generally meant to be read very quickly. You're not going to miss any fine literary shadings if you skim. This accounts for the exceptionally short lines of newspaper columns.

More thoughtful, complex material, meant to be read more slowly, gets longer lines: the op-ed page essay lines are typically 2-3 times as wide as the disaster report columns. And really ponderous academic or legal material, which must be read slowly, typically gets the longest lines of all.

Imagine trying to read Immanuel Kant's 100-word sentences (or RFCs) at 3-4 words per well-leaded line, half of them hyphenated.

Matt McIrvin and Todd Fahrner , together contributed the text above as an original news post in ciwas.

The CSSPG wants to thank you both for your contribution to a better understanding of a very important aspect of web document publishing.

-

Jan Roland Eriksson