‘FontBitch’ Defined

so the debate continues!!!

I wouldn't consider myself a font bitch...

but I am definitely not a Neilsonist either.

Too many people get the idea of "font control" mixed up with "micro fonts". I do agree that there is a fine line to this argument... I much prefer to have a verbal discussion over a topic like this.

I think "Font Fuhrer" or "Font Despot" or "Fontadonna" would be better.

BTW, I also see something wrong with the other extreme: the laissez-faire hippies who think they can just leave the size entirely up to the user.

hhp

At the risk of sounding like a mysogynist, many girls I know don't mind the word "bitch" as it seems to bestow some kind of power on them. And applied to a guy it's a feminizing joke, so I don't have any problem with the label. In fact, it's pretty funny.

Is "butch" the masculine iteration of "bitch"? Or is it "batch"? I can never remember...

There is no masculine iteration of "bitch"...

hhp

hmmm...maybe `FontPimp¥ ?

I don¥t know. Small Fonts are nice for designstuff. But for `real¥ readable content I prefer a size beetween 10 and 12 points.

There is no masculine iteration of "bitch"...

this from one of the guys who invoked cher a couple threads ago. i'm appalled.

(kidding. kind of.)

Surely the masculine of bitch is dog, dawg?

Those who proclaim absolutist laws like "thou shalt never specify in pixels" or "thou shalt always design fluid pages" don't understand all the issues.

Try designing a dense, feature-packed web application aimed at browsers from NN4 up to IE6, and don't specify font sizes at all. Then hear many more users complain about how f'ed up it looks in their browser.

MH, right on.

> I prefer a size beetween 10 and 12 points.

Exactly how are you using "point" here?

The only really useful measurement on-screen is the pixel, and I have concluded that at this moment in time (and for the near future) 16 pixels is the optimal height for readability, in the overall. This translates to "12 point" in the [default] Windows setting of 96 dpi.

hhp

The masculine of bitch is "dirty dawg" or "A-Hole" if your nasty.

the masculine colloquial iteration of bitch is surely bastard. obviously the literal definitions are not parallel, but the gendering and general meanings are.

Yes, Hrent.

You¥re right.

But is the "[default] Windows setting" at 72dpi, isn¥t it?

On Mac¥s it¥s 96DPI, right

No, the other way around.

hhp

>the masculine colloquial iteration of

>bitch is surely bastard. obviously the

>literal definitions are not parallel, but the

>gendering and general meanings are.

No wait, I've got it! The masculine colloquial iteration of bitch is "mitch".

Sorry.

g.

Tanner:

>But is the "[default] Windows setting"

>at 72dpi, isn¥t it? On Mac¥s it¥s 96DPI

Hrant:

>No, the other way around.

The fact that Mac IE lets you switch between these virtual resolutions within the browser preferences confusificates the issue even further.

I think the issue here is not flexibility of choosing a personally appropriate type size, but designing a site to be legible by everyone in the first place.

g.

well...okay

Those who proclaim absolutist laws like "thou shalt never specify in pixels" or "thou shalt always design fluid pages" don't understand all the issues.

Amen to that...

You tell 'em MH

Give the power to the people. Instead of enforcing font choices on users, allow them choice. Sure, you know what's better for a user and there are some incredible difficulties in programming web pages to allow customization, but come on ñ not all of us are stupid, a lot of users can make informed decisions.

Apple values their customers, enough to allow them to change the default system font. I imagine only a small percentage of users change their system font, but allowing a user to change their system font is, in effect, Apple telling users, "I value you enough to allow you to make changes to what we already believe is the best possible users experience we can muster."

Of course 25pt Comic Sans is an inappropriate system font, but at least Apple is not a font dictator (bitch doesn't sit as well with me) although this could be argued. Please allow more control over anti-aliased text Apple!

In any case, web sites that choose to ignore the user are in effect saying, "My design is important, too important to allow the user any control."

Are we going to go through this again?

> designing a site to be legible by everyone in the first place.

Legible is trivial. *Readable* is the trick.

> a lot of users can make informed decisions.

Not concerning something they have insufficient understanding of: how humans really read. Mind you, neither do most designers, but that's beside the point.

> Apple values their customers, enough to allow them to change the default system font.

Apple values the cash it extract from its customers, and one way it does this is by giving them the illusion of power, even if it means taking away true functionality.

Give your kid chocolate cake for breakfast, lunch and dinner, and he'll love you, and even if he makes it to adulthood he'll be so sickly that he won't know to blame you.

A good user interface (OS) takes into account how the user uses things, such as text. The OS should calibrate itself to its user and usage environment, and pass this info to the apps running under it. This is not nearly as tricky as it sounds.

> "My design is important, too important to allow the user any control."

A lot of designers do say that, and they suck. I'm saying "Readability is too important to let the user mess it up by choosing a size that's too large." There is emprical proof that users would do that.

hhp

I am interested in your comments, "The OS should calibrate itself to its user and usage environment, and pass this info to the apps running under it." Please expand on this topic if you have the time or inclination.

Something like this:

First, the OS should collect relevant information about the hardware, like monitor resolution/size. Second, it needs to ask the user (just once) if he has special requirements, like poor vision. Third, it needs to have a "default text size" which is equivalent to 10-11 point in a normal book (which is typically read at about 14 inches): this is the optimally readable size (even though readers think it's too small). Then an app can ask the OS what this size is, and decide what to do with it - ideally allow its change but discourage it. The user should not have to learn what the hell a dpi is and how/why to change it - he just needs to know if he has poor vision or not... Also, the OS needs to have different profiles for different users of the same system.

hhp

Please forward your comment onto Apple. I would appreciate your suggestions applied in an update of Apple's system software.

The thing is, I'm sure they've thought of stuff like this. So why not do it? Because it's not cost-effective: you can't sell this idea neither to the users ("Readawhut?") nor the top brass ("How much money is this gonna make?") so...

Strangely enough, the one company that has invested time in promoting readability is MS, via their Typography Group in general, and Bill Hill in particular.

hhp

Please allow more control over anti-aliased text Apple!

Easily done with Tinkertool, available here.

Much info in their faq (scroll to ‘How can I control smoothing of fonts?’).

Glad you mentioned Tinkertool, Dean. But, as far as I can tell, it still doesn't fix the horrific kerning of un-anti-aliased type in apps like IE.

From the FAQ: Are there unwanted side effects when changing the font smoothing settings? This depends. Some versions of Mac OS X show kerning problems (the spacing between adjacent glyphs is incorrect) when you use particular fonts in certain applications and anti-aliasing is disabled.

...kerning problems...

Ja, thoughI haven’t been able to reproduce that: must be a very specific combination of screenfont, OSX version and IE version – the only place I get arbitrary kerning and widths is, well, in everything else that Microsoft makes.

MS Office: the I Ching of letterfit.

> Tinkertool

Tinkering don't gonna cut it.

We need fonts with hand-made anti-aliasing, and we need it on both platforms.

hhp

Hrant:

Street luge, Rotarians, blood and pancakes.

Also, underpants.

Some rough facts and figures about how I, personally, am looking at the screen and experiencing these fonts that are subjectively too small for me: I just measured the distance between my eyes and the monitor and it is 3 feet. Seeing as my 21 inch is still awaiting repair, my main screen now is a 17 inch - 16 inch viewable - high rez ViewSonic GT775 which although not new is still rock solid. It is running at 1280x1024 pixels, 75 Hz. I cannot give figures on how good my eyesight is, but I've never met anyone in my life who can see better than myself, so I'd estimate my Visual Acuity at 20/30 at least. I'm 42, by the way. At 3 feet away from the monitor - which is supposedly a healthy distance according to the ergonomicists - perhaps too far - I start to have difficulty with black text on a white background at 9 point - using IE 5.something on a Mac. My default font is the superb InterfaceDaMa by Dalton Maag (a free download via http://www.daltonmaag.com ).

This means that, subjectively, for me, almost any other font choice is worse at this size, and that web sites using pixelfonts that are smaller and not anti-aliased fall into the category of font bitch sites.

I guess a lot of the problem is that these font bitches use something other than black or white on a background that is not a hard enough contrast. (And sites with a light text on a dark background have to be *very* special if they expect me, personally, to keep reading and keep returning. One exception in my list of Bookmarks is good old Andy Ihnatko's yellowtext blog: http://www.cwob.com/yellowtext/)

Also, the subjective font choice muddies the waters here - but maybe not. Perhaps once a subject's font preference is taken into account then there's your reference point. After all - we're dealing with subjective experience here. Perhaps this long-winded blather is worthless, but apart from the font choices, maybe some hard numbers can be applied to optimum screen reading distance plus screen resolution/size plus text font size. In a few years - perhaps less than ten - we'll have wall monitors for our computers and the real human question regarding type size will be rendered irrelevant because the monitor's resolution will be off the scale.

Anyhow - there's no getting away from the simple fact that 12 point type approximates human handwriting - at arm's length, whether written or read. And straying from that human scale doesn't help you if you want your text to be read by anyone other than yourself.

Martin, you don't count. The specs you gave in your first paragraph make it clear that you're not human. Seriously: you're *way* off the mainstream target (and I think you know it).

Typical averages:

Distance: a little less than two feet.

Eyesight: a little worse than 20/20*.

Resolution: a little less than 1024x768.

Display Size: about 16" viewable.

I've done the calculations, and if we assume 10.5 type in print to be the most readable, that translates to 16-pixel bitmaps.

* BTW, in better vision the second number is smaller, like I'm 20/15 (although my range is very close - I have slight trouble with freeway signs).

> 12 point type approximates human handwriting

This is moot, especially when you factor in reading distance, which is much greater for reading a screen than it is for writing. Unless you're an albatross.

hhp

Hrant: My rambling post was not exactly the most succinct piece of writing I've done, to be sure. What I think I meant about the handwriting is that perhaps if you scaled up the "human scale" 12 point font size by the same degree as the difference between "arm's length" and screen distance, then perhaps you get that 16 pixel size.

Is this the way you made those calculations?

> Is this the way you made those calculations?

Basically, yeah. But I used 10.5 point, and to me "arms-length" refers to holding a book, not writing.

hhp

As to the subject of type/background contrast: I read a something a while back that noted that less contrast on screen increased readability. The same article also noted that a darker background with lighter type was easier to read. The author explans that the difference between this and the general rules for printed work is that a monitor is made of light not reflecting, but coming right at you. In this way, the darker background with lighter type and less contrast is less tiring on the eyes.

I read this, I think, on Zeldman's A List Apart, I believe, but I couldn't find it just now as I was looking for it.

True/Untrue? Anybody know?

Although I myself haven't seen rock-solid proof yet, since the screen is emissive (unlike print which is reflective) it makes sense that too much white can make things difficult, and that reverse contrast could help too. Reverse contrast also preserves the display hardware better.

In addition, the norm for on-screen fonts is 1 pixel stems, and I feel they "fade out" too easily in the strong white, which is why I use Georgia Bold (2 pixel stems) on my own site. Or you could use an off-white background.

And there are other advantages to 2-pixel stems:

1. It gives more "room" for good anti-aliasing.

2. With 3-pixel bolds (bold being especially uesful on-screen since italics suck) you get a gentler emphasis.

3. It leaves room for: serifs that are not overpowering; stroke contrast.

On the other hand, it's true that user focus groups (like the ones MS Typography has conducted) indicate that 2-pixel stems make the text too dark (at normal reading sizes), but I tend to think that this is a preference that only seems relevant when you make people think it's important for them to choose.

hhp

Its my web. I use a 1800x1440 resolution. I have a 21 inch monitor. I use the right tool for the right job. I use opera most of the time. when Im using IE and cant read something I zoom in with opera. easy.

when I need to drive a nail through a plank of wood I use a hammer. I could also use the butt-end of a screwdriver, a shoe, a brick, a tire iron, a shovel ...

Its my web.

Ryan, are you a TV commercial?

hhp

No. Just pointing out facts.

fact: I use browser X

fact: I use configurations A,B and C

fact: Sometimes I have to use browser Y

fact: Sometimes configuration A,B and C isnt the best setup for the current task

fact: Its my web

And I drink Lapsang Souchong. But the Wendell tea shop is not mine.

hhp

what about people who use the internet like this?