Me too.

hhp

Had to quickly try it out. If I did dream this it was a lucid dream of sorts, for it does seem to hold true. Well, at least in these huge sizes.

I did the same thing once to the inner, once the outer contour (FL: Action > Add Nodes and then Action > Random), and left the other contour clean.
The ones on the left do look noticeably more «broken» to me.
Any thoughts / opinions welcome.

Here's a theory: when you look at a shape, you tend to look at its center point (or maybe its "center of gravity"), and the closer something is to the middle of the fovea the more detail you see in it. Since counters tend to be closer to the center than the outsides, you notice the defects more.

But this only applies to individual large shapes (deliberative, non-immersive reading); in the case of boumas (which kick in during immersive reading) it must be more complicated.

hhp

Hm, that makes sense. But that would mean this is actually pretty irrelevant for text type design, in the sense that a simple conclusion à la «the shapes of the counters are actually *more* important to get right than the outsides» would be bogus. Right?

I would argue it is relevant for text design because being distracted from text might make it seem even more glaring.

Ouah! I'm gonna think about this with reference to Mint which I'm now working on again. Agree the inner contour distortions are way more jarring. I'll come back to you when I have something worth reporting ;)

Here's another theory. Inner counters are much smaller, therefore distortions are proportionally greater than the same distortions on the outer contours.

«Inner counters are much smaller, therefore distortions are proportionally greater than the same distortions on the outer contours.»

Here's a [quick] version where the distortions on the outer contour are proportionally equal to those in the inner one (the outer contour of the right hand glyph is a scaled-up version of the inner contour of the left hand one).
I agree the effect is less pronounced, but isn't it still visible? (Although now another effect kicks in too, because in the right hand one the thickness of the stroke varies more. Uh oh, I see this getting complex.)

Could you show us the same thing on the bold example (where the inner contour is substantially smaller than the outer and the stroke thickness will be relatively less disturbed)?

Well that one's tricky, because in this design the outer contour isn't close to being mathematically scale-able from the inner one; here's an approximation. (And obviously the disturbance in the outer contour is much bigger in the x-direction now, due to the unequal scaling.)

Thanks for letting us judge that. For me the same thing holds true, that the left example is more anxiety inducing (even when looked at with the head sideways). Funny, when I let the eyes defocus the opposite is true, probably because the stroke width changes more as you mentioned before.

Something I read somewhere argued that rather than the counter being background to the black shape, it is more perceived as a second layer of foreground, above the black shape. Perhaps this would account for the more profound effect as disruption of the fore-foreground. I wonder how one could prove or disprove this.

Taking something both from Hrant’s and from Ben’s theories, mine is that, since the inner contour is the relationship between two defined black/white areas, our perception of “error” is greater than the one where (the outer contour) this relationship is between a finite area and a virtually infinite one.

Nice theory, you could test that by blowing up the shapes until they occupy almost the whole visual field.

Well, or by limiting the space around the glyphs by placing them in a box or something.

@James (Dunwich Type): What I meant was, I suspect this effect might not be as obvious/relevant/applicable in text (1) because of scale: at text sizes, the distance and hence the difference between acuity/sharpness with which we perceive inner and outer contours won't be as big as when staring at these huge glyphs; and (2) because of context: depending on where we fixate in the context of other letters, the (sharpest) focus does not need to lie in the center of say an "o", as it sort-of automatically does here. (That's what I make from what Hrant said, and it seems to make sense.)

Nina... as a lettering artist I employ "edge texture" all the time, making sure that "interior shapes" are relatively clean so as not to disrupt the reading process. Now where that concept came from... basically the sum of client critiques and peer review. I have always heard in my circle that the "interior" shapes are critical. When I designed http://new.myfonts.com/fonts/ihof/p22-peanut/ you can see that I in fact kept the interiors clean.

In essence I am offering practical data and not research date, sorry :-)

That's interesting Michael, thanks for mentioning it here!

"basically the sum of client critiques and peer review"
So you'd say that the «cleaner» counters were more popular in smaller sizes as well?

Nina
In all sizes, and not just popular but preferable... it seems to draw the eye when there are irregularities. That is what they went to. When I did Katytude for Letraset I remember having over 200 points on the exterior of the O and just 4 on the inside. I can send you a PDF of Peanut if you wish to look a bit closer.

I remember doing an italic demo years ago and I messed up the l/c o so I redid it. The class was taken aback as the exteriors were essentially the same but the interior had a hiccup in it. They simply said the hiccup caused visual distress.

Michael, it's very interesting that you do that! But doesn't this have different implications in a font that's *supposed* to look weathered? I mean, if you're economizing points (at least I think that's part of what you mean) then isn't putting more on the outside actually the opposite of what Nina is getting at? Or do you think there's an inside/outside difference in terms of distressed versus weathered?

hhp

P.S. I have many old letterpress books and, as you know, there can be a bit of an edge in the printing (interiors). It somehow does not offend me; but in my lettering I really try and clean it up.

Hrant... it is difficult to explain, but my gut feeling is that the smooth interior, of say an O, is less offensive to the "reading/glancing eye" as that is where the focal point seems to be. UNSCIENTIFIC of course :-)

I am not sure that I understand the difference between weathered and distressed. I think they are terms that have been invented to describe what I call edge texture... interior or exterior. Other terms in lettering include notching, cutting, roughing etc.

Distressed has always been to me, taking Helvetica Bold and putting lines through it to make it look old like a rusted sign :-)

It might be a function of our primal being back when we were intensely interested in seeing both food and predator in our horizon. I theorize that the outer boundary is akin to the forest mantle and acts to frame what is inside our point of view. We would not be alarmed to see foliage surrounding our viewpoint, but if a predator or an enemy were to become framed by that vegetative outer area, we would have reason to be quite alarmed and surely more attentive to that potential danger. Although not impossible, it is certainly less likely that a predator or foe could position himself as part of the framing foliage than to intrude on the space that it surrounds. This also may partly be why humans attend more to figure than to ground in things they see. I think we just focus quickly on what we are conditioned to think is crucial to life and sustenance, even if we just ignore all other visual stimuli.

>UNSCIENTIFIC of course

Hey science has to explain what we experience. Our experience controls, not the other way around.

I think this has to do with the "salience" of counters in reading: that the brain really relies on these in identifying letters, as Peter Enneson has argued.

Spoken like a true anthropologist Dr. Chris :-) If I see a loose Cobra in my new font for the NY Zoo peeking out from an O I am out of here!

@ William... when you think of how few clues, as well as how subtle they can be, we need to distinguish letters those very things become hugely important.

"Snakes on the Plane" not withstanding we have more to fear from our own species ;-)

> the brain really relies on these in identifying letters

And not on the outsides? Of course that's very difficult to believe.

hhp