What makes FF Legato so unique?

[Member ken…]

> Are there any text faces that did this previously? (Serious question, I’m ready to be educated. :-) )

Okay, this is separate from the pre-/post-digital tools debate.

I think Cyrus Highsmith's Quiosco demonstrates the same kind of independence of interior/exterior contours.

 
I can't swear that it is strictly "previous" to Legato. Certainly not by much. The first appearance of Quiosco was 2002 in La Prensa Gráfica. I believe that the published date of Legato is 2004. Who knows how long he was working on it or when he started.

I'm pretty sure Cyrus knew Evert, and most likely respected him. But I don't know if there were any specific influences either direction. As I've said, I believe this is more of a zeitgeist thing.

Cyrus's stated influence in this direction was Dwiggins.

-- K.

[Member ken…]

Altaira -- Just noticed your request for more Dwiggins. Ha, there must be plenty scattered around Typophile from through the years. People are probably tired of hearing me wax on about WAD. Try searching for "M-formula." There was a "Chunky" thread, I think. Also, that "Cut and Curved" thread of Eben's from last year. Florian can probably help you out. ;-)

I've got plenty more, if that doesn't sate your appetite or give you the kind of specificity you're looking for. But we'll end up wandering far afield of Frode's interest in Legato. ;-)

-- K.

[Member Wil…]

On the readability of Legato, which Peter again notes, I do agree with many here that Legato is unusually readable for a sans. And I think a lot has to do with Boemsma's success in breaking up the picket fence. But other stressed sans do it, including Lydian, Optima and Cyrus Highsmith's Amira--which also violates pen rules systematically.

Nina, I think that what you report about people reacting saying it looks more pen drawn is actually correct, in that the modulations are there as in a pen-drawn stroke, even though Bloemsma's modulations violate pen rules. For example, Carl Crossgrove's Beorcana is quite readable and has a strong pen-drawn quality.

So while I appreciate the readability of Legato, I don't think the forward stress of the black that is distinctive to Legato is the only way or best way aesthetically to achieve what it does by way of readability.

[Member nin…]

Peter, I have tried reading the post you referenced again, but I have to admit it's still a bit over my head. :-/

Like, would you mind getting back to this?
"And he has done this not by starting from translation, but from an introduction of inner form / outer form counter-rotations in an expansion based context."
I'm guessing I don't understand the concept of expansion here. Trying to reverse engineer to a degree what Bloemsma did, when I rotate the contours back in that "o", there's definitely still some contrast there. I wonder where this comes from. Is it possible/likely to make this with a pen?

(Image: "o" from Legato Regular, outer contour rotated counterclockwise by 4.5°, inner contour rotated clockwise by 0.5°)

I'm now reading up on the connection between notan and readability. Great input.

By the way, do you (or anyone else for that matter) happen to have the first link archived that you cite in that post, "About Legato" from Evert Bloemsma's web site? Archive.org has it, but without the images. Looks like there was a PDF version too (which hasn't been archived).

"In digital fonts it is still the font and not the outline that is designed."
I was not trying to suggest that "we design vectors". My point was probably that the movable, flexible, immaterial, and –most importantly in this context– independent digital outline, which has become the "interface" through which the designer shapes a font (and thus, a metaphor of sorts), has a notable effect on how fonts are seen/thought_of; though not necessarily in the way of total identification of the product with its toolwise representation.

Kent, thanks for posting Quiosco. That does look strange close up! I was surprised to see how well it prints/reads. Although it feels much less radical than Legato; the "independentness" of the inner/outer contours seems to mainly come into play in details (like the way he "thinned" the joins), while some important stems (like say the left one of the "n") are still vertical both inside and out, while those in Legato look pretty… twisted :-). Am I not seeing something?
Oh, and cheers for the WAD keywords. Will do some reading up!

"But we’ll end up wandering far afield of Frode’s interest in Legato. ;-)"
Well he hasn't intervened… yet. I hope he will protest if this veers off too much for his interests… meanwhile, at least I am shamelessly capitalizing on his question for my own learning. ;-)

"the modulations are there as in a pen-drawn stroke, even though Bloemsma’s modulations violate pen rules"
So are you suggesting that it's the concept of modulation in and of itself that makes for its heightened readability, and not necessarily how it's executed – pen-conformingly or not?

[Member ken…]

I guess it depends on how you want to go about defining your "independentness." I mean, ultimately, if a design is successful the various contours are going to be bound together in some fashion.

Yes, in Quiosco there are still straight parallel stems. I was looking more at the interaction/independence of the inner and outer curve of the n arch -- these have a radically different thrust from one another, creating an interesting tension, which was the focus of Cyrus's formal exploration. (Note also the non-parallel right stem that results from this tension.)

> while those in Legato look pretty… twisted :-).

Non-parallel, swelling or flaring stems are nothing terribly innovative in and of themselves, of course.

I believe in this case they arise out of Bloemsma explicitly stated aversion to straight lines. Which makes his aesthetic rather ironically non-digital -- since you could argue that straight, square, plumb, and parallel are in large measure inherent in digital tools.

For another interesting comparison, take a look at what Frutiger was doing in Icone (ca. 1980). Legato on the left, Icone on the right:

[Member enn…]

[Nina] Like, would you mind getting back to this?

The analysis is far too compact. I'll try to get back to it on Sunday.

Here is a stitched screen grab of the original Bloemsma page (be sure to scroll horizontally to see the additional content on the right):

[Member Nic…]

It's a nicely drawn irregular sans.
What makes it unique is the amount of BS that's been attached to it.

As Nina has demonstrated by reversing Bloemsma's rotation cue, there is the tell-tale residue of skew, confirming the face's chirographic construction. So basically it is a humanist sans, with the counters made slightly more symmetrical. I think Bloemsma designed primarily with hand and eye, and that the detailed rationalization is post facto, and not very convincing. Type founders have been thinning out the thickest part of the chirographic stroke from Jenson to Quiosco (compare "p"s).

I suspect that he had the general idea that an irregular sans could be drawn with some tension, by chafing against the usual precepts of simple geometry and chirography, and that there might be some readability benefit in that. But the face still has a predominantly chirographic stress, as can be seen by flipping the "o"--Legato above, Maiandra in the middle, and most subtle of all, Productus, below.

Legato has something of the feel of an old-fashioned humanist sans, say Goudy Sans; its slightly uncrisp execution, and the fact that the letter shapes themselves are not particularly novel, might contribute to its lack of popularity, although it does have the even width and semi-condensed proportions of the contemporary sans genre descended from Meta.

I'm not knocking the face, just questioning the theory.

[Member ken…]

Oh, I have a copy of that PDF about Legato from Peter's screen grab of Bloemsma's page. I don't think I can post a PDF here. If you send me an e-mail off-line, Altaira, I'll pass it to you.

[Member nin…]

"As Nina has demonstrated by reversing Bloemsma’s rotation cue, there is the tell-tale residue of skew, confirming the face’s chirographic construction."

1) Is every font that has this "skew" to its "stroke" by definition chirographically constructed / constructable?

2) Is that a Yes to my question if those "de-rotated" contours I showed could be produced with a pen / single stroke? (I suck at calligraphy, so I won't be able to confirm or disprove this myself.)

Nick, even if your answer is yes to question 2, I don't think that could confirm the face is "chirographically constructed". I think the subsequent de-linking and individual rotation of the two contours matters a lot in this design. & that means the font that resulted in the end cannot really have been "chirographically constructed" – at most, it may have been "deconstructed" from an assumed chirographically-informed base structure. (While of course we're really only looking at one glyph here.)

Peter, Kent: Thank you! Kent, I've dropped you an email.
I wonder, William, what you make of Bloemsma's illustration of how Legato's construction principle and contrast helps readability – that very page seems to make a lot of sense to me, and I'm beginning to suspect it's really about a combination of (a) breaking up the "picket fence" (in terms of doing away with straight verticals, and generally introducing more "movement"), which tends to raise perceived coherence of letter sequences, and (b) its funky contrast/stress, which makes it even less "static" (which I'm actually no big fan of personally, aesthetically speaking, but it's very interesting to see what it can / may be able to do).
That bottom-most illustration in Peter's screen grab is especially telling I think; some of the text seems to be cut off: "The illustration below compares the sanserifs Univers (emphasizing vertical parts), FF Balance (emphasizing horizontal parts) and FF Legato (emphasizing diagonals from the upper left to the lower right). Each one of them builds a different rhythm, creates a different interaction between the characters; a different continuity."

The Icone comparison is interesting. I'd like to get back to this but need to look at it in depth. (Fortunately, I just got myself a copy of this wonderful new Frutiger monograph today!)

[Member Fro…]

But we’ll end up wandering far afield of Frode’s interest in Legato
I'm still here!

I get the importance in Hrant's observation: Moving from the simulation of old tools towards actually taking advantage of the digital tools and methods to create something new, though I'm not sure how it's a revolution in terms of readability.

[Member nin…]

As a side note (but maybe not really), the Icone comparison is quite interesting, also with respect to the relations between design and media. In the monograph (Adrian Frutiger: Typefaces. The complete works), Frutiger explains how the design of Icone (1978, published 1980) was a direct response to the new digital medium, although in terms of striving to offset its negative effects (digital distortion): "Instead of 'lines' that could easily be defined geometrically, [Icone] is characterized by a rather 'wild' stroke design. Curves and asymmetrically flaring terminals are among its key characteristics. There are no traditional serifs, which would hinder slanting." (My translation, I have the German edition.)

Icone on the left, Legato on the right:

Now while there seems to be a bit of similarity between these glyphs, to me they also show the notable difference in the thinking behind their designs: Icone's "t" is "flowing" throughout, and harmoniously. Both sides of the stem are curved, and stretching the imagination a bit, the glyph could still be seen as "painted" - and it was not Frutiger's intention to depart from the chirographic model, he just "gummied" the contours (in fact calling Icone "gummy type") to make it less vulnerable to distortion. In Legato, the "t" has a lot more tension between the contours, a much more strongly differentiated "pull" of the curves - which underpins the notion of the tension between the (more) independent contours being a driving force behind its design.

[Member dan…]

Kent, can I have that pdf too? I give you my e.mail.

From a geometrical point of view I think the operation on the contour of the 'o' is not rotation but skewing (I think it is what Nick Shinn wrote). In that way the upper right and lower left regions of the letter are thickened.
The trick with black/white is, I think, that you have two opposite movements: the black is leaning forward while the inner white is slightly leaning backward. Maybe this “counter-movements” makes the letter appear upright while it is slanted, I don‘t know. But the effect is a diagonal stress forward (maybe the interconnection wanted by Bloemsma) without being an ‘italic’. I suppose that stress (that is the opposite of the red dashed line in the image above) is responsible for the “calligraphic ductus” seen by non type obsessed persons (a more natural 'flow')

Looking at the t that Nina posted, I see the same stress (look at the bottom-left part) obtained in the same way: the outer contour is pulled along a diagonal, while the inner contour remains straight.

What I find interesting is that the conceptualization Bloemsma did is actually based on the interaction of surfaces (black area against white area), that is a conceptualization not done in terms of lines (the so-called anti-chirographic point of view). Obviously you can re-tell the whole story in terms of marking tools and strokes.

For the discussion on the tools. I think that vector drawing softwares, so based on the metaphor of drawing board, with splines etc., are actually a conceptual obstacle for thinking in terms of surfaces: you are forced to manipulate dematerialized ‘curves’ (that is good mathematically, instead.)

[Member Wil…]

>what you make of Bloemsma’s illustration of how Legato’s construction principle and contrast helps readability

I think that the modulation and also relatively wide spacing helps readability, but I don't buy his argument that the "forward motion" of the stress helps particularly. After all, italics have this forward motion, and are generally less readable. His innovation keeps the letters looking upright with some of this italic effect, but I suspect it's not much of a benefit. In other words, if you redrew Legato with just a normal old style stress, I suspect you wouldn't lose anything as far as readablity.

Another interesting example, also obviously influenced by Dwiggins, is John Downer's SamSans, which was until recently the 'Featured Face' here on Typophile. That has a lot of angular inner outlines and smooth outer, and modulation. It is a little narrow for my liking--though a narrow informal sans was indeed his brief--but I think extremely readable for being a narrow sans. Interestingly, his "e" has some of the forward/backward motion of Legato--it is from 1993!--but not generally.

[Member nin…]

"I don’t buy his argument that the “forward motion” of the stress helps particularly. After all, italics have this forward motion, and are generally less readable."
But you're comparing two much different implementations of the idea of "forward motion". Italics have a lot of other features that (I'd expect) influence their readability (the slant angle, and the classically narrower widths, to name but two). Also, Bloemsma isn't just talking about "'forward motion' of the stress" – he's defining that more closely as the tension between the black and the intra-/inter-letter space, that breaks up the isolation of individual letters; and I have no reason so far not to believe (a) what he says, (b) my eyes.

"In other words, if you redrew Legato with just a normal old style stress, I suspect you wouldn’t lose anything as far as readablity."
Now *that* would be the ultimate thing to try. I'd contest it. Hell, I'd bet a bottle of decent wine that you would (lose something in terms of readability, that is). Except even that surely wouldn't yield a clear and uncontested result (who's going to do it, and how exactly, and how are you going to test the results "independently"?).

SamSans: I don't know, this looks a *lot* more chirographically-informed to me than Legato. Yes there are some dissonances between the contours (like in the "s"), but they're much more gentle, and follow a "logically"-appearing, much more easily discernible pattern.

[Member Wil…]

Yeah doing the experiment of an old-style Legato would be an effort--but interesting. I'm not about to do it soon, though :)

As far as SamSans, yes it has the feel of the hand in it, in a good way I think, but also its lines have constant contradiction of what a pen would draw.

I also don't buy the chirographic=bad boogyman of Hrant. I think the pen does some things that helped readability, and some that hurt. And since Jenson, type designers have been modifying those things that hurt. Departing from the pen isn't inherently good or bad. Our letter shapes are based on the broad pen, so going too far away is always tricky to pull off successfully.

Ultimately the test of whether Legato's unique stress pattern is a positive breakthrough is to do designs that follow it, but are more successful aesthetically than Legato. And if there is something else there having to do with notan, then develop that. Each person can have a go. I am just skeptical that this is a promising direction.

[Member enn…]

Nina, I was unable to get to my promised elaboration over the weekend, and I'm in a bit of a time crunch now, but can you let me know if the outlines and control points you show in the images above (12 March 2009 / 3:21 and 12 March 2009 / 9:36) are your approximations or Bloemsma's actual?

[Member nin…]

Peter, the 12.Mar.2009 3.21pm image shows the actual "o" of Legato Regular
as it appears in the font. The 13.Mar.2009 9.36am image is my approximation
of rotating the contours back.

[Member Wil…]

Note that the contours are slightly twisted or wrenched against one another ("verdreht" in Jan's FontShop image) rather than simply rotated. For example, the outer side points have a counter-clockwise rotation, whereas the outer top and bottom points are rotated clockwise. This kind of distortion of an ellipse might be achieved by some combination of skewing (slanting) and rotating. But I suspect Bloemsma just adjusted it my hand and eye until he got the effect he wanted.

Also, when I played with my semi-bold version (a free-bee from FontShop!) the line connecting outer top and bottom points are between 1.5 and 2 degrees off vertical, not 5.5 degrees. Maybe the bold is different. The whole thing is very subtly done.

[Member nin…]

Ha. That would of course render my quick rotating experiment moot.
I've been wondering about what looks like different direction of rotation between the outer side points and the upper/lower extrema. Still I read that "verdreht" as simply "rotated" (when two contours are "verdreht", it can mean they're individually rotated in opposing directions, not necessarily skewed in themselves, even though "verdreht" also means "twisted").

[Member Wil…]

I guess there's no simple description of what he did. Basically, his innovative concept is the black have forward diagonal stress and the white a backward diagonal stress, so that the diagonals dominate, but the character still looks upright.

The result, very masterfully pulled off, is readable, but also awkward looking. That wasn't a problem for him, evidently, as he said he wanted not to be constrained by a "pleasant" look.

[Member ken…]

> The 13.Mar.2009 9.36am image is my approximation of rotating the contours back.

But you didn't reestablish extrema points, right?

[Member nin…]

"But you didn’t reestablish extrema points"

Indeed I didn't – I was wary of messing with the curves. Should I have?

[Member enn…]

I also read verdreht as rotated, or turned, though I see some shearing from a straightword expansion with contrast-reduction base prior to the + & - turning of the outlines. I think the shearing is effected by moving left and right-side extremata up or down / top and bottom extremata left and right.

Bill, I think the result is awkward-looking only at display sizes and when gauged aesthetically. Optical-grammatically in the context of the word at continuous text sizes I think the concept makes great and unobtrusive functional sense. As well, my sense of how Bloemsma works is that he works more systematically than you seem to think. But that's what I'll try to show.

[Member ken…]

> Indeed I didn’t – I was wary of messing with the curves. Should I have?

Not necessarily. You're right, that probably would've compromised the curves in some way.

It just would've been interesting to see how the extrema align in the rotated version. Since many of us are used to seeing the points at the extrema, you have to look carefully to realize those points aren't and adjust your evaluation of the shapes accordingly.

Plus, it might help draw out any pre-rotation adaptations to the ovals. (I mean that figuratively, not literally -- who knows how Bloemsma actually arrived at his final shapes.)

[Member Wil…]

Peter, I think Bloemsma was extremely systematic, perhaps even unwisely so. But I suspect the system was in what he wanted visually, and not in doing the same mechanical steps on things. I don't think you can get things to look the same by doing the same thing to different shapes. The eye is too complicated and changes what we see vs what's on the page or screen in ways we don't fully understand. So I would bet he just fiddled with the glyphs until he got that crossed diagonal effect he wanted.