Two potential font projects — which to pursue?

Christian Thalmann

Hi all,

I'm toying around with two very different design concepts for my next font project, and I'd like to hear your opinions on them to help me decide.

—

(1) Working title: Turicum

While doodling around with my Pilot pen, I've sort of accidentally developed a writing style with high x-height (almost unicase) reminiscent of uncial, but with italic-style serifs and italic-style architecture for many letters (e.g., no round-bowled /m/n/u/w). I've been wondering whether digitizing this would make a worthwhile project. With my recent work on Volantene Script and Brilliance, it wouldn't have to start from scratch, so it would be a comparatively work-efficient project. On the other hand, I'm worried that the market for such an obscure and difficult-to-classify style might be very limited.

The first attached image shows a sample on paper. I find the /e as depicted here too crowded, and have since adopted a more traditional-looking form with only a simple, diagonal midbar. I might be able to make the original form work electronically, though, by "cheating" a bit on the stroke weight. In any case, I'm thinking of adding a selection of stylistic alternates, perhaps even a simplified Textura-style series of capitals (like that /C on the upper left in the image).

—

(2) Working Title: Kensai

I like the notion of Square Word Calligraphy, and thought something of the sort would be feasible as an OpenType font. Rather than arbitrarily long words, I'd just make groups of 1–4 letters that would arrange themselves automatically using ligatures and contextual alternate rules. I've started making some sample characters in Glyphs (see second attached image).

While there are a number of fake asian fonts around, I believe I'd be the first to provide for multi-letter glyphs, so there's a potential novelty factor there that might work in my favor. Then again, the scope of use for such a font would be extremely limited. Unlike Turicum, this would be a lot of from-scratch handiwork, so I'm a bit reluctant to commit to it.

—

What do you think?

Cheers, Christian

James Puckett

The first option. Fake Asian scripts are no longer usable in commercial work because using them causes an immediate backlash. So a fake Asian font has almost zero value as a retail product.

Christian Thalmann

Thanks James! I wasn't aware of that. Does that extend to all imitation fonts of non-Latin scripts?

James Puckett

I don’t know if other ethnic groups get offended by imitations of their writing systems.

Scott-Martin Kosofsky

I'm not aware of any serious studies of the subject, but "ethnic" types have been part of the American scene since the 1880s, at least, possibly a decade or more earlier. I suspect they were influenced by the era of gigantic immigration into the United States, that started in the 1840s, but developed real steam between 1880 and 1924 (when the gates were closed to many groups). MacKellar, Smiths & Jordan not only made such types, but created elaborate border systems on exotic and historical themes to be used with them. The American foundries fell over themselves making faux-Chinese types, as well as faux-Hebrew, Persian, Russian, Mexican, Indian, Siamese and any other culture they could think of. Though they became kitsch almost immediately, they never died out. They were very much around during the Letraset era.

In our super-sensitivity to political correctness, we can find people who, like James, wonder if they're offensive to anyone. I would say the evidence suggests they are not, though in some uses they could be. It's one thing for Chinese restaurant or Jewish delicatessen owners to use them with a wink, or to use them for comic affect, but another thing entirely to see them on an anti-immigration or a neo-Nazi website. There are other types, too, such as some forms of blackletter, that could be used in an offensive way. Maybe even Helvetica--"Swiss Watchmakers, Keep Your Overpriced Crap at Home!"

I don't know how these things play in other countries. It's all a matter of context. The real problem for you, Christian, is whether you can make a faux-Chinese type that isn't kitsch. Some years ago, Cynthia Hollandsworth made a Japanese-influenced roman call Hiroshige, for ITC. It was very well drawn, even as an italic, though it got very silly when have weights were added. (ITC had no shame about that sort of thing.)

Jan Willem Wennekes

Leaving the whole ethnic discussion alone, I'd prefer the first one over the second one just because I feel it's more interesting and more beautiful in potential than the second one. As a designer I'd find more use for the first one than the second one as well. Just my two cents of course...

Christian Thalmann

@ Scott: Thanks for the background!

@ Jan: That's a good point, actually. I can see Turicum in a wide range of applications, only some of which are shared with traditional uncial fonts. Labels for home-made jam come to mind, for example. I guess I'll give it a try, then.

Jan Willem Wennekes

Good luck with it Christian!

Christian Thalmann

I've been doodling around with Turicum a bit:



I guess the /x might need a bit of cheating — right now its stem is right at the implied pen width, but it looks too heavy.

Right now it all still looks a bit haphazard; I might tone down some of the more eclectic glyphs to get more consistent rhythm. On the other hand, that might be partly due to the unnatural clumping of letters in the simple alphabet dump. It looks more rhythmic when used to typeset actual text:

John Hudson

The m is a little too narrow. In this style, with such generous round letters, the counters shouldn't be so compressed relative to the n.

The foot terminal stroke on the r is unnecessary and looks odd.

The e would look better if the hairline connected higher up the left side, with a smaller counter better balanced with that of the a.

The last letter doesn't read as z. The impression is that the lower loop is made counter-clockwise from the descending diagonal. That's not the ductus of a z, which needs to give the impression of a zig-zag even when terminating in a loop.

I like the fact that you're sticking with appropriate forms of the f, g, w and y. A lot of calligraphic typefaces based on historical models are messed up by trying to introduce anachronistic letter shapes based on typographic norms that originated in entirely different styles and periods. They end up looking stiff and artificial.

Craig Eliason

I agree with John about the /m/e/z/. In addition to widening the /m/, is there a way to make it less stiff (and /n/ too)?
I think the /p/ looks curled too tightly at the baseline.
I too like the structure of /f/ though I think its arms get a little too fiddly perhaps.

Eris Alar

FWIW as someone not highly knowledgeable of historical calligraphy, the g is hard to read, to the point of being unreadable for me.

Christian Thalmann

Thanks for the feedback, everyone!

@ John:

I agree on the /m, /z, and /e. The counter of the /e is so big right now because I want to keep the crossbar thin, so it needs to be at pen angle. If I make the eye smaller, the whole glyph shrinks... in my hand-penned samples, I originally used a different form with a small eye and a dedicated horizontal outstroke, but I ended up ditching it because it looked too black and crowded. I suppose I could try to make it work in the electronic version by cheating on the stroke widths...

As for the /r, I'm actually quite fond of it. The blackletter-style foot gives it a more pleasant texture and picks up the theme of diagonal on- and off-strokes that defines the font flavor. Texture is going to be very important to this font; I intend to make a lot of contextual alternates to fill out holes. For instance, the crossbar of /t and the onstrokes of /n-like letters should extend backwards when following an /a so as to fill the whitespace there. I would expect a footless /r to stand out rather garishly, then. In any case, I will make a footless /r and offer it as an alternate. I plan to release several versions of the font with different alternates implemented as default, as with Gryffensee, to make them easily usable in typographically unsavvy environments.

@ Craig:

Concerning stiffness: Currently, all vertical strokes have 100% vertical sides, which is probably why they look stiff. I'm thinking of either giving them a subtle bend, or maybe caving in the sides just a tiny bit as I did for Brilliance. However, I want to avoid getting to "fiddly" with the font, its overall appearance should be one of elegant simplicity.

I see what you mean on the /p; the curl is unique in the font and stands out. The curl of the /j is a similar though less pronounced problem. I guess I could just terminate the /j like the /p's descender, but I'm not sure how to solve the /p's bowl. If I pull the circular stroke through, there will be a black clump at the junction (all other letters manage to avoid thick-on-thick stroke junctions by design). If I terminate the stroke while it is still thin, I fear the glyph will look like a wynn (which I might actually implement as a glyph of its own!). Any ideas here?

I agree that the /f is too fiddly, but how to solve it...? I think I've tried two swooshes as well as two diamonds, and they both look worse than the swoosh-and-diamond solution. Maybe I could get rid of the "serif" on the diamond...?

@ Simon:

I agree that the /g is not easy to read for people used to modern typography. I will certainly offer more legible stylistic alternates for it. The problem is that my choice of broad nib and small extender space make it very difficult to squeeze in a two-storey /g, or even just an italic-style /g. I like the insular-style /g for its efficient use of space.

Christian Thalmann

Alright, I've been doing some tinkering.



I'm not quite happy with the /p yet, and I've only made a small change to the /f (does it look less fidgety now?). I also made a footless /r, but definitely prefer the original.

Here's the alternate geometry for /e I've mentioned, by the way. I find it works rather better than I expected:

Christian Thalmann

Here's an example of the many contextual alternates I intend to implement in order to make the texture more consistent. Note how the crossbar of the /t and /t_t extends backwards after an /a, and forwards before an /o-like curve. Likewise, characters with an onstroke at x-height, like /n, have an alternate that extends the onstroke backwards after an /a. More of that to come.

Christian Thalmann

And the same for /g:

Arthur Reinders Folmer

Seeing your finish on your last projects I'm looking forward to the finished project! (and a bit late on the bandwagon, but you made the right choice which one to pursue.)

The alternates indeed help it tying together (more at least than some heavy kerning would do) Interested to see the caps construction method, but Unical construction can at least lead half the way there. (actually, the construction of your alternate /e already goes into that direction.)

I'm actually quite fond of the short tailed /z, your latest one is an improvement, but perhaps it still is not zig-zag enough for readability (which is relative seeing you could also fit the long s style s's in this project, but that would provide even more interesting construction problems.)

Christian Thalmann

Hi Arthur, thanks for your feedback!

When writing in this style by hand, I sometimes lean towards simplified textura-style capitals and sometimes more towards uncial-style ones. But since I'll most likely release the face in several stylistic versions (like Gryffensee), I might just make both kinds, or even more than just the two. For instance, I'd like a textura-style /G, a simple high-legibility /G, and a /G based on a scaled-up version of the current lowercase /G. I've tried the latter on paper, and I like it a lot.

I personally consider the /z very legible, but that's because I learned the two-crescents shape for hand-written /z in primary school. I'll certainly make a zig-zag version for the high-legibility cut, but I expect it'll look rather jarring. I also can't really make the zig-zag stronger in the current form of the letter without lowering the descender, and there's just not that much space for that. Maybe if I narrowed the whole thing down, though...? Not sure that would still fit the generous pace of the typeface, though.

Christian Thalmann

Made some more conventional italic-style alternates before tackling the caps:



@ Arthur: I do have a /longs already, but it currently lacks the necessary ligatures with /b/d/h/k/l/thorn to work well. I also just made a round /r:

Jacques Le Bailly

Your 'r' is looking too much like an 'e' and/or an upside down minuscule 'a' IMHO. I understand that you want to get the rythm and color balanced, but the elongated stroke at the bottom is too large and the inwards curling flag creates an unnecessary counter.

Christian Thalmann

Very well, I'm making the footless /r the default for the italic-flavored edition. I still prefer the footed one, though, even in the italic-flavored context. But I guess I could still keep it in as a stylistic alternate in that edition...

I also made a bunch of /longs-ligatures and a /germandbls:

Craig Eliason

The "italic-style" /j/ is fine but the /g/ and /y/ look goofy in terms of proportions. I think you have to decide between that structure of descender and that short of a descender space.
(The too-tightly curled /r/ issue that Jacques is talking about is related to my earlier comments on the /p/, and /f/.)

Christian Thalmann

I agree that /g/y look a bit disproportionate, but it seems to work for Pluto, too...

http://cdn.myfonts.net/s/aw/720x360/130/0/66825.png

I don't feel like expanding the descender space just for those alternates. Maybe I can make the descender loop large enough to break the disproportionality instead...

How does this work? Note that I also made the flag of the /r wider and less strongly curled. I tried a diamond-shaped flag before (as in the middle of /f), but it looked stupid.

Craig Eliason

Looks good. That little change on the curls makes a big difference.
On the /g/ etc., remember that you can cheat up the bottom of the top bowl too.

Christian Thalmann

On the /g/ etc., remember that you can cheat up the bottom of the top bowl too.

I already did. ;o)

LeMo aka PatternMan aka Frank E Blokland

Jacques: ‘Your 'r' is looking too much like an 'e' and/or an upside down minuscule 'a' IMHO.’

Looking at the origin of this r, which was in gothic hands used only after round or formerly round letters, I don’t think this shape is too bad. I just made a rough note with a broad nib (below). I also made one for the eszett.



Christian makes a deliberate ratatouille of historic hands, mixing half-uncials with rotunda styles and everything in between. The rotunda has an interesting feature considering the stem letters; the nib is rotated in several cases to get a horizontal ending of the stroke. This could be used for this type too IMHO, to make it somewhat more sturdy. Or something in between, like below.

Christian Thalmann

Hi Frank! Thanks for your note. You have pretty handwriting. )

Looking at the origin of this r, which was in gothic hands used only after round or formally round letters, I don’t think this shape is too bad.

I believe what Craig and Jacques addressed was not the shape of the round /r, but of the regular stem-and-flag /r I had posted further up. The flag was curled too strongly and thus seemingly created a mini-counter. I counteracted that by making the flat longer and less curled-up.

Christian makes a deliberate ratatouille of historic hands

While this is probably an accurate way of describing it — I guess all of these shapes must have appeared in one historic style or another — I prefer thinking of it as developing my own contemporary style. Hiſtoric revivals are not my thing; I know too little about the history on one hand, and on the other hand prefer to keep my creative options unconstrained.

I suck at writing italic with the broad nib for some reason (perhaps my brain prefers to work with shapes rather than strokes?), and while I feel naturally at home with textura, it's just too stern and space-gobbling for many purposes. I wanted something compact, friendly and lively that I can actually write reliably, so this is what I converged on after some penstorming. I did have a nebulous notion of uncial in mind, but I emphasized and sometimes exaggerated the diagonal on- and offstrokes in a way that reminded me of italic. I'm not sure there's an established term for this particular combination — is there? If not, all the better! I might call it "cursive uncial" or something.

The rotunda has an interesting feature considering the stem letters; the nib is rotated in several cases to get a horizontal ending of the stroke. This could be used for this type too IMHO, to make it somewhat more sturdy. Or something in between, like below.

Interesting! I guess this could work on the feet of /n/m, etc. I wouldn't want to do it to the top of /u; I find the current design with two cursive onstrokes particularly fetching and downright prototypical for this style. It can't be too counter-historical either, given that oldstyle serifs have two diagonal serifs on the top of /u as well.

I believe I see what you mean by "sturdy": Your hand-written sample is handsomely grounded and regal. The flatter pen angle adds to that. However, I'm aiming for a lively and flowing rhythm with my typeface, so adding more sturdiness might be counterproductive to that particular goal. I've also actively tried to keep all the shapes self-consistent with a fixed pen angle; I'm worried that a terminal with an "off" angle might break the visual consistency.

I like that triangular /t top! Maybe I should try that for my typeface as well.

Christian Thalmann

First trials with capitals:


Oh, and could a curly-topped /s work better than the tilde-topped one?

Jack Jennings

Without meaning to criticize, I think that you'll have a hard time defending the historical g for in your alphabet the more forms you introduce that are foreign to the uncial script.

Your majuscule A is without precedent. Other than roman sq. caps, different uncial scripts had their own set of decorative majuscule that you could draw from. As far as I can recall, everything I've seen uses the three-stroke A ductus, or leaves the bowl open (sometimes doubling back on itself rather than connecting to the stem). I think that you'll have better luck if the bowl originates closer to the top of the stem in either case.

Jack Jennings

Lastly, I agree with Craig about the alt /g and /y. In my opinion Pluto doesn't pull them off elegantly either.

Last lastly, the "tongue" of /e and /f looks too fussy to my eye. I think that you'll get more milage out of something less wavy, more horizontal. Right now /f looks like /x in some texture hands, and I think that it's because of the diamond-shaped crossbar.

Christian Thalmann

Without meaning to criticize, I think that you'll have a hard time defending the historical g for in your alphabet the more forms you introduce that are foreign to the uncial script.

This is a critique thread, so by all means do criticize!

My rationale for the current /g is that it's a two-storey design that fits the space requirements, the pen logic, the rhythm of the face, and my personal taste. I'd rather not use a single-storey design for the base cut, so as to contrast with the "italic" cut. Anyway, as I said, the final font family will offer many flavors of the face, so there's certainly room for a third or fourth design for /g to suit everyone's taste and legibility requirements.

What shape would you recommend? There's certainly no space for a carolingian /g; I've tried.

Your majuscule A is without precedent.

Apparently not quite: http://www.myfonts.com/fonts/glc/750-latin-uncial/

I don't mind breaking historic tradition, as long as the result is æsthetic and harmonious with the rest of the font. Do you also object to the current design on these grounds? Personally, I like it that way.

I'm wary of trying out a triangular design for /A. The font currently has a pleasant roundedness and bounce, which a triangular /A might spoil. I do intend to make a font with textura-style capitals, but for the base font, I'm aiming for caps that don't try to steal the thunder from the lowercase.

How's this for /e and /f? And what do you think of the idea to use the curl-topped /s as the base form and the tilde-topped one before round letters? And does the larger bowl of /A help?