hrant's picture

Here's the main weight of Patria, my first "serious" Latin text face. It's part of a larger system which contains "counter-balanced" Armenian fonts (it's a long story - as you might suspect).

You'll need to do a fair amount of zooming, and please pardon the occasional outline infidelity that Flash introduces (like in the inner part of the lc "j"'s tail, the top-right of the numeral "2", et alia).

I'm daring to cast Patria as a next-generation newspaper face (or at least a face fit for magazines). Its x-height is somewhat small for this purpose, because I place more value on boumas (word shapes) than most designers. Also, it actually sets narrower than it looks, maybe because the open, generous forms increase its apparent size/width.

I've tried to give Patria a full-bodied rigidity (read: masculinity) that seems to me to be lacking in the field of text face design. The Carolingian structures are generally very fluid/feminine, which is nice, except I feel there's too much of it now. Blackletter is very masculine, but it's not really usable any more (or at least not right now). The post-modern boys use a lot of rigidity, but they tend to apply it too literally to allow for immersive reading - I've tried to be more subtle. On the other hand, in certain places I've consciously sacrificed readability in order to reinforce the design's character (like in making the lc "s" somewhat wide).


I hadn't looked at Patria for over two months, but now that I'm re-evaluating it, I can see a few things I'd like to change, pending feedback from you guys:

1. I want to make the beaks that come off of horizontal bars (like in the UC "F") much larger.

2. I'm going to rethink the numerals. I'm considering making them Carter-style ("hybrid") numerals, but the all-horizontal stress I really cherish. Any ideas?

3. I'm also unhappy with a number of the non-alphabetic glyphs, but I'll wait for you to tell me which ones *you* don't like! And I guess that applies to the *alphabetic* glyphs too...

4. I personally prefer the alternate UC "U", but hesitate to place it as the primary form.

5. Are the accents too small?

6. In the Italics: even though I want it to be rigid like the Roman (in fact, I like it to be as close to the Roman as aesthetically and functionally possible), I think I'm going to make the beaks/terminals (like in the lc "s") softer in contour.

7. The spacing was a rush job - it needs to be redone. Also, I'm starting to think that it's too loose overall, unless I should account for severe ink bleed - in case my original target medium (newspapers) is reasonable. Lastly, I've yet to add any UC-UC kerning (there are just some manual tweaks right now). For this, I'm considering giving all UC pairs wide spacing, since it helps all-caps setting. Crazy?

8. I'm considering making midcaps (like in Fenway). If I do, then I might implement that overall loose UC spacing I just mentioned for the midcaps, but not the regular "full" caps.

9. I will eventually put trapping in the font - the degree of trapping depends very much on the best usage medium for this face.


I have a lot of trouble evaluating my own work (especially something outside Armenian, my specialty), and have to constantly fight myself to find the right compromises, so any insight would be greatly appreciated.


Stephen Coles's picture


I'll start by saying that the italic is very
lovely, but the overhangs on the 'f' and
'r' distract.

The beaks bigger? A la Eureka?

Why hesitate on the 'U'? I assume you
mean a Perpetuish 'U'? Go for it. It works.


mart's picture

An impressive piece of work that is swimming in subtlety. It is very readable, from what I can tell of the small paragraph.

I can tell you what I don't like far easier than what I do, because I like the sum of the parts, but not neccessarily the parts themselves.

I don't like the numbers at all. I think you should go back to the drawing board with them. I don't think any newspaper designer would want to use these as they're more designery (is that a word?) than utilitarian.

The italic is a little too subtly angled. The word "system" hardly stands out at all in that sample paragraph. Also, given the overall aesthetic I don't think it's neccessary to make the italic version so similar in lettershape to the regular.

I cannot see this working as a headline or subheading font, unless you have something up your sleeve for the bold and black versions. The individual letters are not pretty enough, at least not in this regular version, and at least not in my opinion.

The "g": I would have thought that with a vertically stressed font you should avoid the diagonal swoop of the top of the lower bowl. Apart from the "s" this is the only letter that doesn't conform vertically.

That "U" works fine, like Stephen says. But apart from that one instance I can only say that I like the overall effect, but I can't point to anything specific as I think that for the most part you've successfully designed a font that works for word shapes and high readability.

Stephen Coles's picture

Actually, I disagree with Martin on the italics.
I think they stand out *too* much, and only
because the weight is too light. Even it up with
the roman and keep the angle as it is. I love
the angle.


Miss Tiffany's picture

I'm not sure I would 'go back to the drawing board' with the figures, but now that I look at them again, maybe there is too much 'scotch' in them? Well, in my mind anyway. The '2', '4' and '6' seem the most in need a fewer curves??

hrant's picture

Here are the four weights as they currently stand:

1. The word "balance" in a given paragraph is set two weight steps away from the body.
2. I've used the alternate UC "U" in the all-caps samples.


hrant's picture

One more thing:
All four weights are currently uniwidth (a given character is of identical width in any weight), but that will change - after I figure out the proper spacing of the Regular. But I might keep the Regular and the Demi uniwidth, and I will make the set-width differences between all the weights certain fixed tracking amounts, so for instance setting the Bold with -10 tracking would make it uniwidth with the Regular.


hrant's picture

And here's a quick view of Nour ("pomegranate"), Patria's Armenian counterpart, for your enjoyment:


Stephen Coles's picture

I am digging that italic! The soft corners...
the gentle slope... oooooh. But the 'r' and
'f' still hang too far for me, and I'd argue
they do the same in the roman as well.

Not fan of the 'g' either.

Overall, it's very nice Hrant. I'd buy.


designalchemy's picture

Excellent work Hrant. I really like the "4". I can see why you have been busy. Ole

designalchemy's picture

Hrant, can you explain the design concepts of the fl and fi ligatures.

mart's picture

Well I guess it's up to me to continue to play the devil's advocate - apart from the fact that the italic is growing on me in this form.
BUT...there seems to me to be not enough difference in the weights. We're talking infinitessimal amounts, but nevertheless I think there's just not enough difference between the regular and bold. Of course, looking at the screen is less than optimal, and I'd probably be proven wrong looking at the color of a printed page, but I just wanted to raise this concern as it's of primary importance in designing a text family. (And yes - I got the weights wrong when I tried designing my own text family.)

fonthausen's picture

I hadn't much time to look at it. But a few things hit my eye...

* I have a problem with the curve character. You are using more flat curves (Lc 's' e.g.) and very round ones (Lc 'c' e.g.). I would to try to make the 'flat' curves a little bit rounder.

** Lc 'g': I would close his belly, or try to find a more subtle ending. The way it is now, the character is a little bit too much like ones we know from around 1900....

*** If this face is to become a face for newspaper, I wouldn't experiment too much with structure, as you did with Lc 'd'. I'd go more into changing details.

**** the eye of the Lc 'e' is too big, i think. The character is unbalanced.

The font is promising, although it still need a little bit of balance and unified rythm.


hrant's picture

Guys, thanks for all the feedback so far.
It's proving even more valuable than I expected.

> the italic is very lovely

Wow, thanks. I'm surprised to hear that.

> the overhangs on the 'f' and 'r' distract.

In the Italic, right? Well, I think I agree.
Probably the paucity of serifs in the Italic
is causing the beaks to stand out too much.

> The beaks bigger? A la Eureka?

(I assume you mean in the Roman only,
and not the "f" or "r" in any case.)

Eureka is a great design, and there's stuff in Patria (eg the
serifs) that's pretty similar. Some of my beaks are larger, and
some of them *need* to be larger, but I'll note one significant
difference (which is also the case with Foundry's Form, as well
as FF's new Zine, and Parkinson's Azuza): the beaks are abrupt,
not adnate like mine. Formally, that's the way to do it, since
they then match the structure of the serifs. And I almost went
that way myself, except I realized that it makes the design too
mechanical in a way. I wanted to express more depth in my design,
so I'm going for a different level -or maybe flavor- of harmony,
even though I'll surely get some flack for it.

I might still change my mind, but some people have actually
appreciated the duality, so I've kept it. But if I do change
my mind, I'd soften the serifs instead of hardening the beaks.

> Why hesitate on the 'U'?

Because many people hate it. My main reason for preferring it
is authenticity: I think that's how the Romans would've made it.

My "d" is in a similar vein, except I found a compromise
between the pure structural form (like Gill used), versus
the established one (which is way too homogenous with some
other characters -most notably the "b"- for my taste).

Oh, and there's some of my Alphabet Reform in there too... :-)

> I don't think any newspaper designer would
> want to use [those numerals] as they're more
> designery (is that a word?) than utilitarian.

You have a point.

BTW, the best word I've encountered for
what you're describing is "mannered".

> The italic is a little too subtly angled. The word "system"
> hardly stands out at all in that sample paragraph.

> I don't think it's neccessary to make the italic
> version so similar in lettershape to the regular.

This is a complex issue, and I worry about getting too deeply
into it here. I think I'll start another thread (eventually).

For the moment, let me quickly state the following (hoping
that it's not too cryptic): when it comes to text fonts, I
think there's a huge difference between what we consiously
see versus what we think we see when immersed.

> I cannot see this working as a headline or subheading font,
> unless you have something up your sleeve for the bold and
> black versions.

You're right.
The display cuts would have to be much more regular.
Note that my bold is a text bold, not a display bold.
(But as with the italics, a full elaboration of the
argument is certainly needed.)

It's actually an interesting paradox, the text versus
display conflict. Certainly, a text font can be made to
look less ugly than it can get away with at small sizes,
but I ponder the true limits of that. I have this hunch
that any real -I mean really real- text font has to have
a certain ugliness. If that sounds crazy, just look at
some of Tracy's work.

But maybe Patria could be less ugly.

> The "g": I would have thought that with a vertically stressed
> font you should avoid the diagonal swoop of the top of the
> lower bowl. Apart from the "s" this is the only letter that
> doesn't conform vertically.

Not surpisingly, I've thought long and deep about the "g".
My conclusions are somewhat unorthodox, but if I don't
follow them I'd be hypocritical. More to come...

> I think they stand out *too* much, and only because the weight is
> too light. Even it up with the roman and keep the angle as it is.

Note also that the x-height of the Italic is smaller.

The weight calibration issue doesn't bother me, because everything is
on an axis. I just need to do some real print runs from Lino output to
decide the exact weight locations. What is increasingly worrying me,
though, is the extremes: they might be too close, especially the Bold.

> maybe there is too much 'scotch' in [the numerals]?

Now that's interesting. Do you mean that they're too
curly or something? Or that they need to sober up? :-)

> The '2', '4' and '6' seem the most in need a fewer curves??

(What about the "9"?)
I agree on the "2", but the "4" is my darling.
As for the "6"/"9", I'm probably going to change them
completely, making them bowl-with-curved-diagonal-stick.

(BTW, concerning the numerals, I'm sure you can tell that they're coming
from that thread I once started about "Latinized Arabic" numerals.)

> [The 'f' and 'r' hang too much] in the roman as well.

Remember my penchant for promoting boumas (word shapes).

The Roman's "f" especially I'm pretty content with.
But its "r" is admittedly more difficult to justify.

> Not fan of the 'g' either.


> I'd buy.

Cool. Just send me 500 Euro in advance, wire-
transfer to my Cayman Islands account is fine.

> can you explain the design concepts of the fl and fi ligatures

Well, Bob, I'm very happy you asked that question! :-)

Different fonts need different tactics for the f-ligatures,
but for some reason most designers appear to be to shackled
to some strange standards that were established who know how
long ago and for what peculiar reasons (maybe for showmanship).
My ligatures look like that because to me making a font work
is more important than pleasing the establishment. If my x-
height were smaller, I'd probably merge the "f"'s beak with
the "i"'s dot in the "fi". On the other hand, the conventional
"fl" ligature in particular usually looks horrid; it's really
a mannered display-font trick, not something text-worthy.
I decided to chop off the el's top serif.

> there seems to me to be not enough difference in the weights.

This is actually the same deal as with
the Italic issue you were making above.

> looking at the screen is less than optimal

It's difficult to over-emphasize that. The lack of
congruence between the screen and the actual printed
end-result is really staggering. Even a 1200 dpi laser
doesn't cut it.

> it's of primary importance in designing a text family.


> You are using more flat curves (Lc 's' e.g.)
> and very round ones (Lc 'c' e.g.).

> I would to try to make the 'flat'
> curves a little bit rounder.

This opens up another interesting world! Remembering that my
intention is to give Patria a certain rigid masculinity, let
me ask: what's more rigid, a perfect circle, or an ellipse?...

I totally agree that making the "s" more curvy would make
things more harmonious, overall. But a very circular form is
perhaps "closer" to a straight form in a more nebulous way.
Also, a curvy "s" would make the font too feminine. Note the
other important glyphs which have some unexpected straight
segments: "a", "e", "g".

> Lc 'g': I would close his belly,
> or try to find a more subtle ending.

Like I said above, I'll try to elaborate more
in the future, but for now let me just say this:
1. I believe that the closed-bowl bicameral "g" is a flawed
structure in terms of the nature of the Latin alphabet.
2. The strong ending reinforces boumas.

> If this face is to become a face for newspaper,
> I wouldn't experiment too much with structure,
> as you did with Lc 'd'.

That's a good point. Although I feel my structure
is more readable, it might be going too strongly
against expectations. Gill got away with a much
more unconventional form... but that was Gill!

> the eye of the Lc 'e' is too big

It's too big for optimal readability, true. But that's
one place where I decided to sacrifice a little bit of
readability for character. Some other places:
1. The "a"'s middle bar is too rigid.
2. The "a", "g" and "s" are all too wide,
and too big-headed (which are actually
structurally related things).

BTW, check out this "e" (on the bottom-right):
It's butt-ugly, but man, does it ever work like a charm!

Anyway, I should consider making a more tame/readable version
of Patria, but in any case the differences will be very subtle,
at least on the surface. What to call it?...

BTW, very much of these critiques here seems to me to be related
to the display versus text issue. When I hesitate about some of
the good points you all make, it's because I'm cautious in my
hazy understanding of the feral wolf that is text face design,
versus the domesticated hound that is display design.


hrant's picture

> Maybe opening up the curve at the base of the "2" might help unify the three figures?

I think so.

Regarding the weight gradation:
I don't expect people to mix weights that are one position apart (even though it's not out of the question). I envision two pairings: Regular & Bold, and Light & Demi (the latter possibly useful in situations of significant ink gain). The question is: Is there enough contrast between each of those two pairs? Even if I feel that contemporary ideals of weight contrast are exagerated, the answer might very well still be: nope. :-/

I think I need more room on the weight axis at the heavy end for sure, maybe even both ends.


hrant's picture

> I've always been a fan of a wide range of weights

Well, that's certainly the norm these days*, and it can often be very useful. What bugs me about it is that -in a text face- it doesn't really make 100% sense, and it can confuse the spirit of a design, especially if everything is interpolated between the extremes. But if/when I make a display version, it will definitely need to have much further extremes.

* Luc[as] de Groot is the master of this.

> Sexy type!

I like that. Can I quote you? :-)
Thanks for the encouragement.

I'll probably be selling it through MyFonts as soon as it's ready - hopefully in a couple of months.


hrant's picture

> How exciting it is to see that the teachings of Father Catich are still being studied.

Wow - I'm embarassed.
I certainly wish that I'd studied them already (I've only perused his "Origin of the Serif"/Trajan work), but of course now I will - and I'm very happy to have heard this before releasing Patria: chances are I'll be able to improve it as a result.

Well, so much for thinking I was being just a bit original... :-/ I guess it's like what Goudy said about the old folk, one more time. I've gotta check this stuff out - thank you! In the mean time, let me reveal something: Patria italics is derived *directly* (I mean Fontographer-directly, plus tweaks) from the Roman, mostly because I believe in Morison's sloped-roman ideas (in spirit if not details), his one good original (?...) idea, as far as I'm concerned. So that's where it's probably different than Catich's ideas (I hope...).

Also, you could say that Patria has some Gill in it (like look at Joanna's italic), and that might tie in to Catich too. Patria has an inscriptional/lapidary flavor, so maybe that's another relation. BTW, I didn't realize that Catich had "theorical" leanings - I thought he was essentially a classicist.


hrant's picture

The development of Patria was so strange that I get confused myself sometimes! The italic was *not* derived from the roman - they're more like sister and brother. The father being a third design code-named Harrier, which is essentially a pretty unorthodox semi-serif sloped roman. It's a stand-alone design, but it spawned both the roman and italic of Patria (in parallel). Now here's where it gets really strange. Harrier was inspired by Nour (Patria's Armenian counterpart) and subsequently became a Latin design subordinate to it*; Nour itself is [in part] based on the first Armenian type (made in Venice in the first quarter of the 16th century), which was the principal style of type until about a hundred years ago, and still gets a lot of use. And this first type itself was based on medieval manuscripts.

* Patria eventually got a subordinate Armenian design - to balance everything out. But there is no Armenian italic - it is not really needed, thanks to a floating emphasis mark (that can be placed on any vowel).


hrant's picture

You're right, my wording was messy. What I meant by "classicist" was somebody who says "this is the way it used to be done, so this is the way we should do it". With "theory" (and please don't waste anybody's time nitpicking about typos in such an informal place) I was indeed pretty hazy, but what I did mean was a process by which tradition is often refuted. This is also related to the eternal battle between theory and practice, between people who paint signs for hardware stores versus those who try to understand -and improve- the underlying structures.

BTW, not dismiss; dislike, and only on a personal level. Which doesn't mean I don't respect Catich's *work* (just like so many people hated Jan van Krimpen, but admired his designs, and even thoughts). In fact, on my next trip to UCLA I will be checking out a copy of his "Reed..." book, to see what I can learn from it.

Now, two questions (for now) to you:
1. What do you think of Adobe's Trajan?
2. Are you the same person as "Teri"?

If you refuse to answer these questions, then I think I (and most probably others) know who you really are.


hrant's picture

Partly to clarify the development of Patria, but also hoping to spur others into providing glimpses into their work method, I thought to present a partial elaboration of Patria's development. BTW, it's pretty jet-set, so you might even enjoy it!

Nour came first, and although it incorporates much of my thoughts on readability, it's based on the first style of Armenian type (which is still used, but is seen by many as being old-fashioned at best - I won't go there here/now). This first Armenian type was in turn was based on medieval Armenian manuscripts - a sample:

Here's one of the six sheets of drawings I made for Nour, while I was in Yerevan (during the summer of 2000):

Since I consider "authentic" Armenian writing to have a slant (noting again that Armenian doesn't really need italics), but weighing that against the bulk of current Armenian designs (which are upright), and furthermore minding readability, I decided to give the font a slight slant, specifically 1/16 = ~3.58 degrees. And this is actually where the slant of Harrier and Patria-italic comes from.

Also, since I had plenty of time and no handy access to a computer, I went ahead and drew the letters as close to their (hopefully) final form as I could. It feels like I did more erasing than drawing, but as you can see I still made some small markings concerning adjustments. But in the end, only one letter (out of 38 x 2 characters) underwent a structural change.

During ATypI's conference in Leipzig I realized that I simply *had* to make a Latin partner for Nour, and the evening after I went to a bar (called "Letterman"!), and over some Obsolut Mandarin I drew some rough stuff:
I with I could blame the vodka for the poor quality, but the truth is my freehand-drawing skills suck.

A couple of days later I drew these at a sushi bar in Frankfurt:
It was over some Sapporo and a spider roll, so I guess that explains "krab"...

The next day in Amsterdam, a few hours before boarding my plane back to LA, I was having some mint tea in a place called Goa, and I made a breakthrough:
I say breakthrough because I figured out exactly how to apply my beloved "rigidity", and drew the "a" and "e" (a couple of *very* important characters) to satisfaction. You can also see the original "g", which I saw as too rigid in the end. That "g" also had a "rationalist" stroke contrast distribution, but I decided to compromise it with what other type designers expected to see. :-/

Back in LA, I started doing the "scannable" drawings, but since I then had much more time-pressure as well as easy access to a computer, I didn't bother to "normalize" the weight, slant, width, or even the proportions - since those can be done with software. I drew the structures, and experimented a bit too (note the "k").

After a pretty long, sporadic process, I ended up with the beta version of Harrier (around the spring of 2001):

At this point, I decided to make a somewhat complex "balanced" typeface system which combines Armenian and Latin fonts in different ways to maximize versatility. Among other things, this included the creation of the weight extremes for each of the 5 master designs. Harrier changed a little and became the Latin style subordinate to Nour, and it also spawned Patria and its italic, in parallel. After that, Patria got a subordinate Armenian style - which looks a lot like most of those current Armenian designs I mention above!

Although different in proportions, "serif-ness", terminals, and even some characters (like the "a" and "g"), the two are structurally the same. As for the roman of Patria, it also comes from Harrier, but it required deeper changes, like in the arches of "hmnu", the armature of the "k", and other places.

Patria's atypical provenance makes it admittedly somewhat unorthodox, but I hope I've applied everything with enough subtlety to avoid rejection.

One last thing: I've decided to make some improvements to Patria, but one thing continues to escape resolution... Should I reconcile the abrupt serifs with the "adnate" (smooth) beaks (like in the lc "s"), or should I hope that they form a higher order of harmony (but maybe to the point of discord)? And if I reconcile them, should I make the serifs smooth (going towards early 20th century transitional designs, like Times), or should I make the beaks abrupt (like a lot of contemporary work - Eureka comes to mind)? Once I answer this line of questioning (hopefully with your help), I can start thinking about releasing this thing already!


hrant's picture

BTW, here's a rough scan of page 29 (28 is the explanatory text) of "Reed, Pen, & Brush Alphabets" by Catich, showing Petrarch:

Maybe it's kindest for me to simply say: no comment...

But I can't help asking this:
> As a calligraphy student of one of Father's disciples

Who was this disciple, exactly?


gulliver's picture

What a cool lc "g"! >;]

Seriously, I like the "g" a lot, even if (or, perhaps, "especially because") the stress isn't strictly vertical. I think some variations in stress add character to a face, and keep the face from having too much page sparkle and/or monotony. I also think that, with characters such as lc "a", "g", "s", UC "G", "Q" "S", et al., some variations in stress are appropriate to the glyphs' structures.


hrant's picture

> the stress isn't strictly vertical.

As you can clearly see by the Amsterdam sketches, my original "g" had vertical stress, because -just like the logic in my latinized arabic numerals idea- it makes sense. But even after I softened the join it still looked too strange. On the other hand, the conventional structure with the thick, monotone bar was just as bad (even when I canted it clockwise - those totally horizontal ones I find especially abominable). Please ignore those ugly ears...
Left: Original v-stress version.
Right: Mostly conventional version.
Middle: "Balanced" version.

And as you hint, the non-rational (but entirely acceptable) stress in the conventional lc "s" was my guiding light. That and the lc "g" in Zapf International:


hrant's picture

In relation to my quandary concerning Patria's serifs-versus-terminals issue, what do you guys think of this?


David Cabianca's picture


Your font is really beautiful. I have been following the discussion and have learned a lot. I have started a MS Word file and have been able to follow the discussion and study the work via a hardcopy. [Jared and Joseph, if you read this, The forums would make a great edited compilation some day :-) and I'd be interested in helping out.] I was curious to know more about your background... Where did you learn so much about type design? Did you work under a mentor? How much type design/reading did you do in school? Any anything else you can suggest for an amateur.


hrant's picture

Thanks for the compliment.
I know some things about type design, but I'm missing more than I know - and always will!

For many years I was a typographic hermit, but for the past four years I've been discussing type with others and more importantly reading all the "main" books, as well as more "exotic" ones that appeal to me. The value of a good library (especially one with Inter-Library-Loan capability) cannot be overstated. Part of the reason for this is that -even if you're a submissive person, which I'm not- practically nobody will teach you too much these days unless you pay them. And since there's little money in type design now, this doesn't tend to work out...

But even more important than reading and learning is *thinking* - on your own. I do that a lot, and it helps to guide my creativity too. One other thing: some people are better at learning through doing, others (like me) through thinking. You have to do both, of course, but it's important to take account of what kind of person you are - especially if you have a mentor (who might be incompatible with your "style"). And if you do have a mentor/teacher, always keep a core of doubt - see him as an equal, although you might have to keep that a secret from him.

BTW, the Typophile boys *have* made a "compilation" (but just the first 50 fonts, so far):


ricardo's picture

Hello Hrant I didn't sow your Patria work until today and I want to give you the congrats for the excelent work I realy like all versions, absolut great. I'll be attention with the progress of patria.


hrant's picture

If you can point out anything that needs to be fixed, *please* do let me know - I plan on releasing it by summer.


hrant's picture

So I've been ruminating about the Display cut of Patria. I know what structural mods I'll be making to the lc (like making the "d" mainstream), and some global changes too (more delicate serifs, higher contrast, narrower, etc.), but I'd really like to make all the caps line up: no descenders. The "J" is easy, but the "Q"...

I did some rough outlines:
Any preference or comments?


kentlew's picture

Hrant, my thought is that if you're going to take away the descenderness of the Q's tail, then perhaps you should consider doing something inside the counter, either penetrating into the counterspace or perhaps a loop like -- what's a type that does that . . . okay, like Miller for instance. I don't recall the overall characteristics of Patria offhand, so I can't speak to which approach might be more appropriate stylistically.

My point is that without a descending tail, you're limited to how long you can make it, which means that it's going to be difficult to differentiate from the 'O', unless you also do something to distinguish it inside the counter.

-- K.

Stephen Coles's picture

Why no cap descenders in the titling font?

hrant's picture

> perhaps you should consider doing something inside the counter

Yeah, I guess. Pisses me off.

Patria has a certain severity (especially in the caps), so it might be hard to avoid cuteness with an internal tail. But I'll try.

> what's a type that does that

Tankard's Aspect does it very well.

> without a descending tail ... it's going to be difficult to differentiate from the 'O'

To some extent - but remember: Display cut.

> Why no cap descenders in the titling font?

Well, for this Display cut I would actually like to have alternates for the "J" and "Q" with descenders (although not nearly as long as the ones in the Text), but the main forms I think should fit between the lines, because you don't want a nice, solid all-caps setting ruined by a protrusion. And if you set multiple lines of caps (not necessarily long ones), a descender ruins your linespacing.


Stephen Coles's picture

Really? I welcome an occassional descender in cap-only
headlines (see below). But I can't fault you for offering
an alternate.


kentlew's picture

Man, I sure do love Dante. Mmmm.

Hrant, linguistic frequency! A descending J or Q here or there won't ruin your linespacing. If you're spacing tight enough for the odd descender to interrupt, then you've probably got other problems as well.

But, back to your Q. Okay, if a loop would be too cute, then do something different. My point is to change the shape of the counter. Maybe just take your tail and jab it up and in to the counter. That might be severe enough ;-)

Your point is taken, though: perhaps this is not so important in a display cut. Still, a little more something seems to be required in your samples.

If I had to pick one, I'd go with the last example. Since it's almost always going to be followed by a U, can you extend the tail just a little more? (What other languages use Q and what letters follow?)

hrant's picture

Stephen, funny!


> My point is to change the shape of the counter.

I love that type of thing (since it makes the "Q" less of an "O"), and I've drawn some like that for Paphos (although descending*), but I've yet to arrive at a form that's sufficiently mainstream.

> can you extend the tail just a little more?

That would be nice - I'll have a good length when I test the "U"/"u" after it.

> Maybe just take your tail and jab it up and in to the counter.

Yeah, I'll try that.

> What other languages use Q and what letters follow?

Well, in the transcription of Arabic to Latin you can get almost anything after a "Q"/"q"**. But some kerning pairs should be good enough to address such rarities.


** From "Making the Alphabet Dance": tariqa, shuqba, qcepo (medical term), taluqdar, qere, cinqfoil, iraqgate, fiqh, qiviut, feqjakuqu (Albanian place name), aqkiyenik (Amerind), taqlid, luqman, qnaitra, qoph, maqpiato (from Ethology), zaqqum, miqra, buqsha, qtrah, quite, qvint, acqwyte, clerqx (Dutch!), qyrghyz, ahmaqzai.


hrant's picture

So I got a bunch of Display-cut "Q"s for you:

Choose one, or critique, please! :-)


kentlew's picture

I thought the descenderless Q was for all-caps titling. Anyway. . . purely on aesthetic grounds, just taken as they are here, I like numbers 2,3, & 5. (1-12, left to right, top to bottom) The corners are subtle, but they help differentiate from the O. I'd still like to see the tail extended slightly beyond the right edge, maybe 15-20 units. I think the bottom concept works, but without seeing the rest of the caps I can't judge how it fits in. Among the four, I prefer number 12. I'm not crazy about the extra spur inside, but 10 is preferable to 11.

hrant's picture

> I thought the descenderless Q was for all-caps titling.

Well, the descenderlessness (!) of the Display-cut "Q" is to ensure solid lines in case of all-caps setting, but the main usage is *still* mixed-case. I will have an alternate that descends (like the top-left one but a lot shorter), but this is for the main form.

> I think the bottom concept works

Note that the "serif" form (on the outside) is the only such form in the font, which might be a problem. But anyway the only good reason I can think of to have a part on the inside is if otherwise it looks too much like an "O". Since the inside part is a bit disruptive, and the top two rows contain plenty of designs that are sufficiently different than an "O" (especially for a Display cut; and especially the ones with a corner at the bottom-left, like you said), I think I won't go inside.

Thanks for the feedback! I've narrowed it down to like 4 choices, and I'll have those up soon.


hrant's picture

OK, it's down to two forms:

BTW, I've made the tails as long as decency would allow. Or am I being a prude?


Stephen Coles's picture

Those look really nice, Hrant.

hrant's picture

Thanks. But you can't have both...
Wait a second, you're from Utah! ;-)


Stephen Coles's picture

Yes, I want both. And the one with the descender.
I want 3. That's right. 3 'Q's. Give.

kentlew's picture

I like them both, Hrant. The length of tail is just what I had in mind. I suppose I would probably go with the top one. But I don't have the rest of the caps in front of me to consider the context. How much rigidity is built in to the rest of the caps?

They're both fine. Pick one and move on. ;-)

-- Kent.

Miss Tiffany's picture

At smaller sizes (when I shrink the display) I prefer the top one with some internal curve, but then again this is for display purposes so maybe the more dramatic of the two (bottom) with the hard edges. For display I like the bottom. -- I guess the water in Utah does make us greedy. ;) -- And, Hrant, prude isn't what comes to mind when looking at this Q -- I wouldn't make it longer.

hrant's picture

> How much rigidity is built in to the rest of the caps?

What, you want a number? :-)

I think the bottom "Q" is better: it's less of an "O", more in character, and more balanced horizontally.

BTW, I tested the word "IRAQI", since it might actually be the most frequent Q-not-followed-by-u/U word, especially these days... It came out OK, and seemed to strongly favor the second form too.


Christian Robertson's picture

Seriously digging the Italic. I haven't actually done an italic yet, but I'm inspired by the seeming inconsitent slant. Is that an optical thing, or are the vertical strokes of a slightly varied slant? Also, the italic 'k' is just amazing. The shallow slant of the leg is spectacular.

I'm having a hard time with the tail on the g. It kind of meanders without strength or direction. The concept is not that far off; as a matter of fact, I really like the g from harrier. I would recommend consulting harrier, but also drawing a bunch of quick, strong lineal versions of the character and consulting the curves that come from a bold stroke of the hand. It also feels like it's tipping forward a bit.

hrant's picture

> I'm inspired by the seeming inconsitent slant.

Uh, that's a technical defect - the slants are all equal. :-/

Certain forms (like the "i"), which are sometimes called pothooks, appear to rotate counter-clockwise. I thought that because my slant is very slight (1/16 = ~3.58 degrees), and because my serifs are small, I could avoid having to make optical corrections - but I was wrong.

The structure of the "k" comes from my Alphabet Reform work.

Anyway, I'm glad you like the italic - thanks!

> I'm having a hard time with the tail on the g. It kind of meanders without strength or direction.

Well, I hope that the tail has "inner strength": note the rigid section in the middle. But quite possibly the join is too soft - maybe something more angular (like in the cedilla?) would be better. One thing I won't do is make the top of the tail horizontal: that has no place in the Latin alphabet - sorry everybody. Also, I don't mind picturing what the/a hand might do in drawing a letter, but I don't actually draw forms by hand - just the outlines - otherwise it's not really type. The forms are in my head - the less things interfere in their eventual physical manifestation, the better.

> I really like the g from harrier.

Harrier's is the original, and Patria's was derived from it (mostly) my shortening things vertically. Maybe something got lost in the translation (or something needs to be added).

Thanks for the comments!


greatermark's picture

Hi my name is Eka.
I have Djohar for you to critique.
It's my first experimental type.

Please comment.

By the way it still vector format, i haven't convert it to type format so i can edit it again easily.
Sorry for my bad english.

djohar.swf (30 k)

greatermark's picture

Ups why it appear in the Patria?

hrant's picture

I *wish* I had designed that! :-)


rs_donsata's picture

Hrant I nottice some irregularity or inconsistancy in the width and shape of the serifs, specially in a,A,c,C,S.... looks like you used a wide range of possible serif terminations. The shape of the c seems asistemathic to me and the "stems" (couldn

Syndicate content Syndicate content