By the way, the character variants underlined in red are the ones I like best so far.

Nice.

It's the type of thing we need - please do make it real.

hhp

Interesting. Stop sketching, start digitising.

Thanks for the encouragement. :) Here's a first experiment... don't like the curves on the |d| and |b| yet, but I can certainly see it going somewhere.

Looking at the "c" I see the big, central decision you'll have to make ASAP: figure out whether any glyph should be conventional. If you're going to have a mix, spend a good amount of time determining which letter should go to which side (preferably relying partly on letter frequencies). Or if you're ambitious implement OT code to give the user choices.

Maybe don't fully connect the bowl of the "a". Or even have the bowl totally separated.

Try to avoid symmetry like in the "b" and "d".

Try an "e" with a split like in the "f".

hhp

I think those narrow gaps are way too narrow in the digital version.

The cross-bar(gap) of /f/ should be longer at the right.

@ hrant: Yeah, I was planning on leaving a few glyphs conventional. Judging from my sketches, I expect |c i j l o s v z| to remain standard Garalde style simply because I can't seem to think of a satisfying way of introducing that fine-line substructure. They're just so inherently simple characters... do tell me if you have any lightbulbs on those.

On the |a|: I tried that on paper and wasn't satisfied with it, but maybe I just didn't try enough. I do like the current |a|, though; it's a bit of a defining character for this font idea.

On |b| vs |d|: Well, they're not entirely symmetrical. I also just noticed my sketched |d| had the gap in its bowl on the top while the electronic one has it on the bottom; I guess I could change that, though it would just be a different kind of symmetry. Just how much asymmetry are you suggesting? Making the |d| narrower than the |b|, for instance (apparently Garamond does that)? Or making up an entirely different alternate anatomy for |d|? I kinda like the bowl shape I have for |b| and was going to use it for |b p d| (and maybe go a little crazy on the |q|). If I have to make up a completely fresh design for |d| and |p|, I suspect they're going to turn out less pretty.

On the gapped |e|: I already tried that (see image), and I'm not quite sure how I feel about it. Incidentally, I'm not too fond of the gapped |f t| to begin with. I might still change those.

A bit more tweaking on the |b d|, and some new letters. Widened the gaps in the fiddly parts.

I'm not quite sure about that |e| yet; I might end up with a regular epsilon-style cut.

As for the name, Aurichalcum is perhaps too technical. I might switch to Windrose or Navigator, though the latter will be hard to google.

And here's the first complete draft of lowercase. I tried to salvage the |q| from a failed previous font, but it doesn't quite fit. I think I'll have to make a replacement that makes use of that thin “clockwork”. I might keep its shape quirky, though.

Opinions?

Love the texture! Which is the hardest thing in such an effort.
More soon.

BTW, I actually like the name quite a bit.

hhp

This is the kind of type design I like! Setting up a system and then getting stuck in to the problem solving. I think you have a good start. I see the /n/ as the key glyph to get the others working. For example, using the n structure as a starting point for /b/ it seems to me that at the moment the bowl is upside-down (same goes for d and p). The left stem of /n/ can also be used to make a pretty cool /i/ by extending the thin bit up into the tittle, and an /f/ by using it to form the crossbar. I'm sure none of this makes much sense so I did a quick sketch to show you what I mean:

A recent Type ID reminded me of Isadora (http://www.fontshop.com/search/?q=isadora), which could be a reference (even if by contrast) for the possible construction of some letters.

Interesting! Thanks for the feedback.

@ Dave: I can't say I like the |i| and |f| you propose. :\ My goal is not to shoehorn a double stem into every letter; that would somewhat trivialize the overall concept of the font.

Your proposals for |d b| are interesting; I'll definitely look into replacing some of those angled serifs with flat-footed ones. However, I liked the treatment of the bowls in those letters better in my version. Your proportions feel sleek and remind me of something contemporary like Quadraat, while I'm rather aiming at the generous, almost pompous, circle-based proportions I associate with compasses, naval steering wheels, and wood-and-brass engineering in general. The unconventional architecture of my bowls also lend themselves better to the humanist diagonal stress, while yours rather feel modern to me.

I agree that the broken |s z| feel contrived. I might end up with regular Garalde designs for those letters. Likewise, I might return to the simple epsilon-style |e| that I had earlier. I do like my current |f t| designs, though, especially the fact that they have their crossbars on the same height but approach that height from opposite sides.

@ Riccardo: I can see a certain similarity, but the overall feel of the font is extremely different from mine. Isadora is downright... frilly. I do intend to give double stems to my capitals, though, to mimic fraktur-style capitals, so at least that's the same in Isadora.

Fun with capitals!

What about a construction more like this, at least for |F|?

I have no idea how to design anything. I mean design as a process here. But as a visually aware human being – which is the only qualification I can claim supporting the legitimacy of my view – I absolutely adore this typeface! My first thoughts were “Oh my god, is this good”.

Not to add to the noise, but this is quite lovely! I’m going to be watching how this progresses…

@ Riccardo: I've tried out your architecture for |F|, but I'm not so happy with how it turns out... the thin piping adds weight to the lower half of the letter, making it looks somewhat clump-footed. I suppose I'll have to take care to keep the piping mostly on the inside of the glyphs.

I've closed off the tips of the crossbar, though, which I think helps.

@ Flooce: I'm not a professional designer by any stretch either — I do astrophysics for a living — so by all means do voice your opinion. :) My “process” is mainly trying to make pretty things, failing at first, and trying to learn from the feedback so I can make somewhat prettier things next time. ;o)

With a grain of salt: To my eye, closing the /F/ that way isn't helpful at all. Enclosing those spaces produces a hollow shape. Either return to the prior idea or make the middle crossbar one-stemmed and try something in the direction Riccardo suggested. Another option is completely cutting the shape, like you did with /e/ and /f/.

@ Lex: Hmmm. I see what you mean about the closing of |F|. Maybe I could cut the stem but leave the tips connected? Otherwise I'll just return to my first draft.

Meanwhile, I've tried adapting the |c| and |s| more into the emerging style of the font. I think it works very well for the |c|, but I'm not convinced of the |s| yet. I kept the split |z| because there is historical precedent for horizontal bars in that letter.

@ Riccardo: I've applied your proposed design to |L| instead — I think it works perfectly there. :)

@ Dave: So you didn't like the original |w|? That architecture seems rather commonplace in Fraktur, and I'm also used to it from the handwriting in school, but I'm not sure how recognizable it is world-wide. In any case, the premise of this font is to use the overall shapes of Garalde, so I suppose I should remake this letter anyway. I might keep the Fraktur-style |w| as an OpenType alternate. How about this?

And here's the first draft of the complete alphabet. Numbers next...

Wow.

I know it's hard, but release this quick before somebody copies it.

hhp

Thanks for the heads-up. :)

Just for the record, what exactly constitutes a “release”? I was thinking of keeping the font free for personal use, and just ask people who use it for commercial purposes for a donation. (Much as I like the idea of making money with type, I'm grateful for the many free fonts I've been using so far, and would like to repay the favor to some degree.) At which point does a release “count” for legal purposes? Is an alphabet like the one above enough for the first version, with more letters coming soon?

Repay whom? The people who make free fonts are not the ones who will benefit. And you will virtually never get a donation. Magnanimity is wonderful, but not if it stunts a promising future in type design; it's not like you'll get rich from it, but all humans are material creatures to some extent.

In retail, "release" means letting total strangers use it. BTW if you do it on MyFonts (which is easy) and somebody copies you and ends up on MyFonts too (which is statistically the most likely) you can ask MyFonts to take it down and they usually oblige. In the past year or so there have been 2-3 such cases thanks to Typophile. A drop in the bucket maybe, but: we've got your back. :-)

Character coverage: you can release a font with only alphanumerics, but try to complete at least one basic Unicode table.

hhp

Hmmm, you have a point there. If I charge money for the font, I can put that into buying other people's fonts, which has the benefit of actually supporting people who make fonts.

On the other hand, if the font were free, I would have no qualms putting up a limited “alpha version” now in order to stake my claim, and keep adding more glyphs, ligatures, alternates, and careful kerning later. I'd feel bad asking money for an alpha version. (I'll probably go for iKern for the kerning, at any rate.) Also, I assume the odds of having my design stolen and copied are much lower when my font is freely available...?

Magnanimity is wonderful, but not if it stunts a promising future in type design; it's not like you'll get rich from it, but all humans are material creatures to some extent.

Wait, so does making free fonts stunt one's chances in type design? I do have a day job, so I'm not counting on making money with my fonts (though I admit it would be gratifying).

As for coverage: I intend to include all of Latin-1 eventually, but that's going to take some precious time.

Sure, give out a beta version (which means it has known flaws that you intend to fix; alpha means it's known to fail spectacularly :-) for free, and even ask for feedback on it. When you finish it, sell it (maybe giving a discount to people who gave you valuable or at least non-trivial feedback).

Being copied is an honor, especially if you get the credit (which is not assured). But you have to balance that against income. If you don't expect income then being copied can only been good (even if you don't get explicit credit). However: you'd be harming type designers who do want to make money (which can even by a symbolic valuation of something you value - it doesn't have to make one's life much easier). Ergo: it's a social responsibility.

Also, can you predict the future? I can't. You don't know where a casual interest in something might lead, especially if life changes from under your feet.

At the very least you should charge a nominal amount just to erect a barrier against parasitic freeloading.

hhp

So you're saying giving away fonts for free actually harms type designers? Huh. I never would have figured. When I see a font sell for dozens of dollars per cut, I assume it's such a high-quality product that it cannot be replaced by a free font. For me, free fonts were “gateway drugs” to developing an interest in typography in the first place. I would have expected that to be good for business.

There's also the fact that I'm designing fonts primarily as a form of artistic expression, and making a font free vastly expands the target audience. Knowing my design has been downloaded thousands of times might be cooler than knowing it's been bought a handful of times.

What's a nominal price for a single-cut font? $10?

Even $1 is like... infinitely more than zero. :-)

hhp

True, yet even a $1 purchase is a hurdle few people will have the energy to overcome unless they're used to buying fonts on a regular basis, in which case you're probably a graphic designer or aficionado, and would just as likely pay $10.

1$ might be infinitely much more money than 0$, but 0$ is an infinitely more attractive offer than 1$...

Getting used to paying for fonts isn't so bad.

hhp

Aurichalcum is too technical. I'm now thinking of Outbound as the release name of this font. It sounds appropriately adventurous and nautical. It's also a good word to showcase the fraktur-style |O|.

Is it possible to keep updating a given font on MyFonts so that people who bought the first version can download the newer releases whenever they become available?

I think the name should reflect its android/mecha look and feel.

BTW, I wouldn't call that "o" fraktur-style.

hhp

I'm not getting an android/mecha vibe at all. I think it evokes retro-tech from the era of steam boats, the Eiffel tower, and sea travel. The fearless spirit of discovery and adventure of newly invented science.

Yeah I'm misusing the word Fraktur. I suppose I'm referring to the internal vertical structure that |O| has in Old English Blackletter.

So, what about version control on MyFonts? No problem, I assume?

I like where this is going. How about the name Oarlock?

I like how clockpunk this is while still remaining practical and readable.

The original |w| was too obscure, but the new one is too prosaic (and kind of "v-v"). I'd try more ideas and aim for a middle ground.
I like the construction of the |W| but I would enjoy seeing the central spanning serif beam made into an arc sweeping around what would be the crossing.
There seems room in |M| and |Z| for more fanciful touches too, if you can think of some.
|X| appears to be light. I suspect it's not merely from comparison with the doubled strokes in other letters, but that the downstroke is actually lighter than it should be.

Hey all,

sorry for the radio silence. I've completed a larger character set now, increased the descender depth for more appropriate old-style proportions, recut a number of characters that didn't fit the texture of the font well (such as many of the diagonal characters), and had it all kerned by Igino Marini's iKern. I'm now trying to get it published on MyFonts as soon as possible, but it's taking a while.

@ Cerulean: You're right about the |X|, it was too light. Good catch!

I'm running into a weird problem, though. iKern produces OTF files as an output, which work perfectly. However, if I try to save them into FontForge format for further editing (such as making the X heavier) and then generate new OTF fonts for system use, I lose all functionality of my ligatures. WTF? What's going wrong? Do I have to work in OTF exclusively?

EDIT: OK, it works if I make the changes to Igino's OTF files and directly export it into OTF again.

@ Luma: Oarlock is a nice word, but its meaning seems rather pedestrian. ;o)

I was thinking of some nautical, brass mechanical stuff that vaguely resembles the letters and starts with O. I see your forms here:

Call it Astrolabe.

hhp

@ Luma: Those are some nice oarlocks. ;o)

@ Hrant: That was one of the first names I considered, as it happens. The meaning is really cool, but I don't like the name aesthetically. Also, there's a slew of square sans science fiction fonts called Astro-something; I wouldn't want to end up in that drawer.

Taking some names for a ride... frankly, all in all, I still like Outbound best. It has a positive, evocative, universal meaning and just looks nice. I also like Quadrant really well, but I'm afraid it might be too close to the existing font names Quadraat, Quadranta, and Alpha Quadrant.

Some neat words in the conceptual territory of Astrolabe: Orrery, Torquetum, Antikythera.

@ Cerulean: Those sound intriguing, but unfortunately not very aesthetic (Orrery in particular sounds like a speech impediment!)...

However, I believe I have the solution now:

This name has it all: It's reasonably pretty, starts with an |O|, contains a ligature as well as the seminal |a|, and signifies a historic object both nautical and technical in nature, often crafted from brass. What's not to like? :o)

This is a wonderful concept. I wonder when there will be a usable font for download. Please let me know :-)
kyrmse@gmail.com

Hi Kyrmse,

good to know I'll have at least one buyer. ;o)

The font will be available for download as soon as MyFonts get back to me on my application for foundry status. It's been more than two weeks since I sent them my application materials, but I hear it can take up to a month... (Is that normal? I know at least one message from them already got lost in electronic limbo on the way to me...)

Wunderbar! Dieser font hat etwas renaissanceähnliches an sich, kann aber durchaus auch in "modernerer" umgebung gebraucht werden. Ich könnte mir vorstellen, daß beispielsweise Chaucers oder Malorys texte sehr schön in... wie heißt das noch mal?... "Octant" gesetzt werden könnten. Eine ganze seite mit verzierten initialien würde eine attraktive textur aufweisen.
[With due apologies to everybody else - got carried away:]
Wonderful! This font has something renaissance-like, but may be easily used in a more "modern" environment. I am able to imagine texts by e.g. Chaucer or Malory very beautifully set in... what was the name again?... "Octant". A whole page with decorated initials would exhibit an attractive texture.
Christian: do let me know when! ;-)

Just some quick observations:
/t/s crossbar could grow a little more from its left.
The tail of /Q/ is too heavy, I think.
/G/ looks narrow to my eyes.
Some of /W/ serifs are too short.

You’ve got some nice ideas there, reminds me of some of my scribblings :-)

@ Lex: I had already fixed the |Q| on my own, but thanks for pointing out the |W|. I strengthened the serifs and added the missing one on the leftmost arm back on. I suppose the |G| could use a little more width, but wouldn't want to make any more edits that mess up the iKerning at this point. I like the |t| as it is; it's among my favorite glyphs in this font.

MyFonts still haven't replied to my foundry application... :(

This looks absolutely great, I love the fine detail and shape contrasts. The capital Q must be my favorite (now it's tweaked I'm safe to say that!) I'm not too sure on the lower case 'a' though, the bowl could use a bit more space I think, it's extremely tight compared to all other glyphs, especially since you have such an open 'e'. But of course it's added quirkiness.

Also, the s is slightly dull compared to most glyphs, perhaps an idea for some alternates? I think it's good to have a neutral one, but the curves are begging for some of your crazy interventions ;)

Great job!

Alright, the contract with MyFonts is in place, so Octant should start selling any day now.

I still have an acute problem with it, though: Wherever it is used on the MyFonts webpage — both in the font sample windows and on my foundry page — neither the kerning nor the ligatures work.

From the website:

On my local machine, in PowerPoint:

Any idea why that would happen?

(I've started a separate thread for this technical problem: http://typophile.com/node/99929)

The technical problems have been resolved, as far as I can tell. Thus, Octant should be available for purchase within a day or two. Yay!

Congrats. It's really beautiful.

hhp

Very nice indeed!

Congrats! Good luck!

Dear Typophiles,

Octant is now for sale on MyFonts. Yay! :)

http://www.myfonts.com/fonts/catharsis-fonts/octant/

Thanks for your support, everyone!