Handwritten font: am I creating something useful here?

Koos Looijesteijn

Hello! As part of a side project, I've started making a handwritten font. I've worked on this in evenings and weekends for nearly two months now. Discovering how much work it is to do well, I'm wondering if it's worth going the extra mile and offer it for sale. There's a ton of handwritten fonts out there, so I'm wondering what yet another one would have to offer. What purposes do you think a font like this could fulfill?

I also have a technical question. Because the glyphs are all more or less straight on the baseline, the font looks too mechanical to me. I could arbitrarily move each glyph a bit up or done a bit, but I'm not sure if that actually increases the handwritten feel or just makes it look random and wonky. Any tips for making variation in vertical positioning look natural?

I'd definitely enjoy the learning process of working on this font a bit longer. But doing it well enough to release it also requires a bit of tedious spacing and including glyphs I wouldn't need for my own project and the effort of actually promoting it. And I'd need to pay for a Glyphs 3 license.

Anyway, I'd appreciate your expert opinions on how to proceed with the design too. I'm well aware that I'm far from done with spacing, that the lowercase thicknesses don't look coherent and that I need more contextual alternates.

As this is my first post, I'll include a little background about myself. I'm a user interface designer, meaning I have a lot of experience with applying typefaces and with creating vector drawings. Making my own font has been on my mind for about 15 years now. It's every bit as hard as imagined it to be. Learning the technical bits of it has only increased my respect for real typeface designers. But now I'm learning about what makes a font work well, I want to do all that.

George Thomas

In the days of metal type they faced the same question: will anyone like it enough to buy it? In those days they had more at risk, but didn't have ten thousand competing fonts so had some advantage.

With a little more work on it, I would put it out there for sale. After all, you've done the work and if you don't offer it... well, you get the point. At any rate you will have benefited from having made the font if you learned something in the process.

Re: your question about vertical position: typically round letters and letters with pointed tops/bottoms need to extend above and below the baseline/x-height/cap height so they won't look too small next to letters with flat tops/bottoms. In this case given the shapes I see, I would still follow that rule but not as much as I would normally, perhaps four-six units in a 1000 unit em at most -- whatever looks "right" to you.

Thomas Phinney

You said you were aware the spacing needs work, but just to give some feedback as to what direction it needs… I can’t tell if it is just the base spacing, or the kerning on top of that, but overall it looks awfully tight—even if the font is intended for large sizes.

K Pease

Clear and sensible hand fonts each have their own personality, which means another one will always be useful to somebody. Yours is well-proportioned for comics, which makes it all the more likely people will see a handful of reasons to choose it over others very much like it, just because of the nuance of voice or art style they want to match. You know this because you had to create it for your project.

Koos Looijesteijn

@George Thomas & @K Pease thanks, I think I just needed a little encouragement (and more time to finish the work) and will put it on the market some time. I like how the design reminds Pease of comics, because it's very likely that comic lettering influenced me, although I wasn't really aware of it yet.

Regarding the vertical positioning: what I meant is that to me the glyphs all sitting on the baseline makes it look static. Real handwriting often has a wavy baseline. I though of solving that by varying the vertical positions of each glyph. But perhaps that's just not something I should try to solve in the font and leave it to the designer using it instead?

@Thomas Phinney thanks for pointing that out, I see what you mean with 'awfully tight'. I started off with spacing that was more like that between l and a, but in the process everything got squeezed together.

Igor Petrovic

It's looking pretty good so far, please don't give up

Thomas is right, the spacing looks too tight. If you are not already familiar with it, you might want to search for Walter Tracy's spacing method. 

As per your question about vertical positioning. I would say it's not a must when it comes to handwritten fonts. Maybe better without randomization than with a lousy one. Because if you simply make some letters positioned higher, they will be always higher and that is not what natural handwriting looks like. The solution is to make more complex randomization, by making a few alternates for some letters, and then try to randomize their usage by OpenType features. That way you will have i.e. lowercase o sometimes on the baseline, sometimes a bit higher, sometimes a bit lower etc. 

If you go that way, this video might be useful:
https://www.youtube.com/watch?v=eCviu05n_XA&ab_channel=TYPOInternationalDesignTalks

Also, I believe that the fact that there are tons of similar fonts is not a reason not to publish it. Hyperproduction is a broad topic, with different opinions, but type design passed that point long ago.

The market works the other way. Your possible customer is maybe not willing to search through the web ocean for too long (especially for handwritten and similar display "one-weight" fonts). If she/he stumbles upon it, likes it—that's it, the sale is made, just with many other products in the world. No one asks in the supermarket: Do we really needed one more chocolate biscuit in the world?! It's on the shelf, million others are not on that shelf...This topic can be debated, but anyway, not publishing your fonts will not solve the hyperproduction

Bringing something new these days is more on technology than on type design level. There is still a lack of well-developed non-Latin fonts, fonts for advanced "on-screen" usage like VR and similar, and for other special usage cases. If one aspires to bring something really new, it would require advanced-to-very advanced skills. 

But to be honest I've seen some great experimental type designs as well, on Future Fonts and similar. For example "Birra Saison" font by Viktoriya Grabowska. So there is a space to push things forward in type design as well.

Don't be afraid of mistakes, and don't spend too long on your first font. After all, the font is software and you can improve it over time. The most important thing for the first font is to finish it

Marc Oxborrow

@Koos Looijesteijn The pseudo-randomization effect you're after can be achieved with OpenType  contextual alternates. Here's a Glyphs tutorial on the topic, taken from this Typedrawers thread.

Craig Eliason

One option would be to let denser letters like /s or /e be taller (and thus lower at the baseline) so that the unevenness is not just random but serves to even out counterspaces. I discussed this tactic in my TypeWknd talk last year but the video hasn't yet been released. 

Koos Looijesteijn

If you go that way, this video might be useful:

Great! Yes, it even explains how to do the baseline randomization even without adding glyphs. Very much agree with her being annoyed at designs not leveraging glyph variants. By the way, it's the third time I run into Ulrike Rausch's content explaining what I want to achieve! I guess I'll just do everything she recommends and see what happens