Never received any training, I mostly use LaTeX so I'm a little aware of spacing and style.
Had 82/100 as average. What's your score?
I tried a few of the fonts. I never really liked monospace but I really dislike making decisions. If I had to make a font, I'd probably use monospace. My understanding is the whole problem goes away with monospace, right?
Edit:
I am having a difficult time articulating my thoughts on this.
Let me just spell it out: I think I can recognize truly horrible kerning. It makes it very difficult to read. However, the flip side is hard. I don't think about kerning when things work (most of the time). And I can't tell kerning issues from for example when I needed to use some Microsoft Windows Server machine who remembers back when and it didn't have ClearType. All I can tell is it is difficult for me to read. I can't really explain why.
The flip side, recognizing good kerning I think is very difficult for me. I simply never think about it.
For computer science people, maybe think of it as a satisfiability problem? It is fairly easy for a human to do the boolean satisfiability verification (where true is illegible and false is NOT illegible). However, as soon as you turn boolean satisfiability verification to something like k-sat, now the classification becomes (almost?) impossible because it is now subjective? Back to my simple terms, how can you say one kerning is better than another? Does it depend on the opinion of the person reading? Come to think of it, is kerning different in languages other than English? Everything I have said so far is all about English...
The city I'm living in was once famous for its printing history and there's a whole part of town full of old printworks that have slowly been converted to high price housing. One of the converted buildings has a stairwell that's fully visible through windows and it has in large vinyl letters different names of fonts in that specific font all over the walls. Like "papyrus" in Papyrus, "helvetica" in Helvetica, etc.
And the kerning of those words is just the worst. It's so bad, you'd think that there are random spaces in those words.
The kerning hurts my eyes. The irony pains me a few centimeters deeper.
Even in professional software like InDesign, you can choose between ad-hoc calculated optical kerning and the resp. font's kerning.
These can differ considerably, depending on the intend and preferences of the typographer that designed the font and kerned it.
See my other comment here about kerning depending on font size.
A font may be kerned to be typeset at 10–16pt but you use it for headlines at 24p. The optical kerning may be better suited then.
Kerning is hard, not just because of the skill you need to eyeball things well but because there is so much variance to account for. Very tedious.
That said, I think there's an art to it, so it's probably fine that I disagreed with some of the choices.
With Quijote, I wanted the u closer to the Q than the solution did. When I got to Toronto I tried to remember that, and ended up with a low score because the T and o were a lot closer together that time.
I dunno. Hard to nail down. But fun! Nice project.
Not disappointed.
Kerning depends on size. The smaller the font is displayed, the more it likely it is used for text were uniform looking kerning depends more on the upper third of a character.
However, this page displays the fonts large, very large, depending on what display you look at it.
So this has to be taken into account for the 'solution'. And furthermore, since the display is large, like a headline or logo, the kerning should consider the whole character, not just the upper part.
Lastly that font size may we'll be against the intend of the typographer who cut the resp. font.
I.e. if the font's internal kerning is used, that would then become an unreliable measure for 'correctness'.
If you were to tweak the game to build typography skills it'd be about pairs and you would have to look at the typeface at different sizes. In addition there are poorly kerned typefaces out there which show what happens when no or little kerning exists. You'll find all the common places kerning is required.
Also would be fun to consider ligatures (joined letter pairs e.g. fl) that are possible too. A more stylistic way to deal with spacing between letters.
In short, cute game. Not quite a real-world practice tool!
Edit: the best version of this game would be to make a (or edit an open license) typeface :D and kern it
KernType – A Kerning Game - https://news.ycombinator.com/item?id=21913654 - Dec 2019 (36 comments)
Kern Type: a kerning game - https://news.ycombinator.com/item?id=11032912 - Feb 2016 (5 comments)
Kern Type, The Kerning Game - https://news.ycombinator.com/item?id=6791800 - Nov 2013 (23 comments)
Show HN: KernType, a typography game made with Raphaël - https://news.ycombinator.com/item?id=3086320 - Oct 2011 (53 comments)
(now I'm off to see which of the 12 past submitters of the Bézier Game deserves a repost invite...)
My opinion.
