After months of my day job getting in the way, playing with this again.
Trying the descending italic /Y/ again, but it's not working.
Trying a descending italic /k/ leg, which I like--question is, does it fit this face or not?

Thinking of abandoning the asymmetrical serifs--here's what the caps look like after the serif revision.

Well spotted. Better?

The leg of the R suffers from "pen rotation" at the joint: it gets to thin (e.g. start the inner curve a bit closer to the stem). Also, the state of the connection joining that inner curve of the leg and the middle horizontal is not yet settled: is it a smooth or an abrupt transition?

I'm not sure about the leg's end, regarding both R and K. I would make a more didone approximation in matters of shape (without exaggerating) and flatten those bottom curves a bit. But don't take this comment to seriously.

Maybe the descending Y would have worked if the right vertical stroke had been aligned with the slant angle...

Nice work c:

I see what you mean about whitespace balance (also about Cyrillic). I think it works better on K (if a bit receding) that on R (it looks a bit too much like a P with an appendage).
Aesthetically, I still prefer the previous, more classic blade-like K’s leg.

Man, this typeface features a tremendous amount of refreshing elements. I read about half this page and I had some points of criticism along the process but it seems to be taken care of already. There isn't really anything I would change or improve. I love the fact that this typeface seems to be the warmest Century typeface I've seen yet it remains very clean.

Thanks Martin for the compliments and the input!

Subtle is good :-)

Good, but to me it looks a bit unbalanced: R’s bowl rests its weight on the leg, while K’s arm lifts it. (this applies in general, not specifically to your samples)

Not necessarily sturdiness, but, so to speak, more “presence“ (extending past the bowl with part of the leg and not just the foot, for example).
But, as I said before, trust your eyes (and those of others) more than mine.

Here (4) is a sturdier leg, straighter and further out. I put a little corner in the closed counter that corresponds with the leg. Also lowered the middle very slightly.

Almost there, but the curve is still much less apparant than it is in the K. The angle at the top of the leg could be a bit, just a bit, more towards horizontal. The base of /2/ could extend a little further to the right for more stability as well.

Cheers!

Almost there, but the curve is still much less apparant than it is in the K. The angle at the top of the leg could be a bit, just a bit, more towards horizontal

Hmm, I was trying to make the curve less apparent, to gain the sturdiness it was missing! I think I'm reaching the conclusion that a diagonal leg has to be pretty straight (as #4), or else you get the "P with an appendage" effect that riccard0 was mentioning.

The base of /2/ could extend a little further to the right for more stability as well.

Not sure about that. But I wonder if the top where the teardrop is should come down a little further. I really hate drawing /2/s, maybe more than anything.

Does this look like about the right height/weight/proportions for small caps?

More towards petite ones...

I think -for a text font- it's better to be on the small side
than the big side (and the darkening I mention would help a
bit). The size they are now would indeed not work for mixing
with the full caps, but to me that's a secondary usage, better
reserved for a display cut of a design.

hhp

Because it's better to be too subtle sometimes than in-your-face
all the time. But I guess we need to see a PDF. I'm thinking things
like pluralized acronyms will work as it stands, but needless to say
I could be wrong.

hhp

Much better, definitely. The serifs of /E/'s crossbar are perhaps too dark. Make them a little more abrupt. Just now I noticed that all of your /S/' may need more weight at the top left, and loose weight more slowly in it's way to the top; that would ameliorate the thick-thin transition, which should be more gradual IMO. The /2/ also needs work in the right curve :)

Wish the Typophile pdf upload weren't broken!

The curve at the right is a bit too thick and for some reason it's axis looks slanted to the left (counterclockwise). But these are just marginal observations. Ignorable, due to meticulousness, if you like.

PDF just worked: http://typophile.com/node/89140
Maybe it's the size issue?

hhp

OK looking at the PDF (MillennialOldstyle12.pdf is the current one, right?) the weight seems pretty good to me, but I have to agree they look potentially a little too tall now… actually maybe even more importantly I wonder if they seem a little too wide, slightly out of character with the overall narrow elegance of the face? It's hard to tell though, don't listen to me too much, remember the salt & all that. ;)

Should probably give a bit more generous letterspacing.

Indeed :-)

Experimenting with a narrower contextual alternate for /r/, which would appear before /f/t/v/w/y/. (In this sample, 1=all the same /r/, 2=using the r.calt in "Harvey"/"Norway"/"barfed"/"every"/"party.")

@hrant
what you're suggesting is easily accomplished by kerning; why bother to make an alternate to preserve excessive white space? It looks as one step forward, but it's one step back ;-)