bugFree UCS Outline Fonts - Bugs: bug #39917, Greek italic


bug #39917: Greek italic

Submitted by:  Rosella Capriotti <rosy58>
Submitted on:  Sun 01 Sep 2013 03:32:29 PM UTC  
Category:  character range Severity:  1 - Wish
Item Group:  glyph(s) ugly Status:  Proceeding
Privacy:  Public Assigned to:  Steve White <Stevan_White>
Open/Closed:  Open Release:  2012-05-03

Add a New Comment(Rich Markup)

You are not logged in

Please log in, so followups can be emailed to you.


( Jump to the original submission)

Tue 17 Nov 2015 03:29:22 AM UTC, comment #22:

Hi Steve,

I have a proposal for FreeSerif Italic. In my proposal when Theta is used in Greek language, this letter should have a contextual form. The end of tail should be shorter within word, in end of word is longer.

Illustration is here:

(file #35477)

Hsing-yen He <hammerandsickle>
Sun 22 Mar 2015 10:51:19 AM UTC, comment #21:

Hi Steve,

I have a forget. I should mention “there is an online sample to be reference” firstly.

Hsing-yen He <hammerandsickle>
Sun 22 Mar 2015 09:32:17 AM UTC, comment #20:


it wasn't clear to me what you were talking about, with this LinoType website. To make it pertain to this discussion, I did this:

1) remove "Nimbus" from the search field,
2) under "supported languages" check "Greek"
3) type a letter phi into the sample text field.

An impressive list of different forms of the letter phi results.
I don't know that it resolves the issue here, but it might have been nice to look at!


Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sun 22 Mar 2015 05:04:45 AM UTC, comment #19:

I found Linotype Finder can made several sample by inputing texts if you find some font families, such as:

Hsing-yen He <hammerandsickle>
Tue 24 Sep 2013 12:06:55 PM UTC, comment #18:

The name of this topic is wrong: it should be changed to: "Shape of Greek PSI".

Rosella Capriotti <rosy58>
Wed 11 Sep 2013 05:36:01 PM UTC, comment #17:

OK, wait, there's clearly a misunderstanding going on between us, like when we kept disputing about KAPPA, remember?

So, basically, what matters to me is the central stroke, the inversion you did on the alternate glyph, I liked it very much, but I'd like you to implement it even in the standard glyph.

About the arms of the alternate glyph, it's not that I don't like them, it's that the serifs don't "get along" well with other serifs, it seems like you extracted this glyph from another font and the pasted it on FreeSerif, I was suggesting you harmonize the shape to make it more compatible with the general "feel" of the font.

But, I repeat it, what matters really to me is central stroke, not the arms.

Rosella Capriotti <rosy58>
Wed 11 Sep 2013 04:50:42 PM UTC, comment #16:

??? Does not compute.

What is the point of an alternative glyph if it is the same as the default glyph?

As I said, from what my researches have turned up, the current default glyph is standard for this kind of face.

So I'm afraid I still don't understand.

I think you want me to alter the default glyph, but I have reasons not to make any big changes to it.

So I proposed an alternate glyph, but you don't like it either, but I don't understand what aspect of it you don't like. First you said you didn't care about the arms (which were the thing that I mostly noticed.) The alternate glyph was supposed to satisfy your preference for a longer central stroke, while satisfying my preference for outward-bending arms.

But then you didn't like the arms on the new glyph. And now... you complain it isn't the same as the old glyph. How could this be more confused?

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 10 Sep 2013 06:31:44 PM UTC, comment #15:

Out of touch is the new alternative glyph not the old standard one.

Rosella Capriotti <rosy58>
Tue 10 Sep 2013 06:19:55 PM UTC, comment #14:

What I get is:

  • you do not stress the judgment you made about the Unicode sample.
  • you stress again the length of the stroke
  • as a side remark, you find the arms of the alternate glyph out out of touch with the rest of the range.

I only wish I knew what you meant by the last point.

The one thing that seems peculiar to me about the LinoType/FreeFont psi is that it is assymmetrical. Few other upright faces are like this. Nonetheless it is the traditional style for this kind of face, as is the upward-tapering, medium height central stroke.

I could try knocking off the hooks at the ends of the arms... to my eye, although they are like the hooks on the other letters, having two of them might be excessive.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 10 Sep 2013 05:29:24 PM UTC, comment #13:

Dear Steve,

I think there's a misunderstanding between us: I've always been concerned about the central stroke, I'd like it to be longer in both versions of PSI, if possible. As far as the arms are concerned, the serifs you've applied seem to me out of touch with the rest of the glyphs.

I hope I've made myself clear.

Rosella Capriotti <rosy58>
Tue 10 Sep 2013 05:09:41 PM UTC, comment #12:

I am utterly bewildred by your remark.

Could we somehow be talking about totally different things?

I maintain that the shape of the arms of the alternate glyph for phi currently in FreeSerif is more like the shape in the Unicode specimens, than is the current glyph for phi in FreeSerif. Of course I arranged the arms to have proportion and hooks more like other Greek letters in FreeSerif.

This is illustrated by the image
The current phi there is in green, the alternate is black with red dots.

The Unicode specimine is at
at slot u03C8.

Comparing the alternate with the specimine: both have outward-bending arms. I put serifs at the ends of the FreeSerif one for consistency with those on other letters in FreeSerif. The specimine has a very short central stroke -- but as I understand it, you want a longer stroke.

After I posted the image of the two forms, you indicated that you scarcely noticed the shape of the arms, and it was only the length of the central stroke that concerned you. Have you reversed your impression?

You know, I've spent many hours now researching this one letter.
I'm not sure what else I could do, short of making a couple of other alternatives for the one letter -- but that would be excessive.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 10 Sep 2013 09:59:19 AM UTC, comment #11:

The limbs of the alternate glyph don't look at all like those of the glyph in the Unicode specimen. The style of the glyph is also markedly different than all other glyphs in the font. This jumps out at you immediately.

Rosella Capriotti <rosy58>
Mon 09 Sep 2013 09:58:07 PM UTC, comment #10:

Cute... I'm glad it pleaed you Rosella! But cute is not the goal.

As I said, the reversed stroke is unlike existing fonts of the same style, and rather unlike the rest of the letters in the font.

Currently, all the main-range glyphs are very similar to the LinoType font I mentioned, which was the first example of a serifed face I found in a printed book in town. I am very reluctant to deviate from that in a single letter.

Maybe I can find a middle road, reducing the taper at the top (it is awfully light).

The bent-out limbs are likewise very much the style in other kinds of faces, and has the advantage of being like the unicode samples, as well as more like how I expect psi to look. But again, I don't regard myself as a type designer.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sat 07 Sep 2013 12:10:26 AM UTC, comment #9:

I really liked the reversal but the arms don't convince me, it looks rather out of touch with the rest of the glyphs. Anyway, I think you should implement the reversal even in the standard glyph, it's sooooo cute!!!! I like it sooo much! LOL :)

Rosella Capriotti <rosy58>
Fri 06 Sep 2013 11:45:28 PM UTC, comment #8:

OK, I made a tentative alternative psi, in FreeSerif only.
Character Variant 08.

Here is the problem.

Although the form isn't what I think of as a "classical" psi, the form in FreeSerif has a pedigree, and I have seen that form used in professional printing.

On the other hand, the Unicode samples reflect something more like the "classical" psi.

Usually I prefer to err in the direction of the Unicode samples. But here, this is purely a style issue, and the pedigree is so strong, I prefer to leave the FreeSerif psi like its original, and make the more Unicode-like, "classical" form an alternative.

As to your concern about the central stroke, I reversed the tapering so it is heaviest at the top. I have seen this done in at least one other font, to good effect.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Fri 06 Sep 2013 02:10:03 PM UTC, comment #7:


I tried to illustrate the different forms I was seeing with the image. People really do perceive different things.

To me, the comparison with Didiot doesn't go anywhere -- this is an extemely different style -- that glyph would clash terribly with the existing glyphs in the font. The central stroke is indeed higher than the FreeSerif one, but also, it has lobes on bottom and top.

All three (or so) psi designs in Andron feature a central stroke that is both higher than the LinoType design, and also taper from top to bottom -- unlike the LinoType one.

Now, I'm trying to determine if there is any history in this LinoType style of font for the cenral stroke to be longer. So far, I'm not finding that.

I went to town to find modern printed material in Greek. Most of it is in sans-serif (where it seems, the "latinate" kappa is
used universally...) However, the Langesheit dictionary is typeset in a font that looks very much like the LinoType one. And its psi looks essentially identical to the one in FreeSerif, including the approximate height of the central stroke, and the tapering of the stroke from bottom to top.

Since then I did a survery of fonts at my disposal.
Just among the upright styles, there is a big variety of forms for this letter.

I find the one in FreeSerif to be sort of a middle road, between short fat psi's and narrow tall ones, between central strokes with fancy decoration and a simple line, between central strokes that are of the font height, and ones that are below the x-height.

Designs of psi

1 symmetrical, arms like palm branches
A central stroke of x-heigt, arms spread wide
B central stroke between x-height and font height;
tapering from bottom to top, arms narrower
2 asymmetrical, left arm w/serif, right with bulb
(often this is the cursive form)
A central stroke between x-height and font height;
tapering from bottom to top, arms narrower
(OmegaSerifGreek, LinoType Mac Greek,
B central stroke to font height;
bulbs at ends.


Unicode sample is symmetrical, but with a central stroke of x-height.

Cyberbit is like FreeSerif

New Century Schoolbook, Times New Roman has a psi whose central stroke is of the x-height.

LinoType Palatino is more like Didiot in that it has (gentle) bulbs on the stroke. Also all forms are assymetrial.

Very fancy: Agmena
Lots of ligatures, final forms, etc.

ClearType Font Collection,
Calibri, (gothic)
Cambria, (heavy serifs) psi palmtree, very subtle hooks on upsilon, chi eta
Consolas, (curvy gothic)
Constantia, (big wide serifs) serifs rather than hooks, asymmetrical psi
Candara, (thin modulated gothic) sarcely any hooks, asymmetrical psi
Corbel (very straight gothic) symmetrical psi
Meiryo (very straight gothic) symmetrical psi

Some other germane pages:



Let's keep thinking about this.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 04 Sep 2013 11:21:38 AM UTC, comment #6:

Frankly speaking, I cannot see the different types of PSI in Andron, the only thing that caught my attention, the thing that matters to me is the height of the central stroke, that's why I said "perpendicular", while you're concentrating on the design of the "cup". Anyway, my preferred glyph is the one that appears in Theano Didot, which is the standard in the hapla fonts which are standard in Greece. I think this issue should be approached in the same way as you did for KAPPA.

Rosella Capriotti <rosy58>
Wed 04 Sep 2013 08:33:19 AM UTC, comment #5:


"perpendicular" cannot be the right word.

Andron Greek 1 features every one of the different forms of psi. Which one are you referring to?

By "palm tree" I mean the one that is symmetrical, with the two branches that curve up then back down. Only one of the forms is like that. Others are assymetrical.

As I said, I experimented with a "palm tree" form of the letter. I don't think it looks too bad. But I am no artist, and typographers are very picky people.

See attached. In the background is the original, assymetrical psi (which looks like an upright italic to my eye). In the foreground is my own version of a "palm tree" form.

My concern with just changing the letter is, the form in FreeSerif seems to be conventional for a certain tradition (As evidenced by the LinoType Greek Mac Greek (and Omega fonts.) I don't think the form in FreeSerif was a mistake, as such. I can see that this form has a harmony with the other letters.

One option might be to provide yet another character variant feature.

(file #29003)

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 04 Sep 2013 12:15:36 AM UTC, comment #4:

OK for PHI, but I'd like the PSI to be just like in Andron Greek. I don't know what you mean by "palm-tree", what I wish is not redrawing the glyph but just increasing the height of the main stroke, the perpendicular one. I hope you get my point now.

Rosella Capriotti <rosy58>
Tue 03 Sep 2013 11:52:43 PM UTC, comment #3:

Carl Faulmann,
Das Buch der Schrift enthaltend die Schriftzeichen und Alphabete aller Zeiten und aller Völker des Erdkreises
Wien 1880

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 03 Sep 2013 11:10:24 PM UTC, comment #2:

I kind of see what you're saying, but I would say that the roman style is more like an upright italic, than the other way around.

Particularly I would prefer the normal psi to be a symmetrical palm-tree form. I've already experimented with it, but I'm not sure it's the right thing to do.

I really don't see what you're talking about with the phi. There is some variation in stroke modulation and subtle style details, but essentially the phi in serif is very like that in other fonts, within the constraint that it should fit with the other letters.

Here are some more samples to look at:

LinoType Greek Mac
The style looks very much like that of FreeSerif.
Now, the main range of FreeSerif comes from the Omega fonts,
so clearly those and the LinoType fonts share a common heritage.

The Biblical Greek Formum
"The best Greek fonts"
Compares Gentium, Bodoni, NeoHellenic
(Note only Bodoni does the kai replacement)

Andron Greek
various designs for psi
the regular tends to be have the symmetrical, palm-branch style.

There are also several MS Greek fonts to consider.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sun 01 Sep 2013 03:52:40 PM UTC, comment #1:

By the way, I think PSI is in need of an overhaul: I'd like to be more like what it is in Theano Didot & Old Standard TT, just like what you did to KAPPA, because it's the default style in Greece. Also see Old Standard TT to get an understanding of what italic PHI should look like.

(file #28973, file #28974)

Rosella Capriotti <rosy58>
Sun 01 Sep 2013 03:32:29 PM UTC, original submission:


I've taken a close look at the Greek italic glyphs, and indeed as you said more or less a month ago, the hodge-podge origin of glyphs is clearly visible: KSI is very cute, but PSI and maybe also PHI look just like a slanted version of the upright version, maybe (when you've got some spare time of course) you could do something to improve their look.

Rosella Capriotti <rosy58>


(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:


Depends on the following items: None found

Items that depend on this one: None found


Carbon-Copy List
  • -email is unavailable- added by hammerandsickle (Posted a comment)
  • -email is unavailable- added by Stevan_White (Posted a comment)
  • -email is unavailable- added by rosy58 (Submitted the item)

    Do you think this task is very important?
    If so, you can add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.




    Follow 8 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2015-11-17 hammerandsickle Attached File- => Added Proposal for Greek Theta in FreeSerif Italic.png, #35477
    2013-09-06 Stevan_White StatusNeed info => Proceeding
    2013-09-04 Stevan_White Attached File- => Added psi-two-forms.png, #29003
    2013-09-03 Stevan_White Severity3 - Normal => 1 - Wish
        StatusNone => Need info
        Assigned toNone => Stevan_White
    2013-09-01 rosy58 Attached File- => Added PSI in Theano Didot.png, #28973
        Attached File- => Added PHI upright & italic.png, #28974

    Back to the top

    Powered by Savane 3.3