bugFree UCS Outline Fonts - Bugs: bug #58309, Suggestions on selected Cyrillic...

 
 

bug #58309: Suggestions on selected Cyrillic characters. Part 2: the Iotified E

Submitted by:  Matvei Davydov <matt_d>
Submitted on:  Wed 06 May 2020 11:22:02 AM UTC  
 
Category:  individual character(s) Severity:  3 - Normal
Item Group:  character substitution issue Status:  Fixed
Privacy:  Public Assigned to:  Steve White <Stevan_White>
Open/Closed:  Open Release:  development

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)

Wed 09 Sep 2020 04:04:00 PM UTC, comment #19: 

Thank you for the explanation, Steve. (I tried to download the file before. That's why I couldn't understand what was going wrong).
But now I can clearly see the difference. The "modern" form looks wonderful. Good job!

Matvei Davydov <matt_d>
Tue 08 Sep 2020 06:47:53 PM UTC, comment #18: 

Find attached a PDF file showing how that test appears on my system.  I think this shows two different forms of "iotified e".

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 08 Sep 2020 04:53:49 PM UTC, comment #17: 

Hi Steve,

For now in your file I can't notice any difference between an "ordinary" form of Iotated E and a Bulgarian/Serbian one. Is that mean that a "modern" form appears only when using a Bulgarian/Serbian locale?

Matvei Davydov <matt_d>
Mon 07 Sep 2020 05:03:00 PM UTC, comment #16: 

Hava a look at the attached test file.

(file #49754)

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 07 Sep 2020 04:52:12 PM UTC, comment #15: 

Thank you for the news, Steve!

Please, let me know, if you decide to include the "modern" form of the Iotified E into the font.

Matvei Davydov <matt_d>
Wed 02 Sep 2020 08:58:43 PM UTC, comment #14: 

I implemented the iotified ie in sans as well.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 18 May 2020 08:59:27 AM UTC, comment #13: 

Hi Steve,

>If you could please type in the text from file 01.JPG that you posted, I will
>make a test file from it.  I just need the lines containing the two forms of
>the letter.
>

I'm afraid, I don't understand, what exactly you are asking me to type in. Did you mean to type the whole sentence from the book (in Cyrillic with a translation) or something else?

>It's particularly interesting that a specific reform made the "modern" form of
>the letter official.
>

Better to say semi-official. Gerov had published the first volume of his dictionary in 1895, but the first official reform of the Bulgarian alphabet was implemented in 1899, and then the Iotated E was excluded from the alphabet.

>I have already implemented something that may please you.
>

It's nice hearing that. Thank you!

Matvei Davydov <matt_d>
Sun 17 May 2020 04:35:35 PM UTC, comment #12: 

Hi,

This is great.  I need to find a way to incorporate such research into the project.
Strange that I hadn't considered that before.  There are decades of
good e-mails.

If you could please type in the text from file 01.JPG that you posted, I will make a test file from it.  I just need the lines containing the two forms of the letter.

It's particularly interesting that a specific reform made the "modern" form of the letter official.

I have already implemented something that may please you.
I'm a little stuck on a check-in right now due to ongoing efforts in Malayalam script.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sun 17 May 2020 11:11:31 AM UTC, comment #11: 

Hi Steve,

Unfortunately, my colleagues haven't helped me much. So I had to study that specific issue myself.

Well, what we know for now.
1. The variant of the Iotified E, that is presented in the Unicode tables (with 'e' resembling a Greek epsilon) was a primary form, that was used in Russian, Bulgarian anв Serbian manuscripts till the 15th century. Since that times it fell into disuse in Russia. In Bulgaria and Serbia this letter continued to exist, but the use of it became very limited. In the 16th century exactly the same form of the Iotated E one can see in Serbian first printed books, because the fonts in that books replicated manuscripts' handwriting. In the 18th century this letter became obsolete everywhere.

2. In the 19th century Slavic philologists tried to replicate the old texts as close as possible to the original look of the manuscript. (I've mentioned about it already). When the scientists didn't have all the needed glyphs, they ordered additional ones, taking as a model handwriting variants of selected letters. That's why in scientififc papers the Iotated E looked as descibed above ('e' in it is similar to a Greek epsilon).

3. In the middle of the 19th century the Bulgarians began the reform of their national alphabet and orthography. There were several suggestions (that's why the reform lasted nearly 100 years), but the main goal of all the projects was the same - the modernisation of the language, including the modernisation of letterforms.
The author of one of the projects was Naiden Gerov. One of his suggestions was a modernisation of a look of the Iotated E. He not only suggested a new, 'Latin-like' form of it, but also implemented its practical use. Gerov compiled a five-volume dictionary of the Bulgarian language (it was the first work of that kind and still one of the best), and his sympathizers helped him to publish it at the turn of the 19th and 20th century.
After Gerov's death the Bulgarians had simplified the orthography rules several times, and now they don't use the Iotified E anymore.

I've chosen some examples from the Gerov's dictionary and post them here.
At the first image you can see simultaneous use of both forms of the Iotified E - the 'archaic' (with an epsilon-like 'e') and the 'modern' one (with a Latin-like 'e'). As I could understand from the text, here it was Gerov's suggestion to modernize the form of the letter.
At the second image you can see a rare form of a capital 'modern' Iotified E. It's rare, because Gerov used that letter only in word endings.
Two other images demonstrate some cases of use.

As regards another ('archaic') form, I should stress, that in the scientific papers of the 19th century its form was similar to what it is now in the official Unicode tables and in the FreeSerif font. That's why I haven't added the images of it. But I can post that images later, if needed.


Matvei Davydov <matt_d>
Mon 11 May 2020 05:44:36 PM UTC, comment #10: 

Steve!

In general you are right, but one small point. Not "Western Slavic texts/languages" but Southern. Western Slavs (the Poles, the Chechs, etc.) use Latin alphabet, not Cyrillic.

Matvei Davydov <matt_d>
Mon 11 May 2020 01:48:30 PM UTC, comment #9: 

This is what I'm hearing:

The latinized form was most recently used for spoken-language text, primarily in Western Slavic texts.

The epsilon form was used in preceding centuries, and continues to be used to represent historical text.

Now, there is a special OpenType feature, 'hist', specifically for historical forms of letters.

If the Unicode samples had used the latinized form, this would have been easy: I would just make the epsilon form accessible via 'hist'.

There is also the option of an OpenType "Character Variant", but this is devoid of meaning, and intrinsically font-specific -- a user has to know that this font has a variant letter, with a certain index, specified in the font's documentation.  In contrast the feature 'hist' has a meaning.

On the other hand, there is at least one case where I chose to differ with the Unicode samples: they showed a rare form of a letter, and I chose to make the default to be a form in much more widespread use.  However, in the present case, as I understand it, the letter isn't in common use in any language.  It seems a weak reason go against the standard.

Hm.  Could make Western Slavic languages take the modern form by default, and then that could be reversed by specifying 'hist'.  That would satisfy the use-cases.  But it's pretty complicated.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 11 May 2020 06:24:15 AM UTC, comment #8: 

Hi Steve,

What did you mean while using the word "normal"? The more widespread variant? If so, that is the variant with an epsilon-like E.

As regards the circumstances an epsilon-like variant was published many times from the 16th up to the 18th centuries in the Serbian books, and in the 19th century it was widespread in Russian scientific publications of medieval texts.

A modern Latin E variant was in use in the 19th and the early 20th centuries in Bulgaria and Macedonia (maybe in Serbia as well). I can't exclude the possibility that then it was used sometimes in scientific publications as well, but I still couldn't find any examples. I hope that my colleagues clarify this issue soon.

Finally, since the beginning of the 20th century both variants are used extremely rarely.

Matvei Davydov <matt_d>
Sun 10 May 2020 12:50:37 PM UTC, comment #7: 

Hi,

The use of "normal" here can only be confusing.  Is that the one that looks like a modern Latin e?  Or the one that looks more like a Greek  epsilon?

I want to understand the circumstances in which each form has been used.  That doesn't strictly require an image, if we already have good examples.  But an image certainly doesn't hurt.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sun 10 May 2020 06:48:06 AM UTC, comment #6: 

Hi,

1. As regards the images of the Iotified E that was used in Bulgarian/Serbian books.
Because of the covid carantine many people in Russia are in the country now, with no access to the internet, books, etc. I hope, I'll get the answer from my colleagues soon (in a few days or so). I'll report here, when I get any new information.

2. As regards the images of the 'ordinary' variant of the Iotified E. I should stress, that variant has absolutely no differences with a present form of the Iotified E used in the Unicode tables and in the FreeSerif font. In that case, is it really necessary to post here more images of that 'ordinary' variant?

Matvei Davydov <matt_d>
Sat 09 May 2020 08:21:14 AM UTC, comment #5: 

Looks like we're making progress.

I really do want to see more example images!  Preferably of both forms!

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Fri 08 May 2020 09:40:56 AM UTC, comment #4: 

Hi Steve,

> Hi,
>
> My question though, is was the "rounder" form (small or capital), as in the
> Unicode Standared ever printed?  Why and when?
>

I got your meaning finally. Of cource, the form presented in the Unicode Standard was published many times (and for the first time in the late 15th century). Moreover, in the 19th and the early 20th centiries exactly that form dominated in the scientific papers. (In that times scientists usually tried just to replicate the appearance of the old texts. So they didn't touch the abbreviations, superscripts, and used the letterforms that resembled the shape of the original characters). Later on many specific characters like the Iotated E fell into disuse for a number of reasons. And finally, while the Unicode tables beeing created, exactly that stylized and artificial form became a standarded one.

> > That's why I'd better do another thing. I'll consult with my colleagues
> > philologists about the usage of the Iotated E in the modern typography
> > (since the middle of the 19th century) and then inform you what the results are.
>
> Very good.  But again, I am much more easily convinced by evidence than by opinion.
>

When I wrote about the consultations I meant that my colleagues may provide me with useful references and images. And then I could report new information in this thread.

> As to a policy: I do prefer a recent common printed form (NOT modern computer fonts).
>

I fully agree with you.

> If the rounded form is just based, say, on manuscripts, and was never printed, then
> I would definately go for the printed form.
>
> If both forms have been printed... perhaps I could make the form a font feature.
>

Your last assumption is right. Both forms have been in use.

> But there are alternatives.  For instance, the "modern" form might be the
> default in Bulgarian.  But either form might be reachable by some font feature.
> (You can read about features that appear in FreeFont in its documents.)
>
> The problem is, we have at least two forms of these letters, but only one concrete
> use-case (the relatively recent Bulgarian).  Yet I know the letters had been used somehow
> previously, and the form in the standard is different from the Bulgarian one.
>
> From my perspective, this isn't an either-or proposition.  There are various ways
> the situation might be resolved. But to do that, we (that is, I) need a better
> understanding of the use cases.
>
> Thanks!

Nice to hear that. Thank you!

Matvei Davydov <matt_d>
Thu 07 May 2020 04:13:11 PM UTC, comment #3: 

Hi,

    >> Your argument is pretty compelling, especially with the images.
    >>
    >> I would also like to see an image of the capital letter in older print.  And I would like to know if the other form was ever printed.  (Is it printed this way, say in liturgical texts?)
    >>

>It's not so obvious, Steve. The first problem is that in the Slavonic early
> printed books capital letters were rarely used up to the end of the 17th
> century. The second problem: characters in the fonts that were used in the
> early printed books don't have very much in common with the ones in the
> modern fonts. And the last important problem: since the Iotated E was a
> 'rare bird', it'd be rather hard looking for it in old books.


About the capital form: that is clear.

My question though, is was the "rounder" form (small or capital), as in the
Unicode Standared ever printed?  Why and when?

> That's why I'd better do another thing. I'll consult with my colleagues
> philologists about the usage of the Iotated E in the modern typography
> (since the middle of the 19th century) and then inform you what the results are.


Very good.  But again, I am much more easily convinced by evidence than by opinion.

    >> Please compare the case of iotified 'a' discussed in this paper
    >> Unicode 5.1, Old Church Slavonic, Remaining Problems
    >> – and Solutions, including OpenType Features
    >> https://kodeks.uni-bamberg.de/slavling/downloads/SK_Unicode_5.1_OCS_OTF.pdf
    >>
    >> My feeling is: I would prefer the form used most recently in any printed media.
    >>

> I read that Sebastian Kempgen's paper years ago and I can't agree with his
> position at all. I think you know his fonts very well and their weak spots
> either. As for now I should stress only one thing. His desire to just replicate
> the shape of the characters the way they looked in old manuscripts leads to
> nothing good. I don't know anyone in Russia who uses his fonts, people in
> scientific papers use other fonts instead - FreeSerif, Linux Libertine, Tempora, Lytopis, etc.


I know nothing more than what he wrote there. 
I have no personal preference for one form or another.

The point is, he addresses the question, and compares some alternatives.
I need to see the alternatives, and understand where they come from.

As to a policy: I do prefer a recent common printed form (NOT modern computer fonts).
If the rounded form is just based, say, on manuscripts, and was never printed, then
I would definately go for the printed form.

If both forms have been printed... perhaps I could make the form a font feature.

    >> My concern is, we may be looking at a language-specific form.  For instance, FreeFont already has variant letter forms for Bulgarian and Serbian (check out the documentation!)
    >>

> As I understood, you suggest to change the shape of a character depending on
> locale settings. I'm worried that it is not the best decision, because with
> this approach while using Russian locale I won't be able to see a desired
> variant of the Iotated E.


Yes, that was the suggestion.

But there are alternatives.  For instance, the "modern" form might be the
default in Bulgarian.  But either form might be reachable by some font feature.
(You can read about features that appear in FreeFont in its documents.)

The problem is, we have at least two forms of these letters, but only one concrete
use-case (the relatively recent Bulgarian).  Yet I know the letters had been used somehow
previously, and the form in the standard is different from the Bulgarian one.

From my perspective, this isn't an either-or proposition.  There are various ways
the situation might be resolved. But to do that, we (that is, I) need a better
understanding of the use cases.

Thanks!

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Thu 07 May 2020 02:41:39 PM UTC, comment #2: 

Hi Steve,

комментарий №1:

> Hi,
>
> Your argument is pretty compelling, especially with the images.
>
> I would also like to see an image of the capital letter in older print.  And I would like to know if the other form was ever printed.  (Is it printed this way, say in liturgical texts?)
>

It's not so obvious, Steve. The first problem is that in the Slavonic early printed books capital letters were rarely used up to the end of the 17th century. The second problem: characters in the fonts that were used in the early printed books don't have very much in common with the ones in the modern fonts. And the last important problem: since the Iotated E was a 'rare bird', it'd be rather hard looking for it in old books.
That's why I'd better do another thing. I'll consult with my colleagues philologists about the usage of the Iotated E in the modern typography (since the middle of the 19th century) and then inform you what the results are.

> The letter in FreeFont is my design, based approximately on the Unicode samples
> http://www.unicode.org/charts/PDF/U0400.pdf
>

Thank you for the information!

> Please compare the case of iotified 'a' discussed in this paper
> Unicode 5.1, Old Church Slavonic, Remaining Problems
> – and Solutions, including OpenType Features
> https://kodeks.uni-bamberg.de/slavling/downloads/SK_Unicode_5.1_OCS_OTF.pdf
>
> My feeling is: I would prefer the form used most recently in any printed media.
>

I read that Sebastian Kempgen's paper years ago and I can't agree with his position at all. I think you know his fonts very well and their weak spots either. As for now I should stress only one thing. His desire to just replicate the shape of the characters the way they looked in old manuscripts leads to nothing good. I don't know anyone in Russia who uses his fonts, people in scientific papers use other fonts instead - FreeSerif, Linux Libertine, Tempora, Lytopis, etc.

> My concern is, we may be looking at a language-specific form.  For instance, FreeFont already has variant letter forms for Bulgarian and Serbian (check out the documentation!)
>

As I understood, you suggest to change the shape of a character depending on locale settings. I'm worried that it is not the best decision, because with this approach while using Russian locale I won't be able to see a desired variant of the Iotated E.

Matvei Davydov <matt_d>
Wed 06 May 2020 07:06:48 PM UTC, comment #1: 

Hi,

Your argument is pretty compelling, especially with the images.

I would also like to see an image of the capital letter in older print.  And I would like to know if the other form was ever printed.  (Is it printed this way, say in liturgical texts?)

The letter in FreeFont is my design, based approximately on the Unicode samples
http://www.unicode.org/charts/PDF/U0400.pdf

Please compare the case of iotified 'a' discussed in this paper
Unicode 5.1, Old Church Slavonic, Remaining Problems
– and Solutions, including OpenType Features
https://kodeks.uni-bamberg.de/slavling/downloads/SK_Unicode_5.1_OCS_OTF.pdf

My feeling is: I would prefer the form used most recently in any printed media.

My concern is, we may be looking at a language-specific form.  For instance, FreeFont already has variant letter forms for Bulgarian and Serbian (check out the documentation!)

Let me know what you find!

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 06 May 2020 11:22:02 AM UTC, original submission:  

Hi Steve,

Once you have already changed the look of the Iotified A, and you did it right. Kryukov's variant was nothing more but graphic stylization, and it looks ugly when surrounded by modern charcaters (so called Russian Civil Script). And your variant (the combination of modern-like I and A) looks natural in that surroundings. See the first picture attached.

And now I ask you to do the similar thing with the Iotified E, both upper-case (U+0464) and lower-case (U+0465). The main reason is absolutely the same - the combination of I and the Ukranian Ye (U+0404 and U+0454) looks like an alien in Civil Script surroundings. Moreover, there are other reasons as well:

1. In the Iotified E the second part (Cyrillic E character) was just a variant for an 'ordinary' E, that had a specific shape - so called open, or 'anchor-like' E. Nevertheless, both variants of E denoted the same sound.

2. The Ukranian Ye, which is used now as the second part of the Iotified E, denotes itself the same sound as the Iotified E formerly did. So, it's obvious, that the use of the Ukranian Ye in the Iotified E is really excessive (instead of 'ordinary' E).

3. Though in almost all Slavonic languages the usage of Iotified E ended in the 15th century, the Bulgarians, the Macedonians and the Serbs have been using it (rarely) up to the beginning of the 20th centure. And, what's the most important, then the Iotified E looked like the combination of modern Cyrillic I and E. See the second and the third pictures attached (both are from the Bulgarian books).

So I ask you to change the shape of the Iotified E on something like that is drawn at the fourth picture.(I should add, the location of the horizontal connector between I end E is not mandatory. If you think, that it should be placed higher, or it must match with the horizontal line of E character, I'll agree with your decision).

Matvei Davydov <matt_d>

 

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

Attach Files:
   
   
Comment:
   

Attached Files
file #49097:  01.JPG added by matt_d (63KiB - image/jpeg)
file #49098:  02.JPG added by matt_d (30KiB - image/jpeg)
file #49099:  03.JPG added by matt_d (85KiB - image/jpeg)
file #49100:  04.JPG added by matt_d (157KiB - image/jpeg)
file #49008:  Iotified@A.JPG added by matt_d (51KiB - image/jpeg)
file #49011:  Iotified@E@suggestion.JPG added by matt_d (46KiB - image/jpeg)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by Stevan_White (Posted a comment)
  • -email is unavailable- added by matt_d (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 13 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2020-09-07 Stevan_White Attached File- => Added old_cyrillic.html, #49754
    2020-09-02 Stevan_White StatusProceeding => Fixed
    2020-05-17 Stevan_White StatusNeed info => Proceeding
    2020-05-17 matt_d Attached File- => Added 01.JPG, #49097
        Attached File- => Added 02.JPG, #49098
        Attached File- => Added 03.JPG, #49099
        Attached File- => Added 04.JPG, #49100
    2020-05-06 Stevan_White StatusNone => Need info
        Assigned toNone => Stevan_White
    2020-05-06 matt_d Attached File- => Added Iotified@A.JPG, #49008
        Attached File- => Added Iotified@E@Bulgarian_1.gif, #49009
        Attached File- => Added Iotified@E@Bulgarian_2.jpg, #49010
        Attached File- => Added Iotified@E@suggestion.JPG, #49011

    Back to the top


    Powered by Savane 3.7