Junicode-font
Junicode-font copied to clipboard
psb1558
Old Polish b/p quadratum/rotundum as PLK in ss09
These characters are often mentioned in the literature on the history of Polish but cannot be quoted because the glyphs are not available. For more information see pages 8-10 of https://www.researchgate.net/publication/350449200 and/or pages 42-43 of https://pressto.amu.edu.pl/index.php/pspsj/article/view/19852.
Interesting. I'd like to be very clear about what differentiates b/p molle from the regular b/p. For the italic I can see the difference in the images from Kucała, but it would also help (especially for figuring out what the roman should look like) to see what regular b/p look like in the manuscript you cite.
Parkosz had no intention to use regular p/b and the manuscript is a copy made by two scribes with different handwriting styles. Below please find some examples from the first and second part of the treatise: eligibilitatem, scribamus, Topicorum, exempla:
The thread on MUFI-fonts list (accessible only by subscribers): https://groups.google.com/forum/#!topic/mufi-fonts/fmncUGKG9Kw.
Sorry I've been neglecting this. Getting ready for another term on Zoom, and then supplying hints for Elstob.
I'm still having trouble visualizing what these glyphs should look like, but will post some ideas when I get a chance.
I'm a big Gimp fan, btw.
I've prepared a note about encoding Parkosz's alphabet with a tag character. Please contact me (preferably by mail) if you want to have a look.
The Parkoza tract is fascinating, and it's great to have both the MS facsimile and a modern edition to work with. I've been thinking about what glyphs should be added to handle it. This is kind of rough, but a start:
We can discuss the tags once the collection of glyphs is settled.
Very nice! Thanks!
B durum and b molle are easy to distinguish thanks to the loop, but I'm afraid p durum i p molle look too similar for a typical reader. What about e.g. p molle
Or p (and b) square can be "more square"...
I'll fiddle with those shapes a bit. p and b durum are difficult because their shapes are hard to fit into a standard roman design scheme.
The p molle looks to me like an ordinary late-gothic p. Is there a contrast in the MS between p molle and the standard p?
Here's a bit of a change for l + loop, which was too easy to mistake for a form of long s or f:
Is there a contrast in the MS between p molle and the standard p?
You can find in my old comment above two instances of p in Latin words: Topicorum, exempla. For me they look more like p durum. I will search for more examples later.
Here's a bit of a change for l + loop
It's OK for me.
Second scribe:
Okay, I see it (I think). He's exaggerating the roundness of p molle and omitting the top of the left-hand vertical stroke. Makes it look a little like a Greek rho. From MS p. 6:
Exaggerating the roundness won't work very well when the bowl of the Junicode p is already round, but we can certainly omit the top of the right-hand stem:
It's possible to subtly emphasize the quadness of the quadrata letters by making one side of the bowl a little concave:
I've made the tail of this alternate g a bit less extravagant to reduce the danger of collisions.
t's possible to subtly emphasize the quadness of the quadrata letters by making one side of the bowl a little concave:
Looks nice. Alternate g is also OK.
Okay. The roman version of the font (I'll add italics eventually) now has what I hope is a complete collection of Parkosz's special phonetic characters:
The existing tags in Junicode are listed in ch. 5 of the Junicode Manual. There you can see that a character is always defined by a sequence of base character + two tags. The font contains tags for a-z and 0-9 (plus a special "cancel" tag). I have no objection in principle to adding other tags, but it seems unnecessary in this case, since none of the present two-tag sequences begins with p.
For base characters I very strongly prefer the most common possible alphabetic characters rather than IPA or other exotics. This is for accessbility, ease of use, and the searchability of text; also because the tagged characters have got to be duplicated in the cvNN feature set, and there is no room in the set for new base characters.
One could simply use the tags to number the variants, but I prefer mnemonics where possible. So I'd suggest a series of base+tag sequences looking like this:
b&__p;&__s; b square
b&__p;&__l; b with loop
c&__p;&__h; c with hook
g&__p;&__h; g with hook
l&__p;&__l; l with loop
p&__p;&__r; p round
p&__p;&__s; p square
v&__p;&__h; v with hook
By the way, these characters strike me as good candidates for Unicode encoding. I'm thinking of the recent additions of characters from another early work with great importance for the history of a particular language: the 12th c. Middle English Ormulum, which is analogous to Parkosz's work in a number of ways. I wonder if Michael Everson would be interested in helping to work up a proposal.
I should add that it might be desirable to have "p" tags for characters in Parkosz that are already in Junicode and accessible via other tags: m with descender (MUFI U+F223), n with descender (MUFI U+F228), perhaps others.
Italics OK.
I should add that it might be desirable to have "p" tags for characters in Parkosz that are already in Junicode and accessible via other tags: m with descender (MUFI U+F223), n with descender (MUFI U+F228
I fully agree.
So I'd suggest a series of base+tag sequences looking like this:
Your tags proposal is definitely more convenient than mine.
Now looking forward to majuscules.
Middle English Ormulum, which is analogous to Parkosz's work in a number of ways.
You are right, I will look into it.
As for #24, ss will be 09 as mentioned there? What about PLK value? Is it still needed? I understand both answers are negative, ss is 10. Am I right?
I forgot about the "Majuscules" section of your paper.
Are these (on p. 6) fair examples of capital P molle and P grossum?
ss09 is still reserved for language-specific variants, though there aren't any yet (in Elstob I'm using it for Vietnamese and English). But ss09 is most appropriate for variants that are purely stylistic (these are substantive). Also, since the tag sequence is unique, there's no need for a language-specific feature in the font: the tags will do it without the added complication.
Yes, just ss10 to activate the tags.
No diacritics.
Difficult to say which examples of capital are best, especially as there are two scribes. Below are examples from the largest sample of the Polish text (last but one page).
According to Kucała's interpretation all P are capital, the last one is P molle. The interpretation is said to correct scribe error, so in practice you can't take it for granted.