blackfoundrycom / abbagarima Goto Github PK
View Code? Open in Web Editor NEWLicense: Other
License: Other
Some refinements to make for the ሩ glyph would be to maintain the shape of the refined ረ glyph, but do not compress the body to a smaller size (or very maybe just 10%). Then:
Page 057v:
This first example is probably the best one:
Page 057r:
The body shape is good but the foot in this case is a little short and too vertical:
The foot is good in this sample on the same page, but tthe lower stroke extends too far:
The U+1310 glyph features a diacritical mark fairly similar to the modern form. It appears to be attached to the body at fairly consistently the 50% vertical point. I think the slightly triangular shapes are the best style:
From Image 041v:
From Image 056v:
From Image 057v:
From Image 061v:
From Image 080v:
A sample could not be found for the U+12EC glyph, but it can use the diacritical mark developed for ጐ (Ref: #28) and positioned at the same location as the ዩ (U+12E9) diacritic.
The Gospel of Mark in the Garima Gospels has only two occurrences of the letter U+12B2 in the word ወአእኲቶ . One is illegible because the page is too faded. The other appears on Image 079v, it could easily be mistaken for the letters ነዚ.
The diacritical mark here could be used as a basis for companion syllables where samples are missing (e.g. ቊ U+124A, ኊ U+128A, and ጒ U+1312).
The present state of the font uses a roundish head (top loop) for the ጸ syllables, which is found in many examples, however, the more triangular head shape appears to be the predominant form and the most characteristic of the Gospels. This enhancement request is to change to the triangular shape for the entire family and to include a start of stroke mark that will be illustrated herein.
An example of the preferred head shape is see on image 55v:
The above samples show the more triangular head shape very clearly but do not exhibit a stroke artifact that most ጸ syllables will have. The artifact is an extra bit of ink that appears at the start of the top stroke. The following sample from image 57v presents ጸ with the artifact clearly:
The following two images from image 73v are the same image with the top left side mark highlighted for clarity. The images in samples of Issue #11 and #12 also feature the artifact.
The glyph for U+1214 applies an arm-and-ring diacritical mark (appropriate for U+1E7E8), but should use a ring on the right foot. Two samples are below, note that the ring shape is more square-like than in modern form. The word እግዚአብሔር is fairly common and I will continue to look for a representative sample.
The top loop of the ፎ syllable has a frequent variance in the Garima Gospels where the loop may have been a challenge for the scribe to shape consistently. The current design has the loop appearing a little too large than average and in contact with the central vertical stroke. An improvement would be to make the loop a little smaller, more circular, and a small distance from the central stroke. The following images provide good reference glyphs:
Image 69r:
Image 71r:
So far only one sample of U+1314 has been identified in the word "ጒጔጕእ" , and it appears to use the same diacritical style as with U+12B4, ኴ, in #43 .
From Image 063r:
The Garima ዚ glyph appears to have a play bend on the right leg, appearing like a running futball player about to kick a ball. As usual, there are variations in the handwritten form, note the shape of the leg bend and how the "knee" appears above the baseline.
From Image 057r:
From Image 060v:
From Image 75r:
From Image 79v:
From Image 083r:
The glyphs for Ethiopic numbers 8 (፰) and 60 (፷) are very similar to ፐ (U+1350) and ፕ (U+1355) respectively where the bottom of the glyph has a stroke mirroring the top. Ethiopic 9 (፱) has a body similar to ሀ (U+1200) but more narrow.
The modern form of Ethiopic number Seventy, ፸, in modern form is similar to ሮ (U+12EE), in the Garima Gospels the glyph follows this outline but the top loop does not close. Ethiopic number Eighty, ፹, is much like Greek Pi, or like ፐ with two vertical strokes.
All of the numerals should have the same height if the heights are close. The irregular number shape of ፮, would be an exception that can be smaller in height.
From Image 081v:
This sample also has an example of Ethiopic Ninety, ፺, which can be made identical to ን (U+1295) but smaller to fit within numeral bars:
From Image 105v:
From Image 153v:
This sample also has examples of Ethiopic Sixty (፰) and Seventy (ሮ):
From Image 155v:
This sample also has examples of Ethiopic Seventy (ሮ) and Eighty (፹):
The Garima Gospel does not appear to have an example of the ዉ glyph based on available transcription, though we will continue to seek out one.
In the absence of a sample, we should follow the modern form where the diacritical mark appears near the bottom of the glyphs. The Garima ፂ (U+1342) glyph in the font is a helpful example showing a diacritic mark also at the bottom of the glyph (though of different order) that can provide inspiration.
The Garima ጾ glyph is very similar to the ጸ (U+1338) glyph where the right foot appears only slightly raised. Note that the curvature of the right leg does appear to change where it will appear to descend nearly vertically. Good examples of the ጾ glyph appearing distinct can be seen in Image 68v:
as well as in image 83r:
The ኦ (U+12A6) glyph is often nearly the same as the አ (U+12A0) glyph in the Garima Gospels, in part due to a slant in writing found on many pages. The glyphs should be generally identical with the distinction being the shorter right leg of ኦ፡
Image 55r has an example in the word ወወፂኦ which occurs frequently:
Another good example is found on image 56v where we see ኣ (U+12A3) and ኦ (U+12A6) together in the same word. Note that the leg shortening is in the same amount:
The glyphs of the Ethiopic Numerals Forty and Fifty are identical or nearly so to the syllables ሣ (U+1223) and ሃ (U+1203), the shapes may be slightly smaller to then fit within the upper and lower numeral bars. In the Garima Gospels, the numeral forty glyph appears to have the first stroke on the left side start a little lower than the middle or final stroke.
From Image 041r:
From Image 077v:
From Image 113v:
From Image 005, reviewing other samples there appears to be no difference between the U+1203, ሃ, glyph and the U+1376, ፶, body. The top-left stroke may start a little lower than the right stroke, perhaps 50% of the time:
From Image 005v:
From Image 062r:
Samples for the U+1325 glyph follow, they are found infrequently. Note that the final leg appears to lift a little, as was the case for U+1321, ጡ. I'll review to see the raised final leg may also be the case with U+1322, ጢ, it is often subtle.
From Image 081v, note that the size of the diacritic loop is a bit small.
From Image 081r, the size of the diacritic loop is good here though it does not appear lifted, thus combining the two would be ideal.
The shape of the ረ is too close to the modern form and should be adjusted to be very similar to the shape of the sixth form, ር. The images below will show examples of each:
Page 057r has some samples of ረ and ር for comparison. All ረ family are highlighted for reference:
ረ - red
ሩ - green
ሪ - white
ር - blue
ሮ - yellow
The 1st and 3rd ረ samples highlighted above are the best, the shape of the 3rd might be best overall. There is some distinction seen from ር in the shape of the left side arch, particularly on the top half which has some angle to it where the top half of the ር glyph remains more open and squarish like the modern form.
The ጰ family of syllables can follow the work already done for the ጸ (U+1338) family, with the top part added. The existing ጰ glyph in the font is a very good basis. The main consideration for these shapes is the height of the mark on top of the glyphs, in this case, the Garima style uses an extended mark versus modern form, so here I recommend making the mark on ጰ about 50% longer (higher). I'll add samples to this ticket.
In older manuscripts the sixth order ጵ (U+1335) may take on a different shape, so far I have not located a sample in the Garima Gospels so it is OK to follow the ጽ (U+133D) shape here.
Samples from Image 081r of ጲ , these samples show the height and shape of the top mark. Note that there is a small curve in the mark to the right.
Image 091v has a sample of ጰ which again helps illustrate the relative size of the mark:
Image 167r has a sample of ጴ, again offering an example of the top mark:
on commit 4661724
The WAA diacritical mark of U+1313 should be identical to U+12B3, ኳ. Samples from the Garima Gospel appear in the following:
From Image 057r:
From Image 065v:
From Image 067r:
From Image 067v:
From Image 069r -this may be the best sample:
From Image 072v:
From Image 072r:
From Image 080v, a nice shape of the diacritic but unfortunately distorted:
The Garima Ethiopic Numeral 30 is similar to the modern form with the exception that the left-side begins closer to the baseline, perhaps having 1/3rd the total letter height and the right-side stroke is 2/3rds of the total height.
From Image 075:
From Image 076:
From Image 145:
From Image 170:
From 186, the glyph shapes for ፫ (U+136B), ፯ (U+136F), ፰ (U+1370), ፳ (U+1373), ፶ (U+1376), and ፻ (U+137B), are also very good:
A few of the numerals of the Garima Gospels have shapes slightly different from their modern counterparts. The Ethiopic Digit One, ፩, appears something like a rounded Lozenge symbol ◊ (U+25CA). Though in many samples it may appear more like a modern zero. Some useful examples:
From Image 55v:
From Image 75r:
From Image 81r:
From Image 185, here the numeral ፬ is highlighted with a blue underline to help distinguish it from ፩. The samples highlighted with a box are the best ones in this group. The 2nd box, on the right side, would be the best model for the font.
The ፐ - ፗ syllables are relatively infrequent in the Ethio-Semitic languages, in Ge'ez in particular. So far I have only managed to find two samples from the Garima Gospel 2 (HMML numbering) and they are both a little irregular. The samples are useful to show the height of the letters and the width of the top stroke, but are limited as examples of the diacritical marks. The diacritic marks applied to ዘ and ተ could be a useful reference since the marks are attached to relatively vertical strokes.
A sample of ፒ (U+1352) from Image 068r in the word መሀፒል . What is unexpected here is the bottom stroke extending to the left, as if it were a mirror of the top stroke. It could be that the original design was based on the inner numeral glyph ፰ (8), or the scribe may have accidentally added the bottom stroke as a reflex because he was more accustomed to writing the similar numeral than the letter.
A sample of ፔ (U+1354) from Image 060v in the word ስናፔ. It has a hint of a stroke to the left at the bottom of the glyph, but it is more like a large serif:
Unfortunately, these may be the only samples to discover and they are contradictory with respect to the bottom-left stroke. Thus, I think until more samples can be found to resolve the contradiction, it would be best to make the diacritical marks in the modern way, with the more regular diacritic marks applied to the right side.
An example of ፃ appears on image 55r (along with ፁ, ፅ, and ፆ) in the word "ወፃእ" at the end of the 2nd line, and ፆ at the end of the first word, "ወአስተራገፆ", on the 5th line.
Note the short leg, the same size and shape of the ፃ and ፆ bodies, and the end of the strokes in the top left corner of ፃ (a serif?) and lower right corner of ፆ .
Another example is on image 69r in the word "ታፃእ" on the 2nd line, and a ፆ example on the start of line 6 in "ስተራገፆ":
applying rounded corners in countershape 44b6040
The ኁ glyph can be created from its base ኀ (U+1281) with the addition of the "u" diacritical mark. Note that the diacritical mark is attached to the body at a higher than usual position, essentially at the bend in the leg stroke.
Two examples are seen on image 63v and are highlighted in the attached image:
The Garima ፊ glyph is essentially the ፈ (U+1340) glyph with the addition of a diacritic mark in the form of an upward stroke at the lower end of the glyph. Note that it is a nearly straight upward stroke and not curled inward as per the modern practice.
An example from the right side of image 67:
Two examples from the bottom of the right side of image 75:
The Garima Gospels has an irregular labialized -WEE diacritical mark. Earlier I had misidentified it as the -WAA diacritical mark. Thus, from this error, the font currently has the U+12B4 glyph in the position of U+12B3. The correction would be to move the U+12B3 glyph over to the U+12B4 position.
From Image 072v, we see an example of the U+12B4, ኴ, glyph:
From Image 063v, we see the correct U+12B3, ኳ, glyph:
The Ethiopic Digit Two glyph (U+136A) appears like a Greek Capital Beta, β.
From Image 067r (each letter in this sample is beautifully written), the Ethiopic Numeral Eighty, ፹, here is also a good quality sample:
From Image 75r:
From Image 151v:
From Image 170v, the sample here for Ethiopic Numera One (U+1369), is also very good:
The Garima glyph for the Ethiopic Digit Six is unlike the modern form, it appears something like a small skewed "z". Where the Ethiopic Digit Seven in the Garima Gospel is more like the modern form where the body is similar to ጊ (U+130A).
The following samples may include other numbers to help as a reference for size comparison. The U+136E glyph (Ethiopic Six) is underlined in red, and U+136F (Ethiopic Seven) is underlined in blue.
From Image 19r:
From Image 107:
From Image 116v:
From Image 117v:
From Image 121v:
From Image 141r:
From Image 145v:
From Image 147r:
From Image 155r:
From Image 173r:
From Image 174r:
From Image 185:
An example from image 55r at the bottom of the left column in the words "ንቲኣሃ" and "ወአንሥኣ". The 2nd word offers a comparison to አ the (U+12A0) glyph:
Another good example is found on image 56v where we see ኣ (U+12A3) and ኦ (U+12A6) together in the same word. Note that the leg shortening is in the same amount:
I couldn't find an example and it seems it may look very similar to ፍ (U+134D) because of the base shape that extends left on top
The Garima Ethiopic Digit Four glyph is indistinguishable from the Aynu-A, ዐ (U+12D0), except possibly slightly smaller to fit within numeral bars -this is also true of the numbers 50 (based on ሃ, U+1203), 40 (based on ሣ, U+1223), 90 (based on ን, U+1295), and 100 (based on የ, U+12E8).
Image 81r (below) shows ፬ in blue underline, and ፭ in red underline. Note that the back stroke of ፭ will lean forward more in the Garima form and the curvature may have the top slightly hang over the middle stroke:
More samples from Image 121r (note the shapes of ፺ and ፻) are good here:
From Image 039r:
From Image 115r:
From Image 129r:
ጥ is one of the less consistently written symbols, it may appear nearly identical to ጠ in some cases. A clear form of it will have slightly shorter left and right legs than ጠ , the middle leg will then have the regular length with a "krick" toward the bottom (below the mid-point of the leg).
Two examples are seen on image 67r in the words "ወተሰጥዎ" and "ጴጥሮስ".
Another example from page 65 in the word ጥምቀታተ፡
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.