Mike Hebrew V1.01

Mike Thompson · Post by **Mike Thompson** » Mon Sep 13, 2004 4:47 pm

Hello everyone, I have a new version available on
http://www.mikethompsonpaintings.com/font

This features a font family of regular, italic and bold.
Currently there is no bold-italic.

Sections of bold or italic should now look ok when embedded
in regular text. The design of the three has diverged a bit since
the last release. Italic has become a bit more idiosyncratic
while regular is becoming, well, more regular.

Comments are welcome.
Please let me know when something doesn't work or is missing!
Mike

Dick Pape · Post by **Dick Pape** » Wed Sep 15, 2004 6:34 am

Hello Mike Thompson --

I can't read the language but do appreciate the effort you have made designing the fonts. Good work.

I'm asking about the mapping relationships of the Hebrew letter to the Latin letter to the glyph. I am using the Unicode table as a guideline, so the names below may be wrong...

1. I assume your order or sequence of glyphs is correct, but it seems out of place (but it could be right) to have a "letter with accent" before the "letter". Glyph 33 is HEBREW LETTER BET WITH DAGESH whereas glyph 65 is HEBREW LETTER BET. Is there a Hebrew "collating sequence" which arranges the letters in an ascending order?

As it is confusing, don't use A-Z, a-z, but put them in the correct order and serially assign Hebrew/unicode glyph names.

2. Glyph 33, the letter "A", has been mapped to (decimal) 65 LATIN CAPITAL LETTER A and 64305 HEBREW LETTER BET WITH DAGESH.

a. Why doubly map the glyph to Latin and Hebrew? Of what value or use is the Latin map?
b. Why not use BETDAGESH rather than "A"? Sounds more Hebrew!

3. Glyph 64, the letter afii57664, is mapped to 96 GRAVE ACCENT and 1488 HEBREW LETTER ALEF.

a. (Same question on double maps).
b. Why use the AFII code versus another alternative such as alef or alefhebrew?

4. Glyphs 152 to 177 are standard unicode characters which apparently have random names assigned. If they are relevant to your Hebrew font, they should be assigned to the unicode standard maps. That is, glyph 152 (§) should be called "section", not Ecircumflex and mapped to decimal 167, not 202.

Based on my rather limited knowledge, I would not include the latin maps and limit the definitions to the Hebrew names. It introduces a Latin-tinge which is not to be used I guess. (Setting the keyboard entry is another question I know even less about.)

FCP: re-Validate the glyphs to clean up the "off-curve extremes" without changing the shapes. You still end up with zero errors which is good. I sorted the glyphs and the order changed, I'm not sure where or why. Both of these tools would help standardize the font.

You've done all the hard part, so it's easy to nit here and there.

Good luck with your Bold Italic!

Dick

Yehuda · Post by **Yehuda** » Wed Sep 15, 2004 10:31 am

Since Mike hasn't answered this yet, I'll field it for him.

The problem with Hebrew is that there are lots of different mappings that have been used for it -- seven-bit, eight-bit DOS, ANSI, Unicode, proprietary mappings, .... (This is also true of other non-Latin scripts.) While Unicode is now supposed to be the standard, not everybody uses it.

This particular font is designed to be used with a Hebrew/English word processor (DavkaWriter) which uses a proprietary mapping. The Latin-letter mappings that Mike has are the right ones for the program. I would assume that Mike put in the Unicode mappings so the fonts can also be used by other programs. As for the 152-157 range, these are the mappings used for Hebrew text by DavkaWriter. I agree that they could/should have done things differently, but that's life.

BTW, I can't get to the font. I keep getting a "server unreachable" message

Erwin Denissen · Post by **Erwin Denissen** » Wed Sep 15, 2004 11:55 am

A couple of days ago Mike asked me to upload them, so here are the direct links located on this server:
Mike Hebrew Regular
Mike Hebrew Italic
Mike Hebrew Bold
All fonts are v1.01.

Mike Thompson · Post by **Mike Thompson** » Tue Sep 21, 2004 10:58 pm

Dick,
I just returned from a visit to relatives. I'll be studying your comments shortly.

What I did was to take a font supplied by DavkaWriter and replace their glyphs with my own. The Davka fonts make no mention of Hebrew in the mappings. This makes things confusing for a font designer.

After I had got it to work on DavkaWriter, I added just enough Unicode mappings to get it to work on another Hebrew wordprocessor Mellel which runs on Mac OSX.

Mike

Mike Thompson · Post by **Mike Thompson** » Tue Sep 21, 2004 11:09 pm

Yehuda,
Thanks for your explanation of the oddities of my font's mapping.

A few days ago, I had the chance of downloading my fonts onto a laptop used by an Israeli. It had a Windows operating system and Hebrew/Latin keyboard. I opened Microsoft Word and used it to type a simple New Years greeting in Hebrew. This worked fine. Do you have any idea why it worked? What mapping would it have used?

Sadly it did not understand my font family and generated a bold and an ugly italic from my Regular. Again, any idea why?

Mike

Dick Pape · Post by **Dick Pape** » Wed Sep 22, 2004 3:55 am

Hello Mike -- not sure I should make any more comments, but I have Yehuda keeping me honest!

I opened Microsoft Word and used it to type a simple New Years greeting in Hebrew. This worked fine. Do you have any idea why it worked?

!?! Jeepers Mike -- you worked hard for it to work. And it did! You did good!! Your glyphs and the keyboard mapping worked well together.

As far as your "bold and ugly italic" I found when loading all three of your fonts into WinXP fonts (via FCP), the regular and bold versions worked correctly in MS Word, Excel and Powerpoint. The Italic didn't show up at all although it was in the Fonts folder like the other two.

I created a document with Mike Regular which was correctly displayed. I used Word's Bold option and it displayed your Mike Bold font. When selecting Word's Italic option it took the regular version and applied some algorythm to oblique the characters. When I asked for a Bold Italic, it took your Mike Bold and slanted it to the right.

I autonamed your Italic in FCP and reloaded but that didn't change the outcome. Needs something more - like someone with more knowledge. But you're close.

On the other hand, if you wish, you could let Microsoft create your Bold Italic. Might not be pretty, but it's done.

Dick

Yehuda · Post by **Yehuda** » Wed Sep 22, 2004 6:11 pm

A few days ago, I had the chance of downloading my fonts onto a laptop used by an Israeli. It had a Windows operating system and Hebrew/Latin keyboard. I opened Microsoft Word and used it to type a simple New Years greeting in Hebrew. This worked fine. Do you have any idea why it worked? What mapping would it have used?

It's using the Unicode mapping that you added. (BTW, you should add the Unicode mapping for the shekel symbol and for other Hebrew characters.)

I have no wisdom to offer on your Word problems.

Erwin Denissen · Post by **Erwin Denissen** » Wed Sep 22, 2004 7:14 pm

I've tried a lot of different things to get your italic font to work, but unfortunately to no avail.

It seemed to work when I only installed the regular and the italic version, but the italic was replaced by an "italic imitation" of the regular font as soon as I installed the bold version.

Maybe a reboot helps. Let us know your results.

Mike Thompson · Post by **Mike Thompson** » Thu Sep 23, 2004 3:02 am

Thanks to all those who are helping me with this!

I'm looking at Times New Roman because it is possible to distinguish by eye if a word processor really is using the true italic, true bold etc.
The top of k and j can be used to distinguish true bold from fat regular.
The z can distinguish a true bold-italic ( a fun glyph indeed!). The a and y can distinguish true italic.

So far I haven't discovered why it works with MS Word but Mike Hebrew doesn't.

By the way I see that all members of the Times New Roman family have Hebrew unicode characters (and lots else besides),
Mike

Font Forum

Mike Hebrew V1.01

Mike Hebrew V1.01

Any idea why it works?