I've got Kindle 1.0.1 instead of 1.0.4

[FourOhFour]

I got 1.0.1 just now. Waited a few minutes and the update your kindle menu item became available again. It's updating again now.

Maybe 1.0.1 is a fix to the updating code necessary to get 1.0.4?

[rationalbiker]

Maybe this thread could be retitled?

1.0.1 to 1.0.4 to Star Trek Love

[TallMomof2]

1.0.4 was installed yesterday. All is well.

[exreader]

1.0.1 installed just a bit ago. no new manual. while hustling for a camera for the funnyyupdate screen, it rebooted. rats.

then put it to sleep, and it did it again, 1.0.4 downloaded, got some okay pix. nothing stunning, but the messages are neat, and the strange icons down the right side of the screen are intriguing - hidden mode?

exreader

[jimad]

You can find a Unicode Glyphs test file for Kindle at http://www.freekindlebooks.org under "Unicode". Simple answer is that Kindle doesn't seem to support much other than a few DOS dingbats above U+0250 "Latin" and even then Kindle leaves off the diacriticals on many of the high latin letters. I have Kindle version 1.0 -- does anyone know how the versions have changed and why?

[DaleDe]

Originally Posted by jimad

You can find a Unicode Glyphs test file for Kindle at http://www.freekindlebooks.org under "Unicode". Simple answer is that Kindle doesn't seem to support much other than a few DOS dingbats above U+0250 "Latin" and even then Kindle leaves off the diacriticals on many of the high latin letters. I have Kindle version 1.0 -- does anyone know how the versions have changed and why?

Thanks for the link. I will update the 'special characters' in the MR wiki based on this data. There are discussions of all the Unicode and other formats beginning with a discussion of 'characters' on the wiki. There is also a discussion of 'fonts'. Hopefully this will help in your quest.

Dale

[DaleDe]

I tried you test on a PC running Mobipocket Reader and found that it supported practically everything. I do not have a Kindle to test this with. Can you capture some pages and display what you see on the device? On the PC I would expect it would depend on the fonts you have loaded.

Is it true that 7E looks like an ndash rather than a ~?
Did you intend to leave out U+00AD from you list?

On a PC I get Greek, Armenian, Coptic, Hebrew, some but not all Arabic, Syriac, Thaana and NKo

All of the Canadian and beyond are poorly formatted.

Beginning at U+2580 the graphics characters are spotty but it is likely the font set.

Dale

[wallcraft]

Originally Posted by DaleDe

Can you capture some pages and display what you see on the device?

I enclose screenshots from a Kindle (smallest font size). The first is the first page, the 2nd contains U+0250 and the rest contain the only characters defined above this page. I have version 1.0.4.

[wallcraft]

With FBReader which Unicode Glyphs get displayed depends on the selected font. I assume the same would be true on a Cybook Gen3.

On the iLiad, Greek looks ok and Cyrillic has only a few missing glyphs (e.g. U+04FA to FF), but coverage above Armenian (above U+0589) is hit and miss.

[DaleDe]

Originally Posted by wallcraft

I enclose screenshots from a Kindle (smallest font size). The first is the first page, the 2nd contains U+0250 and the rest contain the only characters defined above this page. I have version 1.0.4.

That is not very good coverage. I have updated the special character data on the wiki site to reflect your observations. In a few cases I see the wrong characters are sometimes displayed and I did not include those. The special character page uses decimal numbers while your page uses hex numbers.

Note that the really interesting glyphs from a eBook perspective are in the range of 200C hex to 2044 but your original sample did not include these. On my decimal page these are the ones from 8204-8260. They have things like curley quotes, emdashes, etc. You might want to try those ranges and see what you get.

Dale

[wallcraft]

Originally Posted by DaleDe

Note that the really interesting glyphs from a eBook perspective are in the range of 200C hex to 2044

Unfortunately, this range is in the part of the document that is corrupted. I enclose screenshots of all the remaining pages that seem to have defined special characters on them (some are named differently, and some display the wrong character).

[Nate the great]

I'm curious. Have you tried other versions of Mobipocket Reader? (That's what the Kindle is, basically.)

Do they exhibit the same mistakes?

[jimad]

What Glyphs Kindle supports or not is not a "mistake" it is a design decision. Now we may disagree with Amazon about their design decisions -- I would have been much happier if Kindle also had at least support for Greek and Hebrew. Re the test file you can find it at freekindlebooks.org under "Unicode" and if you open it in a recent version of Windows using Mobipocker Reader you will find that Windows implements virtually all the 60,000 or so Unicode Glyphs. So the fact that Kindle only implements about 250 of these is a Kindle design decision. Actually, I take back what I said before about this not being "a mistake". Kindle actually implements a bunch of the high latin Glyphs which are suppose to have diacriticals or marks by leaving off those diacriticals and marks. Now THAT *IS* a mistake! You can choose to implement or not implement a Glyph, but implementing it *incorrectly* -- now that is a very bad idea!

[Nate the great]

Originally Posted by jimad

What Glyphs Kindle supports or not is not a "mistake" it is a design decision. Now we may disagree with Amazon about their design decisions -- I would have been much happier if Kindle also had at least support for Greek and Hebrew. Re the test file you can find it at freekindlebooks.org under "Unicode" and if you open it in a recent version of Windows using Mobipocker Reader you will find that Windows implements virtually all the 60,000 or so Unicode Glyphs. So the fact that Kindle only implements about 250 of these is a Kindle design decision. Actually, I take back what I said before about this not being "a mistake". Kindle actually implements a bunch of the high latin Glyphs which are suppose to have diacriticals or marks by leaving off those diacriticals and marks. Now THAT *IS* a mistake! You can choose to implement or not implement a Glyph, but implementing it *incorrectly* -- now that is a very bad idea!

Thanks. I didn't know that.

[wallcraft]

Originally Posted by jimad

What Glyphs Kindle supports or not is not a "mistake" it is a design decision.

Another way of saying this is that the Kindle only supports the Caecilia font, for example here is this font's General Punctuation coverage.