by Michael S. Kaplan, published on 2013/10/11 07:01 -04:00, original URI: http://blogs.msdn.com/b/michkap/archive/2013/10/11/10455822.aspx
Calling all keyboard authors!
Are you excited about the possibilities behind chained dead keys?
Did the Cherokee Phonetic keyboard layout inspire you?
I want to hear from you!
Now I know how to fill in spaces around your layout to bulletproof your work.
That is the part I'm good at.
But I need to know your scenario, your language.
Are you trying to build an intuitive Vietnamese layout that properly handles diacritics and tone marks?
Or is the plan for a keyboard to handle all of the modern Hangul, built up from the Jamo on each key you type?
Perhaps your aspiration is to handle Old Hangul, intuitively building up Old Hangul from all the underlying Jamo?
Or is properly building up Arabic or Cyrillic the goal?
You may be thinking about the many Indics that we have wrong in every version we have ever shipped since Windows 2000.
Or maybe Amharic or Tigrigna? They deserve better, also...
Maybe your angle is completely different from any of these things, using some other language entirely!
I want to hear from you, talk to you.
In some future version, this has got to be easier to do. Way easier.
To make that happen, I need to know what you think you want to author, what you think you want to build.
Are you ready to help make MSKLC great? Because that is what my goal is here.
I need to start with you to make it happen....
If you are one of the group that is interested, leave a comment telling me what language and/or keyboard has you inspired and we will go from there...
John Cowan on 11 Oct 2013 12:15 PM:
Are you kidding? I'd love to have this! I just introduced a whole mess of unlovely hacks to my US Moby Latin keyboard in version 9 <www.ccil.org/.../MobyLatinKeyboard.zip> in order to handle double-accented Vietnamese vowels. I'd much rather type "AltGr+^ AltGr+J a" to generate ậ (a with circumflex and dot below) rather than the current kludge of "AltGr+J 2".
Doug Ewell on 11 Oct 2013 2:53 PM:
What John said. I'm a convert to his keyboard, but I'll keep experimenting with some of the scenarios you mentioned. Vietnamese is the poster child, but there are plenty of others.
Michael S. Kaplan on 12 Oct 2013 4:21 AM:
I know that there were a lot more people who were excited about chained dead keys that haven't volunteered their thoughts, their plans? Why is everybody being so shy all of a sudden?
Michael S. Kaplan on 12 Oct 2013 6:17 PM:
So @John and @Doug - do I hear people offering to beta test it when it's ready?
John Cowan on 13 Oct 2013 8:11 PM:
Definitely. That way I can remove the kludges and add lots more characters.
Andrew_Cunningham on 13 Oct 2013 8:55 PM:
I can think of a range of uses for chained dead keys in a number of east and West African languages, some SE Asian languages, esp Vietnamese ethnic languages. Quite a few of my keyboard layouts could be optimised with dead key chaining. Do you want a comprehensive list?
A.
George on 14 Oct 2013 5:58 AM:
I am currently designing some keyboards for easier scientific typesetting and international languages (e.g. easy typesetting of all European languages using the Latin alphabet). Chained deadkeys would certainly be interesting, as would greater support for the nonbasic-plane Unicode characters.
Doug Ewell on 14 Oct 2013 8:54 AM:
Beta-test a new release of MSKLC? Hellll yeah.
Doug Ewell on 14 Oct 2013 1:37 PM:
George: "as would greater support for the nonbasic-plane Unicode characters." I'm afraid the inability to map a dead-key combination to a non-BMP code point is a Windows limitation, not something that can be fixed in MSKLC. A pity.
Michael S. Kaplan on 14 Oct 2013 4:05 PM:
No, it is an MSKLC limitation, a bug I would like to fix very soon!
Andrew_Cunningham on 15 Oct 2013 4:59 PM:
I'm sure there are a few of us readily willing to beta test for you
Doug Ewell on 16 Oct 2013 11:14 AM:
Michael, I could have sworn you once said the non-BMP dead-key thing was a Windows problem. All the better if it's an MSKLC problem after all, since now there's hope it'll be fixed. (Who will do the first hieroglyohic or pan-emoji keyboard?)
Doug Ewell on 16 Oct 2013 11:21 AM:
No, wait, I'm not nuts. See your blog "Dead keys are not intuitive" from 2004-12-17:
"Also, you can only have each pairing of keystrokes produce a single UTF-16 code point. [...] This is not an MSKLC limitation, it is a core limitation in the keyboard layout architecture [...]"
Michael S. Kaplan on 16 Oct 2013 1:48 PM:
I was wrong. It *is* an MSKLC limitation...
referenced by
2013/11/07 MSKLC wonderings