Discussion:
about descchar binding
(too old to reply)
高生旺
2009-01-31 05:06:05 UTC
Permalink
Hi

If I bind descchar to keypad, then how to define the character possition?
I can use chrlt/chrrt to define cut point, but brltty doesn't know this
for descchar.
Dave Mielke
2009-01-31 05:27:27 UTC
Permalink
[quoted lines by 高生旺 on 2009/01/31 at 13:06 +0800]
Post by 高生旺
If I bind descchar to keypad, then how to define the character possition?
It describes the character where the cursor is.

Note that descchar is already defined in keypad.ktb, as are lots of other
functions.
Post by 高生旺
I can use chrlt/chrrt to define cut point, but brltty doesn't know this
for descchar.
All of the character-based functions, including the four cut ones, use the
location of the cursor when invoked from the keyboard.
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 06:36:40 UTC
Permalink
Hi:

Did you change the program?
I really can perform cut/paste by chrlt/chrrt in prevous. But now it
can't.
Post by Dave Mielke
[quoted lines by 高生旺 on 2009/01/31 at 13:06 +0800]
Post by 高生旺
If I bind descchar to keypad, then how to define the character possition?
It describes the character where the cursor is.
Note that descchar is already defined in keypad.ktb, as are lots of other functions.
Post by 高生旺
I can use chrlt/chrrt to define cut point, but brltty doesn't know this >for descchar.
All of the character-based functions, including the four cut ones, use the location of the cursor when invoked from the keyboard.
Dave Mielke
2009-01-31 07:25:22 UTC
Permalink
[quoted lines by 高生旺 on 2009/01/31 at 14:36 +0800]
Post by 高生旺
Did you change the program?
I really can perform cut/paste by chrlt/chrrt in prevous. But now it
can't.
It's possible that older revisions allowed that via the position of the
leftmost character on the braille display. That was never the intent, though.
It's just that we were enhancing the whole capability without worrying about
smaller details at first. The intent, which is the way it works now, is to use
the position of the cursor.
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 07:31:34 UTC
Permalink
But under console or web page, how to move cursor to possition the
character?
Post by Dave Mielke
[quoted lines by 高生旺 on 2009/01/31 at 14:36 +0800]
Did you change the program?>I really can perform cut/paste by chrlt/chrrt in prevous. But now it >can't.
It's possible that older revisions allowed that via the position of the leftmost character on the braille display. That was never the intent, though. It's just that we were enhancing the whole capability without worrying about smaller details at first. The intent, which is the way it works now, is to use the position of the cursor.
Dave Mielke
2009-01-31 07:43:15 UTC
Permalink
[quoted lines by 高生旺 on 2009/01/31 at 15:31 +0800]
Post by 高生旺
But under console or web page, how to move cursor to possition the
character?
The plan is to add a virtual cursor which can go anywhere on the screen and
then to use it. That hasn't been done yet.
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 07:52:13 UTC
Permalink
For Chinese, may I add more information to descchar?
Post by Dave Mielke
[quoted lines by 高生旺 on 2009/01/31 at 15:31 +0800]
But under console or web page, how to move cursor to possition the >character?
The plan is to add a virtual cursor which can go anywhere on the screen and then to use it. That hasn't been done yet.
Dave Mielke
2009-01-31 08:02:24 UTC
Permalink
[quoted lines by 高生旺 on 2009/01/31 at 15:52 +0800]
Post by 高生旺
For Chinese, may I add more information to descchar?
Right now we use the description associated with the character as supplied by
the Unicode database. It seems to be rather generic for the ideographic
characters, and, therefore, doesn't seem to be very helpful. For the benefit of
other readers, the Unicode character 9000, for example, has, for its
description, the rather useless phrase "CJK UNIFIED IDEDOGRAPH-9000".

Does each of those characters have a specific meaning, or can a character have
one meaning in one language and another meaning in another language?

Do you have a list of what each of those characters actually means?
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 08:29:06 UTC
Permalink
Right now we use the description associated with the character as supplied by the Unicode database. It seems to be rather generic for the ideographic characters, and, therefore, doesn't seem to be very helpful. For the benefit of other readers, the Unicode character 9000, for example, has, for its description, the rather useless phrase "CJK UNIFIED IDEDOGRAPH-9000".
Does each of those characters have a specific meaning, or can a character have one meaning in one language and another meaning in another language?
Because in Chinese braille, some characters represented as same braille
pattern. zh-tw-ucb is only my private experiment. The common solution is
give each character a meaningful description.
Do you have a list of what each of those characters
actually means?
I'm creating it.
Samuel Thibault
2009-01-31 12:18:47 UTC
Permalink
Hello,
Post by Dave Mielke
[quoted lines by 高生旺 on 2009/01/31 at 15:52 +0800]
Post by 高生旺
For Chinese, may I add more information to descchar?
Right now we use the description associated with the character as supplied by
the Unicode database. It seems to be rather generic for the ideographic
characters, and, therefore, doesn't seem to be very helpful. For the benefit of
other readers, the Unicode character 9000, for example, has, for its
description, the rather useless phrase "CJK UNIFIED IDEDOGRAPH-9000".
Does each of those characters have a specific meaning, or can a character have
one meaning in one language and another meaning in another language?
Do you have a list of what each of those characters actually means?
That list already exists: there is a more descriptive field in Unicode,
for instance for U+9000 it says:

`Definition in English: step back, retreat, withdraw
Mandarin Pronunciation: TUI4
Cantonese Pronunciation: teoi3
Japanese On Pronunciation: TAI TON
Japanese Kun Pronunciation: SHIRIZOKU SHIRIZOKERU
Tang Pronunciation: *tuə̀i
Korean Pronunciation: THOY'

Samuel
Samuel Thibault
2009-01-31 12:49:19 UTC
Permalink
Post by Samuel Thibault
Post by Dave Mielke
[quoted lines by 高生旺 on 2009/01/31 at 15:52 +0800]
Post by 高生旺
For Chinese, may I add more information to descchar?
Right now we use the description associated with the character as supplied by
the Unicode database. It seems to be rather generic for the ideographic
characters, and, therefore, doesn't seem to be very helpful. For the benefit of
other readers, the Unicode character 9000, for example, has, for its
description, the rather useless phrase "CJK UNIFIED IDEDOGRAPH-9000".
Does each of those characters have a specific meaning, or can a character have
one meaning in one language and another meaning in another language?
Do you have a list of what each of those characters actually means?
That list already exists: there is a more descriptive field in Unicode,
`Definition in English: step back, retreat, withdraw
Mandarin Pronunciation: TUI4
Cantonese Pronunciation: teoi3
Japanese On Pronunciation: TAI TON
Japanese Kun Pronunciation: SHIRIZOKU SHIRIZOKERU
Tang Pronunciation: *tuə̀i
Korean Pronunciation: THOY'
I haven't found how to get them from libicu yet, but from libgucharmap
these are returned by gucharmap_get_unicode_k{Definition, Cantonese,
Mandarin, Tang, Korean, JapaneseKun, JapaneseOn}.

This seems to be quite specific to CJK, I haven't found similar things
for other scripts. Maybe that could be an additional preference, to let
the user choose which language to replace CJK COMPATIBILITY IDEOGRAPH
with.

Samuel
高生旺
2009-01-31 12:50:52 UTC
Permalink
So interesting. Where is that list?
Post by Samuel Thibault
Hello,
Dave Mielke, le Sat 31 Jan 2009 03:02:24 -0500, a ?crit :> [quoted lines by 高生旺 on 2009/01/31 at 15:52 +0800]> > >For Chinese, may I add more information to descchar?> > Right now we use the description associated with the character as supplied by > the Unicode database. It seems to be rather generic for the ideographic > characters, and, therefore, doesn't seem to be very helpful. For the benefit of > other readers, the Unicode character 9000, for example, has, for its > description, the rather useless phrase "CJK UNIFIED IDEDOGRAPH-9000".> > Does each of those characters have a specific meaning, or can a character have > one meaning in one language and another meaning in another language?> > Do you have a list of what each of those characters actually means?
`Definition in English: step back, retreat, withdrawMandarin Pronunciation: TUI4Cantonese Pronunciation: teoi3Japanese On Pronunciation: TAI TONJapanese Kun Pronunciation: SHIRIZOKU SHIRIZOKERUTang Pronunciation: *tu??iKorean Pronunciation: THOY'
Samuel Thibault
2009-01-31 12:59:32 UTC
Permalink
Post by 高生旺
So interesting. Where is that list?
The raw data is UniHan:

http://www.unicode.org/Public/UNIDATA/Unihan.zip

But it'll probably be better to just have brltty use a unicode library
to fetch it.

Samuel
高生旺
2009-01-31 13:08:22 UTC
Permalink
It's interesting but still not enogh to solve my problem.
I still need to creating our own list and should be included in brltty for
getting exact meanings in Chinese braille.
高生旺, le Sat 31 Jan 2009 20:50:52 +0800, a ?crit :> So interesting. Where is that list?
http://www.unicode.org/Public/UNIDATA/Unihan.zip
But it'll probably be better to just have brltty use a unicode libraryto fetch it.
Samuel Thibault
2009-01-31 13:20:07 UTC
Permalink
Post by 高生旺
It's interesting but still not enogh to solve my problem.
Then it's a bug in unihan and we should report it.
Post by 高生旺
I still need to creating our own list and should be included in brltty for
getting exact meanings in Chinese braille.
I'm reluctant to follow such a way: that kind of information is surely
not specific to brltty.

Samuel
高生旺
2009-01-31 14:19:33 UTC
Permalink
Post by Samuel Thibault
Then it's a bug in unihan and we should report it.
No. It's fault of Chinese braille.
Our braille system is based on Pronunciation
In unihan, you may find that many characters have same Pronunciation in
kMandarin, such as 9000 and 99FE, 3703 and 3751 and 37CC and 37FF and 385B
......... and so on.
You may imagine how it make us trouble in reading.
Dave Mielke
2009-01-31 16:09:23 UTC
Permalink
[quoted lines by 高生旺 on 2009/01/31 at 22:19 +0800]
Post by 高生旺
No. It's fault of Chinese braille.
Our braille system is based on Pronunciation
In unihan, you may find that many characters have same Pronunciation in
kMandarin, such as 9000 and 99FE, 3703 and 3751 and 37CC and 37FF and 385B
......... and so on.
You may imagine how it make us trouble in reading.
Since the braille already shows the pronunciation, what if we just show the
English definition. That should clarify which word it is. Would that be
sufficient?

Another question for my curiosity: If the braille is the same for two
characters then wouldn't that also mean that the spoken pronunciation is the
same? If so, then wouldn't spoken Chinese have the same problem? How is that
resolved?

I know that Chinese shows the difference between words by the tone in which
they are spoken. Does the braille representation also show the tone?
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 16:44:27 UTC
Permalink
Since the braille already shows the pronunciation, what if we just show the English definition. That should clarify which word it is. Would that be sufficient?
For a few people that's sufficient, but for most others it isn't.
Another question for my curiosity: If the braille is the same for two characters then wouldn't that also mean that the spoken pronunciation is the same? If so, then wouldn't spoken Chinese have the same problem? How is that resolved?
We always understand them by context.
I know that Chinese shows the difference between words by the tone in which they are spoken. Does the braille representation also show the tone?
Yes.
Dave Mielke
2009-01-31 17:01:29 UTC
Permalink
[quoted lines by 高生旺 on 2009/02/01 at 00:44 +0800]
Post by 高生旺
Since the braille already shows the pronunciation, what if we just show the English definition. That should clarify which word it is. Would that be sufficient?
For a few people that's sufficient, but for most others it isn't.
What information should be displayed?
--
Dave Mielke | 2213 Fox Crescent | The Bible is the very Word of God.
Phone: 1-613-726-0014 | Ottawa, Ontario | 2011 May 21 is the Day of Judgement.
EMail: ***@mielke.cc | Canada K2A 1H7 | 2011 Oct 21 is the End of the World.
http://FamilyRadio.com/ | http://Mielke.cc/bible/
高生旺
2009-01-31 17:12:52 UTC
Permalink
Post by Dave Mielke
What information should be displayed?
At least a phrase contains 2 or 3 characters or more details.
How about let users add the definition list to the 4th column of ttb or
ctb? If no provision then automatic fetch the standard list.
Samuel Thibault
2009-01-31 18:17:33 UTC
Permalink
Post by 高生旺
Post by Samuel Thibault
Then it's a bug in unihan and we should report it.
No. It's fault of Chinese braille.
Our braille system is based on Pronunciation
There's a misunderstanding: I'm not actually talking about braille. We
are talking about descchar showing information about the unicode
character being displayed. I have said that unihan probably has useful
information, but I now realize that it only provides the english
meaning and mandarin/korean/japanese pronunciation, and you actually
already know the pronunciation since that's precisely what the braille
representation expresses. So Unihan is missing a piece of information
that you want: some text that describes the character itself, and that's
the bug which I'm talking about.

I wonder: of course it's not fine for chinese people who do not know
english, but wouldn't be the english description enough? (i.e. for
U+9000, "step back, retreat, withdraw", you can just grep kDefinition
Unihan.txt to get them all).

Samuel
高生旺
2009-02-01 03:52:04 UTC
Permalink
I wonder: of course it's not fine for chinese people who do not knowenglish, but wouldn't be the english description enough? (i.e. forU+9000, "step back, retreat, withdraw", you can just grep kDefinitionUnihan.txt to get them all).
For me, it's helpful and enough. But for my friends, they'll say: "why
don't say Chinese?".
I think that make two version of displayed information, standard and
localization (depand on braille table), is a beautiful solution.
Samuel Thibault
2009-02-01 03:56:50 UTC
Permalink
Post by 高生旺
I wonder: of course it's not fine for chinese people who do not knowenglish, but wouldn't be the english description enough? (i.e. forU+9000, "step back, retreat, withdraw", you can just grep kDefinitionUnihan.txt to get them all).
For me, it's helpful and enough. But for my friends, they'll say: "why
don't say Chinese?".
Ok so that's what I had inferred. Actually, all descriptions of unicode
characters should ideally be translated. I know that this is done at
least for french. I will discuss with unicode & libicu lists.

Samuel
Samuel Thibault
2009-02-02 23:51:02 UTC
Permalink
Hello,
Post by Samuel Thibault
Post by 高生旺
I wonder: of course it's not fine for chinese people who do not knowenglish, but wouldn't be the english description enough? (i.e. forU+9000, "step back, retreat, withdraw", you can just grep kDefinitionUnihan.txt to get them all).
For me, it's helpful and enough. But for my friends, they'll say: "why
don't say Chinese?".
Ok so that's what I had inferred. Actually, all descriptions of unicode
characters should ideally be translated. I know that this is done at
least for french. I will discuss with unicode & libicu lists.
The result of the discussion is that there is no standard translations
etc. People from ***@unicode.org recommended to just take existing
dictionaries (e.g. those references in unihan.txt) and plug them in.

Samuel
Samuel Thibault
2009-02-03 00:06:13 UTC
Permalink
Post by Samuel Thibault
Post by Samuel Thibault
Post by 高生旺
I wonder: of course it's not fine for chinese people who do not knowenglish, but wouldn't be the english description enough? (i.e. forU+9000, "step back, retreat, withdraw", you can just grep kDefinitionUnihan.txt to get them all).
For me, it's helpful and enough. But for my friends, they'll say: "why
don't say Chinese?".
Ok so that's what I had inferred. Actually, all descriptions of unicode
characters should ideally be translated. I know that this is done at
least for french. I will discuss with unicode & libicu lists.
The result of the discussion is that there is no standard translations
dictionaries (e.g. those references in unihan.txt) and plug them in.
It was also proposed to use examples using the character from word lists
like CEDICT, i.e. "<this kanji> as in <this word>".

Samuel
高生旺
2009-02-03 02:55:45 UTC
Permalink
I think that unihan.txt is too large. The ideal way is only fetch English
definitions for standard, then give users to put their local definitions
in the braille tables.
Samuel Thibault
2009-02-03 03:02:49 UTC
Permalink
Post by 高生旺
I think that unihan.txt is too large. The ideal way is only fetch English
definitions for standard,
Oh yes for sure. I never intended to include all of unihan, just what
we need :)

Samuel

Loading...