Here is a Debian bug report (#432297) reported by Andreas Metzler. From: Andreas Metzler <ametzler@downhill.at.eu.org> To: Debian Bug Tracking System <submit@bugs.debian.org> Subject: xkb-data: key <AE12> generates <acute> with de+nodeadkey on PC keyboard (as in #241127) Date: Mon, 9 Jul 2007 11:03:12 +0200 Package: xkb-data Version: 0.9-4 Severity: normal Hello, after ugrading to etch I have refound #241127 (now in xkb-data instead of XFree86). With this setup: Section "InputDevice" Identifier "Generic Keyboard" Driver "kbd" Option "CoreKeyboard" Option "XkbRules" "xorg" Option "XkbModel" "pc105" Option "XkbLayout" "de" Option "XkbVariant" "nodeadkeys" Option "XkbOptions" "compose:menu" EndSection The key <AE12> (between "scharfes s" and backspace on a german keyboard) produces a acute accent over space instead of apostrophe/singlequote. Please fix this to produce an apostrophe. Patch is simple afaiui: --- /usr/share/X11/xkb/symbols/de.bak 2006-10-20 22:26:15.000000000 +0200 +++ /usr/share/X11/xkb/symbols/de 2007-07-09 10:07:28.000000000 +0200 @@ -43,7 +43,7 @@ name[Group1]="Germany - Eliminate dead keys"; key <TLDE> { [asciicircum, degree, notsign, notsign ] }; - key <AE12> { [ acute, grave, cedilla, cedilla ] }; + key <AE12> { [apostrophe, grave, cedilla, cedilla ] }; key <AD11> { [udiaeresis, Udiaeresis, diaeresis, diaeresis ] }; key <AD12> { [ plus, asterisk, asciitilde, macron ] }; key <AC10> { [odiaeresis, Odiaeresis, doubleacute, doubleacute ] }; A single acute has no use, it is afaik always the wrong character both in German and English language. Most people confuse it with the apostroph and will accidentally type stuff like can´t instead of can't. Apostrophe otoh is a frequently used character, especially as single-quote. This is also inconsistent with linux-console (de-latin1-nodeadkeys.kmap.gz)
Created attachment 10640 [details] [review] Replaces acute with apostrophe
Thanks, committed
Hi there, I noticed this change because of investigations concerning this wine bug: http://bugs.winehq.org/show_bug.cgi?id=11122 el pointed out in this bug that the issue was tied to the keyboard layout and also presented two "fixes" for it. I'm reopening this bug here, because at least from my point of view the fix done here (changing acute to apostrophe) is wrong. And therefore the tables in wine's keyboard.c are correct. Three reasons why the fix is wrong: 1) The normal german keyboard layout has the acute/grave key next to the backspace. The apostrophe is located next to the ENTER key (SHIFT + #). With the current config in /usr/share/X11/xkb/symbols/de the apostrophe is on TWO keys of the keyboard and the acute on NONE. It really doesn't make sense to me to have two keys generate the same character. Especially since it's not possible anymore to generate the acute accent. 2) The labeling of the keyboard clearly indicates the the acute/grave key does/should produce a acute/grave accent symbol. See the wine bugreport for a camshot of my german laptop keyboard. Clearly the key next to backspace shows the acute accent on the lower part and the grave accent on the upper part. And also very clearly: The upper symbol on the #-key next to the ENTER key is an apostrophe. Mapping should be based what the use sees on his keyboard, that's currently not the case. 3) The fix made nodeadkeys and deadkeys layout inconsistent: With deadkeys the mapping is like that (from /usr/share/X11/xkb/symbols/de) key <AE12> { [dead_acute, dead_grave, dead_cedilla, dead_ogonek ] }; With nodeadkeys the mapping is like that: key <AE12> { [apostrophe, grave, cedilla, cedilla ] }; If the key produces an A with acute in deadkeys mode, then it should produce a single acute in nodeadkeys mode. Andreas Metzler's point that the creation of a single acute accent has no use isn't valid in my opinion. Again the mapping should be based on how the keyboard labeling is done. Of course this should include a change to the linux console nodeadkeys layout, since the keys are also mapped differently there. What is the opinion of you people? Greets, Tobias
Mohammed, would you please comment on that?
I've contacted Andreas to get him react here.
Tobias Jakobi wrote: > Andreas Metzler's point that the creation of a single acute accent has no use > isn't valid in my opinion. Again the mapping should be based on how the > keyboard labeling is done. -------------- Could you elaborate why this shouldn't be a valid point? I really think it is infinitely preferable to have a very useful key at two locations (including a easy one without shift-key) instead of blocking a prime location (non-shifted!) with a useless key. Also as previously noted this is long standing practice. I am not aware of a situation where acute over space is useful in either German language or a computing context (The latter being the main target for no-deadkeys layouts). Perhaps <compose>-key <AE12>-<space> should produce acute as a trade-off? Thanks, cu andreas
Hi Andreas! (In reply to comment #6) > Could you elaborate why this shouldn't be a valid point? I don't think that it's valid because it's based on your opinion about what keys/characters are useful or not. From my point of view the generated characters should resemble those printed/labeled on the keyboard. With your reasoning I could e.g. request that the ², ³, ° mappings are changed, because I'm not using them at all - and therefore are not useful (for me) with the current mapping. > I really think it is > infinitely preferable to have a very useful key at two locations (including a > easy one without shift-key) instead of blocking a prime location (non-shifted!) > with a useless key. I'm against the idea to map one character to multiple keys, no matter how useful the key might be. I also doubt that the apostrophe is a useful char in german. See: http://de.wikipedia.org/wiki/Apostrophitis It might be useful in english, but we're talking about a _german_ keyboard layout here. The apostrophe itself is rarely used in german texts. > Also as previously noted this is long standing practice. I don't see where you have stated that? > I am not aware of a situation where acute over space is useful in either German > language or a computing context (The latter being the main target for > no-deadkeys layouts). Nor am I aware of many situations where the apostrophe is useful in a german text. Quoting is usually done with the normal (double) quotation marks ("). I think your point can be reduced to: Mapping another apostrophe to the acute makes sense when typing english text on a german keyboard and when doing coding/computing/latexing work. > > Perhaps <compose>-key <AE12>-<space> should produce acute as a trade-off? > > Thanks, cu andreas > I want to emphasize that I'm not really interested in another (different?) change of the mapping. I just encountered this bugreport because of the mapping problems I found with wine. I just want to know if this decision is final. We already screwed this one up by the last change of the mapping. I have submitted a patch to wine that covers both mapping variants, so wine continues to work with both older and newer versions of the xkeyboard-config / xkb-data package. Greets, Tobias
(In reply to comment #7) > (In reply to comment #6) >> Could you elaborate why this shouldn't be a valid point? > I don't think that it's valid because it's based on your opinion about what > keys/characters are useful or not. > From my point of view the generated characters should resemble those > printed/labeled on the keyboard. With your reasoning I could e.g. request that > the ², ³, ° mappings are changed, because I'm not using them at all - and > therefore are not useful (for me) with the current mapping. Hello, I think the difference to my reasoning is that although ², ³, ° or § are not used frequently by everybody there are actually useful applications. (I even personally have used all of them ;-) Contrary to that I claim that acute over space has no use (except as a bad example: "This is an acute - do not use that, you want an apostrophe.") in either regular German or even English. >> I really think it is infinitely preferable to have a very useful >> key at two locations (including a easy one without shift-key) >> instead of blocking a prime location (non-shifted!) with a useless >> key. > I'm against the idea to map one character to multiple keys, no > matter how useful the key might be. I obviously do not feel that strongly a this. :-) > I also doubt that the apostrophe is a useful char in german. > See: http://de.wikipedia.org/wiki/Apostrophitis It is a useful character in German, although used *a* *lot* less than in English of course. (e.g "Andreas' Jacke". See http://de.wikipedia.org/wiki/Apostroph for more examples.) > It might be useful in english, but we're talking about a _german_ keyboard > layout here. The apostrophe itself is rarely used in german texts. [...] It is rarely used in German texts, but Acute is never used. And apostrophe is used *a* *lot* in computing context, it is a very frequently used key ("single quote") if you regulary do shellscripting or programming. >> Also as previously noted this is long standing practice. > I don't see where you have stated that? I am sorry I did not, however I submitted the bug in July 07 because the layout changed. - Previously (XFree86) I had apostroph, then suddenly Acute. [...] > I think your point can be reduced to: Mapping another apostrophe to > the acute makes sense when typing english text on a german keyboard > and when doing coding/computing/latexing work. That is part two of my reasoning, part one is "Acute is useless". I do think there should be some way to produce Acute even with no deadkeys, how about compose-space-AE12? cu andreas
My patch is now in wine git master and does properly detect both "versions" of the keyboard layout mapping. For me this issue is fixed :)
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.