[Inkscape-devel] Window Manager Modifiers

[Carsten Haitzler (The Rasterman)]

On Sat, 4 Aug 2007 18:51:10 +0200 jiho <jo.irisson at gmail.com> babbled:

> Hello everyone,
> 
> First of all, it was not my intention to start a flame war or  
> anything. It is just my personal opinion that what lacks the more  
> cruelly to Linux desktop currently is consistency and keybindings is  
> one particularly clear example. I hoped that mentioning it on this  
> list, where the problem is regularly brought up, would get the ball  
> rolling to the correct people (be they freedesktop or someone else).  
> Anyway, I happily use Mac OS X (where keybindings are quite  
> consistent and can be edited easily anyway) and won't "fight" for  
> linux to get better... I just hope it will, because I like the  
> development model more ;-)

here's the problem. in windows or OSX - you have no CHOICE. you get your "wm"
with the bindings it comes with - and that's it. sure - some of these bindings
can be changed - and that is no different to x11 and window managers -the
difference being there is not just 1 wm. there are many - with many many many
different features, ideas and design goals. they will all probably have
different needs for bindings based on these features. asking every wm to NOT
use some modifier or to all use exactly the same isn't going to happen - just
like you won't get OSX and Windows to use the EXACT same modifiers. with key
bindings it is possible to find if it's been stolen (just try to grab it
yourself - if the grab fails, it's taken - then ungrab again as you only want
to use it as a normal key event within your focused window). button bindings
can't be easily detected (in theory they can - walk the window tree to every
parent window of yours until you hit root, do this after a reparent, and strut
grab a button/modifier combo, then ungrab, and if at any level this fails -
someone else has stolen it).

> That said, here are my two more cents (or maybe three...)
> 
> Some quick math first: suppressing a key from the three available to  
> applications does not suppresses only a third of the possibilities,  
> indeed those things are multiplicative. With 2 modifiers and 10 keys  
> on a keyboard you have

i know. :)

> 	2 * 10 + 1 * 10 = 30 possibilites
> (2 modifs, 1 combination). With three you now have
> 	3 * 10 + 3 * 10 + 1 * 10 = 70 possibilites
> (3 modifs, 3 two-way combinations, 1 three-way combination).
> Now add the mouse and things are getting worse.
> So the decision here is important as it severely impacts the  
> applications.

and has impacted them for a long time. this is not a new issue. the important
thing is that applications have to live with it. as above it is possible to
detect these things (it's ugly - but would work). you need to just keep clear
of commonly "stolen" button/key bindings. you need a way to be able to
re-configure them for your application, the same way wm's provide ways to
change them (or most do). the wm authors have mostly done their bit to
accommodate apps by allowing configuration, but it is rare apps do the reverse.
to my mind that is not playing a "good citizen".

> On 2007-August-04  , at 05:31 , Ted Gould wrote:
> > On Sat, 2007-08-04 at 12:02 +0900, Carsten Haitzler wrote:
> >> well... this is no new problem. it's been around for years. in the  
> >> end - the
> >> wm's win, because you need one for a sane desktop. whatever they  
> >> steal in terms
> >> of key (and mouse) bindings is what a user comes to EXPECT from  
> >> functionality
> >> on ALL their windows (eg - alt+left mouse to move, alt+right mouse  
> >> for wm
> >> window menu etc.). there is a kind of pseudo-standard (just by  
> >> virtue that a
> >> lot of wm's come this way by default) that there seems to be a  
> >> common convention
> >> that alt + button is what wm's take - anything else if open for apps.
> >
> > I guess my real concern is that it's alt today, meta tomorrow and
> > control next week?  It seems like in general that there should be some
> > consistency here.  That way someone knows where their mouse events are
> > going (though I doubt they'll think about it that way).
> 
> If I recall correctly (and I am quite sure I do) Gnome also uses CTRL  
> in some keybindings so this can indeed be an additional problems (see  
> the quick computation above)
> 
> >> now i ask - why do u think it's bad apps use met - but you now  
> >> promote wm's to
> >> use it instead of alt? you are basically saying "i don't want to  
> >> use meta - so i
> >> want to force wm's to do the thing i am not willing to". at the  
> >> end of the day
> >> - this is highly unfair as your point about the meta key being  
> >> "not always
> >> there" is valid - and all you do is put the problem onto the wm's.
> >
> > No, the reason that I was suggesting that a WM use meta is because in
> > general, I think the desktop knows a lot more about the hardware that
> > it's working with.  Application do, and should, know very little.
> > Things like accessibility and keyboard configuration are done at a
> > desktop/WM level -- so I thought it would make more sense to also use,
> > or not use, the meta key at that level.  I believe that there are also
> > internationalization issues here, though I'm ignorant to other  
> > keyboard
> > layouts myself.
> 
> Let's assume everyone agrees that WM/Desktop and Applications  
> keybindings must be separate (I think that's quite safe). Now the  
> question is which modifier for which, and the choice is between ALT  
> and META. Here are a few reasons why I think ALT is better for  
> applications and META for WM:
> - keyboard proximity. CTRL is *the* application level shortcut and  
> ALT is closer to CTRL than META on all keyboards I've seen, so CTRL 
> +ALT combinations for applications seems more natural than CTRL+META
> - WM keyboard shortcuts they are kind of special since they apply  
> across the board and to me META carries this idea of "special".
> - META "uncertainty". as some pointed out, META isn't always  
> available (though I think it must be a very small minority of cases,  
> since I've seen some pretty exotic Sun and IBM keyboards and there's  
> always a META key somewhere, sometimes where CAPS LOCK usually is) or  
> may change "appearance" (it is the WIN key on windows PC keyboards,  
> the apple key on macs, a real META key on some others etc.). So it  
> may be less accessible.
> 	. as Ted just pointed out, it is easier to fix its absence (ie.  
> remap some other key) at the desktop level than at the application level
> 	. the annoyance it creates when it changes position or form is  
> restricted to one area (windows manipulations) instead of applying to  
> several keybindings in several apps. it is, in my opinion, easier to  
> re-learn.
> 	. even if it is completely absent, or difficult to reach, most  
> window managing actions can be performed with mouse clicks and drags,  
> on windows or in menus. In fact, most people do not actually use  
> window manager keyboard shortcuts. This is not the case for many  
> places where ALT is currently used in Inkscape (or in Gimp for that  
> matter). Some important things are simply *not possible* if you don't  
> have a second independent modifier in Inkscape or Gimp. And this is  
> not a problem of design in either apps, it's just that there are so  
> many things to do that you need many modifiers (I tried to report  
> everything that can be done with the selector tool in Inkscape and  
> its associated modifiers and this is already very complex and nearly  
> saturated).
> I am not saying that WM/Desktop keyboard shortcuts are useless, I am  
> just saying that they are probably  less numerous than what a large  
> app would need so it is less pain if they are difficult to reach or  
> need to be re-learned.
> 
> So there is history carrying over its early habits and there is what  
> some "unifying" projects like FreeDesktop can specify, and maybe  
> change, in the name of better overall usability. I think this issue  
> is typical of the second case and I hope my examples have convinced  
> some about this.
> 
> Cheers,
> 
> JiHO
> ---
> http://jo.irisson.free.fr/
> 
> 


-- 
------------- Codito, ergo sum - "I code, therefore I am" --------------
The Rasterman (Carsten Haitzler)    raster at rasterman.com
裸好多
Tokyo, Japan (東京 日本)