[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: efficiency of windows managers



On Thu, Sep 27, 2007 at 07:25:48PM -0500, Ron Johnson wrote:
> On 09/27/07 18:58, Amit Uttamchandani wrote:
> >> I have to say, the screenshot has me intrigued.  It's got a 
> >> pleasingly geeky sort of retro text GUI look, like something
> >> Xerox PARC might have come up with in the 1970s. ;)
> >> 
> > 
> > It is very efficient. Especially tiling modes. I can't use any
> > other Windowmanager now. I just can't. With DWM, you're focused
> > on the application and the task at hand. Try it out.
> 
> With other WMs, you're not focused on the app and task?

I've not used dwm but I assume its similar to wmii being written by
the same guys. 

I agree with Amit. With other managers one spends a lot of time dealing
with manipulating the windows themselves -- maximizing, resizing,
shuffling them so they fit the way you want on the screen. I
personally find it particularly distracting if working on things
side-by-side: manipulating check registers while reading banking
websites; editing html while refreshing the page in a browser and so
forth. There are other instances where I find it distracting as
well. 

With a keyboard oriented tiling WM a few things happen. 

1) the initial placement and sizing of windows is completely out of
   your control (you can hack scripts to get things to default the way
   you want, but that is doable in just about any wm). You'll say that
   this is already the case, and its true, but the default is pretty
   sensibly different. Instead of splashing the new window up on the
   screen willy-nilly, the new window gets its own frame and its own
   piece of screen real-estate without interference from the
   others. Unless you specifically tell the wm that you want to stack
   windows on top of each other (that is, perpendicular to the plane
   of the screen), then it will *not* do that. So, for example, if I'm
   working in an x-term and I want another one, when it opens, the
   first one gets resized to half the screen and the new one gets the
   other half. Open another, and they each get squeezed into 1/3,
   etc. Its a little wierd at first, but after practice, it really
   begins to shine. 

2) Manipulating windows is done from the keyboard. This is a huge
time/energy saver if you already are keyboard oriented. You never have
to leave the keyboard to get things organised the way you like. Its
similar to using screen. You toggle between windows with keystrokes,
resize windows with keystrokes, split the frame with keystrokes
etc. So that's all good, but the real beauty of this is that the WM
actually makes the decision about how to interpret your commands
(configurable, of course). When you tell it to move a window one
direction or another, it just does it and maximizes everything as much
as possible. There is *no* grabbing of frames, or toggle maximize or
anything like that. 

3) this part answers your question... once you get used to it and get
it configured as you like it, your work method changes. (at least mine
does). I only ever have on the screen the things I'm currently working
on. I spend essentially no time manipulating windows as I've
configured it to behave the way I want in terms of window placement,
column and row size etc. I switch between tasks with ease without the
mouse. In short, I spend more, better focused, time on the work I'm
doing instead of the overhead of controlling the work environment. 

I know that many WMs allow keyboard control, but something like wmii,
and I assume dwm, do it so much more simply and elegantly, it really
is amazing. 

I will admit that they don't work so well with heavily mouse oriented
tasks (gimp for example), but for just about anything else, its
great. IMO. 

A

Attachment: signature.asc
Description: Digital signature


Reply to: