[PATCH 0/1] improvements to the lightweight desktop example

[PATCH 0/1] improvements to the lightweight desktop example

[Leo Famulari]

I think we should remove xmonad from the lightweight-desktop OS
declaration example. Xmonad is not "lightweight" in terms of disk space
or bandwidth, since it requires GHC, which is 859 MiB uncompressed.

Also, the i3 window manager is not very useful without dmenu and
i3status, so I think we should add them.

Leo Famulari (1):
  gnu: lightweight-desktop.tmpl: Remove xmonad and complete i3-wm.

 gnu/system/examples/lightweight-desktop.tmpl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 
2.11.1

[PATCH 1/1] gnu: lightweight-desktop.tmpl: Remove xmonad and complete i3-wm.

[Leo Famulari]

* gnu/system/examples/lightweight-desktop.tmpl (packages): Remove
xmonad. Add i3status and dmenu.
---
 gnu/system/examples/lightweight-desktop.tmpl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gnu/system/examples/lightweight-desktop.tmpl b/gnu/system/examples/lightweight-desktop.tmpl
index 131c43af7..39314c896 100644
--- a/gnu/system/examples/lightweight-desktop.tmpl
+++ b/gnu/system/examples/lightweight-desktop.tmpl
@@ -33,8 +33,8 @@
 
   ;; Add a bunch of window managers; we can choose one at
   ;; the log-in screen with F1.
-  (packages (cons* ratpoison i3-wm xmonad  ;window managers
-                   nss-certs               ;for HTTPS access
+  (packages (cons* ratpoison i3-wm i3status dmenu ;window managers
+                   nss-certs                      ;for HTTPS access
                    %base-packages))
 
   ;; Use the "desktop" services, which include the X11
-- 
2.11.1

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Mathieu Lirzin]

Hi,

Leo Famulari <leo@famulari.name> writes:

> I think we should remove xmonad from the lightweight-desktop OS
> declaration example. Xmonad is not "lightweight" in terms of disk space
> or bandwidth, since it requires GHC, which is 859 MiB uncompressed.

Makes sense.

> Also, the i3 window manager is not very useful without dmenu and
> i3status, so I think we should add them.

Agreed.

While we are on this lightweight template, what would you think of
replacing ratpoison with something less "exotic" like openbox? :)

> Leo Famulari (1):
>   gnu: lightweight-desktop.tmpl: Remove xmonad and complete i3-wm.
>
>  gnu/system/examples/lightweight-desktop.tmpl | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

-- 
Mathieu Lirzin
GPG: F2A3 8D7E EB2B 6640 5761  070D 0ADE E100 9460 4D37

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Clément Lassieur]

Mathieu Lirzin <mthl@gnu.org> writes:

> While we are on this lightweight template, what would you think of
> replacing ratpoison with something less "exotic" like openbox? :)

What do you mean by "exotic"?  Isn't that subjective? :)

Here are the results of "guix size":
  - ratpoison: 172.6 MiB,
  - i3-wm: 263.4 MiB,
  - openbox: 322.9 MiB.

So Openbox is almost twice as big as Ratpoison, which I think matters
for a "lightweight" template.  Furthermore, Ratpoison quite is
convenient to use for those of us who use GNU Emacs and GNU Screen.

Re: [PATCH 0/1] improvements to the lightweight desktop example

[John Darrington]

[-- Attachment #1: Type: text/plain, Size: 1028 bytes --]

On Wed, Feb 22, 2017 at 03:02:39PM +0100, Cl??ment Lassieur wrote:
     Mathieu Lirzin <mthl@gnu.org> writes:
     
     > While we are on this lightweight template, what would you think of
     > replacing ratpoison with something less "exotic" like openbox? :)
     
     What do you mean by "exotic"?  Isn't that subjective? :)
     
     Here are the results of "guix size":
       - ratpoison: 172.6 MiB,
       - i3-wm: 263.4 MiB,
       - openbox: 322.9 MiB.
     
     So Openbox is almost twice as big as Ratpoison, which I think matters
     for a "lightweight" template.  Furthermore, Ratpoison quite is
     convenient to use for those of us who use GNU Emacs and GNU Screen.


If you want a small, yet easy to use window manager there is always
twm: 104 MiB

J'

-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Mathieu Lirzin]

Hello Clément,

Clément Lassieur <clement@lassieur.org> writes:

> Mathieu Lirzin <mthl@gnu.org> writes:
>
>> While we are on this lightweight template, what would you think of
>> replacing ratpoison with something less "exotic" like openbox? :)
>
> What do you mean by "exotic"?  Isn't that subjective? :)

Sure it is!

What I meant is that Ratpoison is not the most "intuitive" WM for non
GNU Emacs/Screen users.  As a consequence adding it in an example
configuration which is likely to be copy and paste, is maybe not the
most welcoming thing.  :)

IMHO Openbox or anything which is able to launch a program by "clicking"
seems more friendly as a default (modulo the accessibility issues which
to my knowledge are not addressed by any of the "lightweight" WMs).

> Here are the results of "guix size":
>   - ratpoison: 172.6 MiB,
>   - i3-wm: 263.4 MiB,
>   - openbox: 322.9 MiB.
>
> So Openbox is almost twice as big as Ratpoison, which I think matters
> for a "lightweight" template.  Furthermore, Ratpoison quite is
> convenient to use for those of us who use GNU Emacs and GNU Screen.

As a former dedicated user of Ratpoison, I know and appreciate all its
niceness.

-- 
Mathieu Lirzin
GPG: F2A3 8D7E EB2B 6640 5761  070D 0ADE E100 9460 4D37

Re: [PATCH 0/1] improvements to the lightweight desktop example

[John Darrington]

[-- Attachment #1: Type: text/plain, Size: 3121 bytes --]

On Wed, Feb 22, 2017 at 04:48:55PM +0100, Mathieu Lirzin wrote:
     
     Sure it is!
     
     What I meant is that Ratpoison is not the most "intuitive" WM for non
     GNU Emacs/Screen users.  As a consequence adding it in an example
     configuration which is likely to be copy and paste, is maybe not the
     most welcoming thing.  :)
     
     IMHO Openbox or anything which is able to launch a program by "clicking"
     seems more friendly as a default (modulo the accessibility issues which
     to my knowledge are not addressed by any of the "lightweight" WMs).
     

I don't want to take sides in this debate.    At the same time I think "clicking
is more friendly" is a fallacy.  It's more friendly to people who are used to
it, but decidedly frightening to those who are not.

Some years ago, I was volunteering at an organisation which taught basic computer
use to the computer illiterate.  
A lot of the students I taught were: Elderly, female and had arthritis in the fingers
(but all that I dealt with were of sound mind).

Many of them had, in their younger years been employed as typists, so a keyboard
was nothing new to them and were quite happy with it.

The mouse on the other hand was a challenge:

1. I would start by standing over the student's shoulder and give them an 
exercise to alternately push the mouse away from their body and pull it back
towards them, whilst observing the curser ascending and descending the screen.
(Don't tell them to "move the mouse up"!  If you say that half of the students,
will lift the mouse from the surface of the desk!)

2. The second exercise would be an extension of this idea.  I would instruct
them now to move the mouse to the left, and observe the mouse cursor also move
to the left.  Almost invariably the student would first rotate the mouse 90 degrees to
the left and THEN push the mouse in that direction.  This instinct is very 
common with the ladies - It's unintuitive to them to have an object moving
in a direction other than the way it is facing. (Have you ever noticed how men
hold maps with North at the top, whereas women turn the map to that the top
is in the direction of travel?)

3. The next exercise would be to get them to click on a button.  Here 
the arthritis was sometimes an obstacle even if I had set the mouse sensitivity 
very low - but normally with a bit of effort the student could place the 
cursor over the target screen area.  So I would instruct them to press the
mouse button.  This was a often big problem.  The wrist would shake so much 
that when the click came the cursor had moved from the target.

So most of them found "clicking" extremely unfriendly.  Please be very 
carefull when making generalisations like "GUIs are intuitive" "mice are
friendly" etc.  As we said before - it depends uponn the user.


J'



-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Mathieu Lirzin]

John Darrington <john@darrington.wattle.id.au> writes:

> On Wed, Feb 22, 2017 at 04:48:55PM +0100, Mathieu Lirzin wrote:
>      
>      Sure it is!
>      
>      What I meant is that Ratpoison is not the most "intuitive" WM for non
>      GNU Emacs/Screen users.  As a consequence adding it in an example
>      configuration which is likely to be copy and paste, is maybe not the
>      most welcoming thing.  :)
>      
>      IMHO Openbox or anything which is able to launch a program by "clicking"
>      seems more friendly as a default (modulo the accessibility issues which
>      to my knowledge are not addressed by any of the "lightweight" WMs).
>      
...
> So most of them found "clicking" extremely unfriendly.  Please be very 
> carefull when making generalisations like "GUIs are intuitive" "mice are
> friendly" etc.  As we said before - it depends uponn the user.

I don't think I have made this generalisation.  :)

I haven't said that "clicking is more friendly".
I have precisely said "'clicking' seems more friendly as a default".

"defaults" are not adapted to everybody, especially in the case of people
who are computer illiterates.

-- 
Mathieu Lirzin
GPG: F2A3 8D7E EB2B 6640 5761  070D 0ADE E100 9460 4D37

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Leo Famulari]

On Wed, Feb 22, 2017 at 11:33:24AM +0100, Mathieu Lirzin wrote:
> Hi,
> 
> Leo Famulari <leo@famulari.name> writes:
> 
> > I think we should remove xmonad from the lightweight-desktop OS
> > declaration example. Xmonad is not "lightweight" in terms of disk space
> > or bandwidth, since it requires GHC, which is 859 MiB uncompressed.
> 
> Makes sense.
> 
> > Also, the i3 window manager is not very useful without dmenu and
> > i3status, so I think we should add them.
> 
> Agreed.

Okay, I'll wait another day or so for more comments.

> While we are on this lightweight template, what would you think of
> replacing ratpoison with something less "exotic" like openbox? :)

This is really a separate discussion! :)

Re: [PATCH 0/1] improvements to the lightweight desktop example

[Leo Famulari]

On Wed, Feb 22, 2017 at 11:48:54AM -0500, Leo Famulari wrote:
> On Wed, Feb 22, 2017 at 11:33:24AM +0100, Mathieu Lirzin wrote:
> > Hi,
> > 
> > Leo Famulari <leo@famulari.name> writes:
> > 
> > > I think we should remove xmonad from the lightweight-desktop OS
> > > declaration example. Xmonad is not "lightweight" in terms of disk space
> > > or bandwidth, since it requires GHC, which is 859 MiB uncompressed.
> > 
> > Makes sense.
> > 
> > > Also, the i3 window manager is not very useful without dmenu and
> > > i3status, so I think we should add them.
> > 
> > Agreed.
> 
> Okay, I'll wait another day or so for more comments.

I split it into two commits and pushed the master branch up to
0d7feebb622b0cf54ab435f308df347008d945b3.

> 
> > While we are on this lightweight template, what would you think of
> > replacing ratpoison with something less "exotic" like openbox? :)
> 
> This is really a separate discussion! :)