[xmonad] The future of PerWorkspace

Andrea Rossato mailing_list at istitutocolli.org
Thu Jan 31 00:02:53 EST 2008

On Wed, Jan 30, 2008 at 07:29:58PM -0500, Brent Yorgey wrote:
>   2. Andrea's LayoutCombinator class and core patch to move the result
>      of the description function into the X monad will help with some
>      but not all of the problems.  In particular it still won't work
>      with xinerama.

I'll read more carefully your interesting message, but I would like to
point out that my request to change the description method of the
LayoutClass has not been done to fix PerWorkspace but because I argue
that description is buggy, since it prevents me to write a
LayoutCombinator class and a CombinedLayout instance of the
LayoutClass that is general enough to be a ... class.

Since, as you and some users noted, the present state of the
repository, and of PerWorkspace specifically, is broken, I'm going to
push a patch, in the next few hours, that removes the LayoutCombinator
class and reverts PerWorkspace to its pristine "PerWorkspace
[WorkspaceId] Maybe Bool l1 l2 a" state, at least till when a decision
of the description type signature has been taken. (As a side note, I
would like to see someone step out to show me I'm plainly wrong with
my claim about the relationship between that specific type signature
and the expressiveness of a LayoutCombinator class. I wonder whether
I'm asking too much).

Now I go read the rest more carefully.

Let me take the opportunity to thank you for the documentation you are
producing on the wiki (and, if I understand correctly something is due
somewhere else... a novel!) about the xmonad internals.


More information about the xmonad mailing list