Re: Finding packages to enable by default

[Stefan Monnier]

>> cua-mode

I don't think this can be default in the foreseeable future.  We can
(and do try to) make it super-easy to enable it, but it's hard to go
much further than that, given the inherent difficulties and associated
occasional quirks.

>> cua-selection-mode

IIRC this is basically delete-selection-mode, so I'll disregard it.

>> column-number-mode

We could enable this.  I hadn't noticed it as a popular choice, but
if people like it, I have no particular objection and AFAIK the code is
ready for it.
[ I don't use it myself, but I probably wouldn't bother disabling it
either.  ]

>> delete-selection-mode

I do not want to enable delete-selection-mode in its current state.
I would welcome a new interactive-form element (along the lines of "^")
which would cause deletion of the region if active, and which could be
added to self-insert-command, yank, and a few others.
Maybe under the control of a new value of `delete-active-region'.

Once this is done, I'd have no technical objection to enabling it by
default, tho I don't know if that's really the desire of the majority.

[ I don't use it myself, and I'd probably disable it for my own use.  ]

>> desktop-save-mode

IIUC there's no technical hurdle either.  I dislike the functionality
(the whole point of restarting Emacs, for me, is to get rid of the
umpteen frames and buffers ;-).  But if most people like it...

>> dired-x

No opinion on this one (I basically don't use dired).

>> global-linum-mode

linum is buggy, so it's not an option.  nlinum-mode would be a better
choice, but the potential performance issues, together with the extra
screen real-estate used up... I have a hard time believing that most
people would want it enabled everywhere.

>> global-subword-mode

Hmm... never thought about it, but I guess I could see why that could
be popular.  I haven't looked much at the implementation, but other than
that I could go along with it.

>> icomplete-mode

I see icomplete as "the way to add ido/iswitchb functionality to the
default completion", so I like it in this sense, but I'm not sure about
enabling it by default.  I had it enabled in my .emacs but got rid of it
at some point, can't remember why.

>> iswitchb-mode

Obsolete.

>> recentf-mode

I see no reason not to enable it, indeed.

>> savehist-mode
>> saveplace

Actually, for these (as for recentf-mode), I do remember one reason:
when multiple Emacsen as running at the same time, they all want to
change the same file(s) and end up stepping on each others's toes.
So if we want to enable these, we'd first need to address this
technical issue.

>> show-paren-mode

Too much in your face for me.  It doesn't seem to be that popular.

>> size-indication-mode

Uses up too much mode-line estate, IMO.  I instead use the relative size
of the scrollbar thumb as an indicator of the total file size.

>> uniquify

Already enabled by default now.

>> which-function-mode

It can be a bit unreliable, but enabling it might be an opportunity to
fix those issues.  It does seem to call for a improvement on mode-line
length (probably by auto-contracting some of its elements, such as the
buffer name, the mode names, ...).

>> winner-mode

I could go along with that.  Tho I'd also like to see more generally
improved handling of window-state handling (ability to save&restore
window-states.  We can do that via window-configurations in registers,
but it only works in a single frame, and can't be saved, and I feel like
registers are "too hard to find").


        Stefan