[ESS] Polymode is on MELPA

[Andreas Leha]

Hi Phillip,

Phillip Lord <phillip.lord at newcastle.ac.uk> writes:
> Andreas Leha <andreas.leha at med.uni-goettingen.de> writes:
>> Very nice!  I am looking forward to such functionality for org->R !
>
> I'll add that next if you want it. The configuration for org->R is the
> same as org->el with a different comment character.
>
>

Very nice., I'd be interested in this.  Just to be sure: are we talking
about https://github.com/phillord/linked-buffer?

In that case I'd still need to buffers open for the same file, correct?
I'd just have them open permanently instead of creating a source buffer
when I want to edit source code?

>> More general question: What is the problem with indirect buffers? They
>> seem to work nice in org.
>
> I talked about this with Stefan Monnier on emacs-help I think, so the
> discussion should still be available.
>
> The bottom line problem is the one Vitalie alludes to. Indirect-buffers
> share text AND text-properties. This is why syntax highlighting breaks
> -- both buffers try to change it at once. The end result is that, while
> indirect-buffers seem like an incredibly cool ideal, in practice they
> aren't. This isn't fixable because of the way that the Emacs data
> structures work.
>
> The bottom line problem is that text properties are too tied up with
> strings. This does meant that they are very fast and scale -- with
> overlays (which do the same things as text properties effectively) the
> start and stop position of each overlay has to be recalculated every
> time the buffer changes. If font-lock used overlays, the problem would
> not be there. But it does, and so do many other things.
>
>
>> Will (have to) be replaced in other places (like inside org) as well?
>
> Don't quote me on this, but I think org-mode has already replaced them.
>

I won't ;-) My back-log on the orgmode mailing list is 2000 messages
now.  That is also the time of my latest git-pull....

Andreas