Re: Support (or not) for Emacs 21, and XEmacs

[Carsten Dominik <dominik@uva.nl>]

Hi,

I have made a test version of Org with the compatibility code
for XEmacs and without some of the left-over compatibility
code for Emacs 21.

I would appreciate if some people could check out the
`remove-compatibility-code' branch from repo.or.cz and report back
if they experience any problems with Emacs 22, 23, and 24, either
during compilation or while running.

Problems with Emacs 21 and XEmacs do not have to be reported.... :-)

Thank you very much

- Carsten

On Apr 14, 2010, at 10:15 AM, Carsten Dominik wrote:

> Hi everyone,
>
> I am, once again, reconsidering support for Emacs 21 and XEmacs
> in Org-mode.
>
> Supporting older versions of Emacsen does make the code in
> Org-mode more complex, needs time to rewrite or copy features
> that do exist in newer versions of Emacs.  Also, it slows down
> the process of updating the version of Org-mode distributed
> with Emacs, because the compatibility code makes it often hard to
> comply with Emacs rules and practices.
>
> What I would like to do is the following.
>
> We are now working toward release 7.0 of Org-mode, which will
> hopefully see full integration of org-babel.  I am considering
> to make Org-mode version 6.35 the last version of Org-mode
> supporting XEmacs and Emacs 21.  That would mean that I would
> still fix critical bugs in 6.35 and keep that package on the
> website indefinitely.  So people stuck with Emacs 21 and XEmacs do
> have an excellent version of Org-mode available.  Git users will
> be able indefinitely to select this version of Org-mode to support
> and Emacs 21 or XEmacs installation.
>
> For version 7.0 I would then not only stop actively supporting
> these older Emacsen, but I would actually remove some of the
> twisted code that has accumulated over the years in order to
> support older versions.  Important examples are all the
> stuff in org-compat, but also key bindings and other stuff.  There
> are about 100 cases of (featurep 'xemacs) in the sources, and many
> other places where I have to check for the availablility of a
> specific feature or function.
>
> If you object to such a development, please step into this
> discussion.
>
> - Carsten
>
>
>

- Carsten