> From: Dan Espen
> Sent: Saturday, April 21, 2012 12:31 PM
> To: help-gnu-emacs@gnu....
> Subject: Re: Emacs: Problems of the Scratch Buffer
>
> Chiron <chiron613.no.spam.@no.s...> writes:
>
> > On Sat, 21 Apr 2012 12:08:25 -0400, Dan Espen wrote:
> >
> >> I'm having a hard time figuring out what _would_ be useful in a scratch
> >> buffer.
> >
> > I use it to check out lisp statements as I try to learn the language.
> >
> > I suppose it would be very helpful if I were to want to make changes to
> > emacs using elisp, to test and debug my stuff.
>
> I've tried that on occasion but I soon end up in my .emacs or some .el
> file and evaluating things there is easy enough.

I have similar feelings.  It might use *scratch* more if it were in text mode, because mostly my spur-of-the-moment uses of Emacs in my Windows development environment are to "neutralize" copied text before pasting into some tool that's paying too much attention to formatting.  (I almost always just want the simple text, and sometimes it's surprisingly hard to get the stuff back to simple text after pasting it.)

I know how to change the mode and how to make that permanent, but haven't because I have another personal journal-like file open in a buffer that is in text mode all of the time (so use it for neutralizing text), and for those few occasions when I need Lisp interpretation, I switch back to *scratch* and do so.

If there were something like M-x scratch, I probably would just delete *scratch*, and in this mythical world, if there were a further option to not create *scratch* unless there is nothing else for Emacs to show, I would use that too.

Now, can anyone tell me why *scratch* has RET bound to (newline) instead of (eval-print-last-sexp)?  Seems that the bindings of RET and C-j are reversed for common uses....

Cheers,
Mark