Many innovations in source code manipulation have appeared first in Emacs, and I think we should turn to Emacs as the most practical research venue. Sadly, it seems we have an immediate obstacle in the font support built in to Emacs. I recently wasted a very unhappy day trying to configure Emacs to change font reliably, and I'm fairly convinced that the font management is very broken, perhaps needing reimplementing from scratch. Would it be worth the effort of fixing this as a first step toward a non-VT100 program editor?
Actually, I'd be inclined to cherry-pick the useful bits of emacs into something more modern, such as eclipse, which I've been using for some Java work recently, and which seems to be picking up some momentum (though I will have a look at the other things suggested here too).
Now, I'm not suggesting that we somehow try to reverse the emacs wagon train into the eclipse garage. Rather, I think it might be worth trying to identify those things that, from a user's perspective, make emacs powerful enough to keep being used despite its flaws, and see if they could be re-created within a more up-to-date presentational framework.
There's a whole bunch of little, practical things I like about emacs: it's keyboard-centric, has incremental regexp searching, can parse compiler errors, can open a zillion files at once, can edit files on remote machines, has a reasonable amount of undo history, a fairly lengthy set of mostly-working syntax-highlighting- and-indenting modes, can nearly make sense of revision control information, is programmable for when you're desperate or bored or inspired, and is just about customisable enough, as long as you're happy in its big-monospaced-terminal-window-with-ugly-widgets world. And it's pretty zippy and lightweight, despite its heritage.
As Alan has found, though, how it plays in the modern GUI world is pretty clunky, and so I think it might be worth separating all the functional bits from the presentational bits, actually throwing the presentational bits away, and seeing if the functional bits can't be transplanted into some young and healthy GUI body with everything to live for, and all the right connections into some modern cross-platform desktop environments, such as qt or bonobo or something.
Of course, I realise I'm probably suggesting the functional equivalent of the Strategic Defense Initiative in the editor wars, and if the outcome is only going to be the ability to delete slowly moving italicised red text, then it's probably going to be folly, and we risk succumbing to the harsh boot of Vi-land.
But, dammit, it's a risk worth taking. We must ask ourselves not what emacs can do for us, but what we can do for emacs. Only then can victory, and ultimate cross-platform source code readability, be assured!
Death to character mode!! Vi users are dorks! Hey, what the ...?
// KERNEL TRAP 101: subscriber enthusiasm overflow. // Please delete this message and forget all about it. // Thank you for your cooperation. You have been warned. -- Mr Grumpy, cleaning up the mess left by Frank Wales [EMAIL PROTECTED]
---------------------------------------------------------------------- PPIG Discuss List ([EMAIL PROTECTED]) Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss Announce admin: http://limitlessmail.net/mailman/listinfo/announce PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/
