Make Emacs Better

I love emacs. I'm also aware that emacs is a really complex piece of software with staggering list of features and functionality. I'd love to see more people use emacs, but the start up and switch cost is nearly prohibitive. I do understand that getting through the "emacs learning curve" is part of what makes the emacs experience so good.

That said, there really ought to be a way to make it easier for people to start using emacs. Think of how much more productive some developers and writers would be if the initial experience of emacs was less overwhelming. And if emacs were easier to use, developers could use emacs as a core (embeded, even) component of text-editing applications, for instance, some sort of specific IDE built with emacs tools, or a documentation creation and editing toolkit built with emacs. I'd go for it, at least.

To my mind there are three major challenges for greater emacs usability. Some of these may be pretty easy to change non-intrusively, others less so. Feedback is, of course, welcome:

1. The biggest problem is that there's no default configuration. While I appreciate that this provides a neutral substrate for people to customize emacs for themselves, you have to write lisp in order to do pretty much anything in emacs other than write lisp. And customize-mode is well unmentioned, but not particularly usable.

Perhaps one solution to this problem would be to create a facility within emacs to build "distributions," that come configured for specific kinds of work. That way, emacs can continue to be the way it is, and specialized emacs can be provided and distributed with ease.

2. Improve the customize interface. I like the idea of customize, but I find it incredibly difficult to use and navigate, and end up setting all configuration values manually because that's easier to keep track of and manage. I'd prefer an option where you configure your emacs instance the way you want (through some sort of conventional menu system), and then have the option of "dumping state" to an arbitrary file that makes a little more sense than the lisp structure that customize produces. Then, as needed, you could load these "state file(s)," But then I've never used the menu-bar at all, so perhaps I'm not the best person to design such a system.

This strikes me as a more medium term project, and would make it easier for people who want to modify various basic behaviors and settings. I don't think that it would need to totally supplant customize, but it might make more sense.

3. Improve and add the ability to extend emacs beyond emacs-lisp. I initially thought emacs-lisp was a liability for emacs adoption and I don't think this is uncommon, but I've since come to respect and understand the utility of emacs lisp. Having said that, I think offering some sort of interopperability between emacs and other languages and interperators, might be a good thing. Ideas like ParrotEmacs and using the Guile VM to run existing emacs-lisp in addition to other new code would be great.

This is a longer term project, of course, but definitely opens emacs up to more people with a much more moderate learning curve.

I've been working (slowly) on getting my base configuration into a presentable state that I can push it to a git repository for everyone to see and use, which (at least for me) might start to address problems one and two, but three is outside of the scope of my time and expertise. The truth is that emacs is so great and so close to being really usable for everyone, that a little bit of work on these, and potentially other, enhancements could go a long way toward making emacs better for everyone.

Who's with me? Let's talk!

comments powered by Disqus