I think TextMate's Rails mode came before emacs-rails and inspired it. That's probably part of the reason. Another part is that TextMate behaves much more like a Mac app, and Rails was started by Mac users. So it's partly history, partly momentum.
I also tried TextMate, and do still use it for some things (I like the personal text wiki extension, which I hacked a bit to use MediaWiki syntax), but found myself programming back in Emacs in short order. TextMate's extension mechanism is incredibly obtuse compared to elisp (the Lisp-intolerant claim this is a strength, but just try making sense of how to customize a non-trivial TextMate mode and see how far you get). Running external scripts just to do basic things seems pretty inefficient to me. Also, its UI needs lots and lots of work. The inability to simultaneously look at two buffers in one project is a deal killer for me. I'm sure it'll improve in the next version.
Back on topic: native Unicode in Emacs 23 sounds great. So now, after only a few months of running the released version of 22, I guess I'll be back to running Emacs out of CVS. :)
I also tried TextMate, and do still use it for some things (I like the personal text wiki extension, which I hacked a bit to use MediaWiki syntax), but found myself programming back in Emacs in short order. TextMate's extension mechanism is incredibly obtuse compared to elisp (the Lisp-intolerant claim this is a strength, but just try making sense of how to customize a non-trivial TextMate mode and see how far you get). Running external scripts just to do basic things seems pretty inefficient to me. Also, its UI needs lots and lots of work. The inability to simultaneously look at two buffers in one project is a deal killer for me. I'm sure it'll improve in the next version.
Back on topic: native Unicode in Emacs 23 sounds great. So now, after only a few months of running the released version of 22, I guess I'll be back to running Emacs out of CVS. :)