Hacker News Reader

Comment by robocat | Hacker News Reader

I love this: “Or, as Richard Stallman put it”: ‘It is unfortunate that the people who implemented the newer editors chose incompatibility with Emacs.’

It is interesting to see a comment that is 100% at odds with the brutal evolution of ideas that occurs in a good ecology of software.

If the same mindset applies to the rest of emacs then they really stand no chance or real progress. (edit: the rest of emacs, and not just key bindings which the RMS quote I have removed from context applies to).

View on HN

Yes, what really pisses me off about emacs is the mind-numbing arrogance - they're basically saying "people ought not to have broke from our button scheme, so we'll just ignore it to our users' detriment!".

There's this implicit argument that convention is some sort of monarchist hierarchy where the successor mustbenamed as heir first, and what convention is most common is somehow irrelevant. It's infuriating.

Emacs has lasted 40 years with a stable (but small) market share. It's outlived every other editor with the exception of Vim. It will probably outlive Atom, and it looks like it's going to outlive Netbeans.

I'd argue stubborn resistance to most new concepts regardless of their sheen is a contributor. I think people underrate that as a trait. You want some bastions. Loads of it is insane but they also don't make many new mistakes.

I don't think so. Emacs developers are quite friendly and accept opinions and advices of anybody, feel free to write to developers: https://lists.gnu.org/archive/html/emacs-devel/

That isn't Emacs, that's Stallman, who is known to be stubborn.

It already caused a major fork, XEmacs, which used to be important, I suppose, but a combination of GNU Emacs catching up with the features XEmacs people wanted and XEmacs simply sputtering out has made GNU Emacs the last full Emacs standing again.

I suppose rms can't understand why anybody would think anything could be better than what he chose to do 40-something years ago.

A lot of things are different without being better.

Before ctrl-x/Ctrl-v became the standard for Cut/paste, it was Ctrl-del/Ctrl-insert in many dos and early Windows editors - which makes a lot more mnemonic sense.

It’s better to standardize on something, but that something isn’t necessarily better and is often worse.

Ctrl+Ins for copy, Shift+Del for cut and Shift+Ins for paste. Still works in Windows. Ctrl+Del deletes a word.

DOS editors often used WordStar bindings. C-k C-b to start a block, C-k C-c to copy or C-k C-v to move. Turbo Pascal used these bindings. They're also the default bindings for the `joe` editor.

And Ctrl+Ins, Shift+Del, Shift+Ins work out of the box in (windowed) Emacs, too.

C-x and C-c make some sense as cut and copy. But C-v, C-z, C-y as paste, undo, and redo just make little sense. The first two make sense for their proximity to x and c in the QWERTY layout. y only makes sense if it happened to be unassigned.

Keyboard shortcuts will always fall into one of two categories: That makes perfect sense (usually mapping to a mnemonic like C-a for "select all"); That's totally arbitrary (C-y). It makes sense (and cua-mode offers it) to offer the opportunity to rebind the keys to the mainstream arbitrarily selected shortcuts, but it also makes sense to keep what's been used by emacs users for 40 years since that's what they're used to at this point (oh hell, I've been using it for 20 years, time has flown).

Mnemonic keyboard shortcuts must be a bastard to learn if you don’t speak English.

And shortcuts based on finger distance must suck for anyone using a non-qwerty keyboard (especially a non-latin character based language!).

Indeed. I use the Dvorak layout (a choice I made twenty years ago and have stuck with since my typing speed did improve, by virtue of finally learning to touch type properly), and many shortcuts make no sense. On the other hand, I actually found emacs much more comfortable to use.

The P and V operations on semaphores are from Dutch. I find this difficult, "The V kind of looks like a down arrow.. but wait no that's increment. The P could mean plus.. wait no that's decrement..."

Note that redo is often ctrl+shift+z which is much easier to type. However redo is a weird case because traditionally most GUI applications (at least on Windows) didn't have explicit redo even until late 90s/early 2000s and redo was just undo a second time (ie. undo was just a flip between the last and previous state). When applications started to add an undo stack and redo command there wasn't a commonly accepted shortcut key for it so they used whatever made sense for the developers - which is how ctrl+y (previous to letter z) and ctrl+shift+z (in Windows shift often inverts the action, e.g. alt+tab vs alt+shift+tab) came to be (though those weren't the only ones).

A third category is those chosen for their physical position.

ctrl x/c/v is best because you can do it with your left hand while your right hand is using the mouse.

Yes! Exactly!

Those might make more sense, but even if my MacBook had Insert & Del keys, I doubt I could Ctrl- them with one hand.

FWIW it was Shift+Del for Cut, Shift+Insert for Paste and Ctrl+Insert for Copy and AFAIK that came from IBM's CUA. Also these shortcuts are still supported by Windows (and personally while i use Ctrl+X and Ctrl+C for Cut and Copy, i tend to use Shift+Insert for Paste :-P).

The only version of Emacs I ever used regularly was (Lucid|X)emacs. According to folks on the Lucid side working with RMS was difficult[1] to say the least. In the end, Xemacs lost traction (and development) and I moved on from Emacs, it just didn't suit how I wanted to work well.

Emacs has the developers to correct the problems but they need to let go over a 40 year old vision and evolve a bit.

[1] https://www.jwz.org/doc/lemacs.html - Of course, this is biased towards to Lucid side of the story. But I remember following this.

Reply on news.ycombinator.com