Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!samsung!usc!snorkelwacker.mit.edu!ai-lab!life!burley From: burley@geech.ai.mit.edu (Craig Burley) Newsgroups: comp.sys.amiga.introduction Subject: Re: Emacs for the Amiga 1000 ? (MicroEmacs?) Message-ID: Date: 15 Jan 91 14:05:37 GMT References: <17399@cbmvax.commodore.com> <1991Jan15.091854.24428@zorch.SF-Bay.ORG> Sender: news@ai.mit.edu Distribution: comp Organization: Free Software Foundation 545 Tech Square Cambridge, MA 02139 Lines: 91 In-reply-to: xanthian@zorch.SF-Bay.ORG's message of 15 Jan 91 09:18:54 GMT In article <1991Jan15.091854.24428@zorch.SF-Bay.ORG> xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) writes: Frankly, as destructive as that binding is for a keyboard with the delete key away from the main keyset, rebinding it should be the subject of the first paragraph in the GNU emacs manual. I have no intention of _ever_ becoming so expert at the arcana of any editor that I have to reprogram the editor before I can use it. It is, as Humpty Dumpty once said, a matter of who is to be master. EMACS key bindings were developed on terminals with huge keyboards and lots of keys with names like META and such. It is frustrating to rebind backspace to delete and such, but that's the EMACS history. I used the Xwindows "xmodmap" to do the rebind, so I get it everywhere (EMACS isn't the only environment I use that assumed DEL, not BS, meant delete previous character). It is just too much to expect a touch typist to use the delete key for emacs, and the backspace key for everything else typed. Rotten human engineering counts for a lot, and that one takes the cake. I've sent that message to Stallman at least twice, and his total lack of a repair tells me all I want to know about the chances of "free" software ever being user friendly. The profit motive accomplishes a lot. Unfortunately, the Free Software Foundation is still only at the stage of doing programmer-oriented software, and for that audience, EMACS is hardly "rotten engineering". Few programmers use EMACS without modifying it their own way (I've never seen any programmer use it without changing it). Even rms, who apparently decided to make all the GNU EMACS defaults be whatever he wanted as his own environment, has some stuff in his startup file (though the smallest amount I've ever seen in a programmer's EMACS startup!). Note that DEL and BS have distinct meanings in the ASCII standard, and given that all the other one-key strokes have "standard" meanings under EMACS, providing a one-key "help" required sacrificing either DEL or BS. Since BS almost never does what is expected (backspace, allowing overstrike, as on a typewriter) but DEL does (delete previous character), and since all pertinent keyboards allow control-H but not all allow a control- sequence to produce DEL (and these latter ones typically provide DEL as the canonical delete-previous key), providing the critical help function on a key that is always available probably made a perverse kind of sense. The problem is a combination of problems -- no standard keyboard layout and meaning for other than alphanumerics, and an overly crowded map of single-character keystrokes in standard EMACS. When the developer-oriented GNU environment is largely built, the FSF presumably will turn its attention to user-friendly, ready-right-out- of-the-box end-user applications like word processing, draw programs (in X? sigh), personal financial programs, and the like. After making the typical bunch of programmer-centric mistakes, I expect they'll get it right -- after all, there will be so many people like you who'll just hack the code to make it do the right thing if it doesn't already! The FSF will need that kind of input and bug-fixing even more than it does now. (Which makes me wonder -- why don't you just make a version of GNU EMACS available that has YOUR favorite key bindings as defaults? It really isn't a bad idea -- you could essentially start from scratch and make a much more sensible set of bindings for typical UNIX keyboards, or maybe even for the Amiga itself. Or at least provide a canonical startup file for free, for people who don't want to get into programming EMACS. A lot of people might really appreciate that.) With this miserable exception, you can rebind the keys to whatever you want, and Gosling's binding was in wide use for years before the GNU project's version became available, so deciding which is the true standard is a) pointless and b) meaningless, and c) a religious question anyway. The true standard, if any, would be the original version of EMACS. I don't remember whether BS was bound to a help function and DEL bound to delete previous character, but as I used DEC and those aforementioned huge-keyboard terminals back when EMACS was first written and did play with EMACS then, my guess is that DEL indeed was the delete keystroke. However, I'm quite sure that ^X-^F was always "find file", not "write changes and exit" (which I think is what Gosling's EMACS does). And obviously the original EMACS predated Gosling EMACS. And I believe GNU EMACS is more faithful to original EMACS, which is to be expected, of course! (Of course, the extension language is a "leeetle deeeferent", but that goes for Gosling EMACS as well. :-) In any case, the reason deciding which is the true standard is a) pointless and b) meaningless is precisely BECAUSE everyone is expected to make their own extensions and rebind some keys. It would be critical to decide which is the true standard if we all had to live with the key bindings as delivered -- but then if that were the case, we probably wouldn't all be using EMACS (i.e. its audience would be much smaller), would we? -- James Craig Burley, Software Craftsperson burley@ai.mit.edu