From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <463CF37F.5050808@conducive.org> Date: Sun, 6 May 2007 05:13:35 +0800 From: W B Hacker User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1.2) Gecko/20070221 SeaMonkey/1.1.1 MIME-Version: 1.0 To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu> Subject: Re: [9fans] what a surprise References: <8a9c0087a8daf8bbf071cec82a340899@proxima.alt.za> In-Reply-To: <8a9c0087a8daf8bbf071cec82a340899@proxima.alt.za> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Topicbox-Message-UUID: 5ae10ea8-ead2-11e9-9d60-3106f5b1d025 lucio@proxima.alt.za wrote: >>> - the more familiar with *any other* CLI environment, (possible exception of >>> Oberon/Aos) the *more difficult* it is to adapt to acme/rio. >> i doubt this is the case. it wasn't for me. > > Not even as bad as getting used to VI. There lies a benchmark, if > there was one. ACK, Best fixed with replacement of /bin/vi with the binary of an editor. Any editor. Even those that lisp... ;-) > And the fact that VI can become second nature is I think the legal profession calls those 'repeat offenders', psychologists 'masochists'. :-) > enough to persuade me that icon-lovers will simply never know what > they are missing on the other side of the fence. > ..'alternative lifestyle' for that one... I don't really care one way or the other about 'icons', and I do see the merits of some of the initially off-putting anomalies - IF one cares to stay around long enough, read and listen - to find out the 'why' of them. But by no means all of them make equal sense, even after research... It would be 'nice', for example, to have a screen that defaulted to scrolling UP as it filled instead of printing off the bottom of the view window.... Or a slider-bar on the same side of the window as the mouse and cursor manip keys *commonly* sit. Or an (optional?) CLi history buffer. And even a hardware TTY is smarter than to run its paper bass ackwards, hiding what I haven't yet read while preserving what I have already read.... ..have yet to grok the 'why' of that one... Mind - not because these things are right, wrong, sideways, or mandated by the Gods somehow. But because, like flush toilets, most folks *expect* them to act a certain way. Hard to get onboard with a different - possibly better - way of working if the first thing the new environment does is confuse a person with contrarian responses. Shouldn't be rocket science to provide a less hostile tolset, if only for transition. Yet it seems to not have been done, lo thjese many years. And the 'why' of THAT is even harder to grok... > But the real crux lies in measurement and no one has yet suggested > what it is that ought to be measured here, Presence of waste motion. or NOT. Excessive mental/physical 'context switching' in wetware. or NOT. > nevermind how to measure > it. 'Therbligs' applied to swapping mouse-kbd-mouse, hands for starters. The 'mental side is harder... possibly a suite of diverse tasks to key-up - timed tasks exercising a wide range of input techniques. But these would need to be such as can be done with *potentially* the same total key / click / drag count in either the Plan9 way AND various 'other' ways, not a test of 'plumbing' vs ifconfig _ /etc/ edits, NFS / SMBFS mounts & such. Those measure the OS architecture and toolset, not the 'hmi'. And, sorry - they are not really 'married' to each other. If they were, shell scripts would not exist. > If we could set down some criteria against which one rates a user > interface (I seriously doubt that popularity should be such a > criterion), Not sure. 'Popular' designs got that way by being generally learnable/acceptable to a great many 'casual' users, not 'coz the 'pro' particulalrly cared for them. And at least part of that relates to manual dexterity or lack thereof. Really 'focused' users very often prefer methods that the general public consider weird, 'coz there is a payoff - center-mounted accelerator pedals on race cars for either-foot heeel-and-to control, for example. And acme/rio have some of that sort of flavor - for coders anyway. > at least we'd have the option of a meaningful discussion. > But I suspect that we are all voicing subjective opinions on some very > personal idea of "comfort". > > ++L > Doubt that part would *ever* change - regardless of testing. Habits die hard. After all - people are still buying cigarettes. And worse. While it could *be* tested, (Google 'therblig' and 'Industrial Engineering'), there is probably no gain in going there. It isn't that hard to 'go thou and do *otherwise*'. Love acme/rio and use it as-is or simple replace or re-code it to something you like better. Personalization is one of the neater things about software... Bill