OK. I'm already prone to RSI when I am programing C and Perl. However, when I spend more and more time in CL, the symptom is getting worst on my left forearm because of the extended use of the left shift key (to reach '(' & ')')
(BTW, since you can use ' and qw// to delimit string in Perl, I can get away by not using L-shift most of the time, but again in Lisp this is not an option.)
This pains me to the point where I have to stop programming and research on solutions tonight.
Basically what I did is swap 1234567890 with !@#$%^&*() [] with {} ;; for C programming and ' with "
I immediately tried it out and to my surprise I got used to it in less than 5 minutes.
I found that after swapping the keys I no longer use C-c C-q to close all the parens as typing the ) manually doesn't hurt at all.
Also I'm more inclined to type the doc string (:documentation "blah blah") because " isn't hard to type now. So is * when typing global variable *current-process*.
The only thing that I still confused myself is the single quote which every so often I need to type (You'll, can't) etc in comments.
But other than that I can live without the shift key 90% of the time (I occasionally type numbers using the shift key but if I can I prefer the numpad instead)
One thing that bugs me though is that this program doesn't work with X-Win32 so I can't use it with my x-session, and it'll crash SecureCRT (my favorite ssh client). I works okay with most win32 applications including putty (that's the only two I found not working). Too bad this program is no longer maintained. (They never get to release it as shareware)
I just want to share this. If you have RSI (or developing it) you might want to try if this is gonna help.
There are other programs for win32 that can remap the caplock to ctrl by just changing the registry. However I'm not aware of a solution that can remap the shifted keycode by just hacking the registry (this Keyboard Remapper runs a residual process to translate the keycode).
BTW, I know in x-windows there's a xmodmap. I only know how to remap the caplocks key to ctrl. If you know how to remap the keys like I described above, please share!
"email...@gmail.com" <email...@gmail.com> writes: > OK. I'm already prone to RSI when I am programing C and Perl. > However, when I spend more and more time in CL, the symptom is getting > worst on my left forearm because of the extended use of the left shift > key (to reach '(' & ')')
if you use emacs this can be done for on all the platforms you use emacs on. you won't get the effect outside of of emacs, but that's just another reason to never leave emacs :)
-- -Marco Ring the bells that still can ring. Forget the perfect offering. There is a crack in everything. That's how the light gets in. -Leonard Cohen
"email...@gmail.com" <email...@gmail.com> wrote: > OK. I'm already prone to RSI when I am programing C and Perl. However, when > I spend more and more time in CL, the symptom is getting worst on my left > forearm because of the extended use of the left shift key (to reach '(' & > ')')
Since dvorak wouldn't help with parens, you should use pedals to bind Shift, Control and stuff like that. This is not a joke. See for example http://www.bilbo.com/bilbo.html. I also have the project to write a small application to use my many MIDI foot controllers (for musical instruments) for generating X Events, including keyboard ones. I don't suffer from RSI, but this would increase the typing efficiency (especially under XEmacs ;-).
On 20 Jul 2005 02:02:25 -0700, "email...@gmail.com" <email...@gmail.com> wrote:
> There are other programs for win32 that can remap the caplock to > ctrl by just changing the registry. However I'm not aware of a > solution that can remap the shifted keycode by just hacking the > registry (this Keyboard Remapper runs a residual process to > translate the keycode).
>> OK. I'm already prone to RSI when I am programing C and Perl. However, when >> I spend more and more time in CL, the symptom is getting worst on my left >> forearm because of the extended use of the left shift key (to reach '(' & >> ')')
> Since dvorak wouldn't help with parens, you should use pedals to bind > Shift, Control and stuff like that. This is not a joke. See for example > http://www.bilbo.com/bilbo.html. I also have the project to write a small > application to use my many MIDI foot controllers (for musical instruments) for > generating X Events, including keyboard ones. I don't suffer from RSI, but > this would increase the typing efficiency (especially under XEmacs ;-).
From their webpage:
STEP ON IT! Pedals: Order Information This product is D I S C O N T I N U E D
"email...@gmail.com" <email...@gmail.com> writes: >OK. I'm already prone to RSI when I am programing C and Perl. >However, when I spend more and more time in CL, the symptom is getting >worst on my left forearm because of the extended use of the left shift >key (to reach '(' & ')')
Amazing.. I never use the left shift key for (), actually, I don't use the left shift key much at all...
But I guess, you'd then get RSI in the right arm ;-(
-- * Harald Hanche-Olsen <URL:http://www.math.ntnu.no/~hanche/> - Debating gives most of us much more psychological satisfaction than thinking does: but it deprives us of whatever chance there is of getting closer to the truth. -- C.P. Snow
Harald & Didier, that's interesting. But I have to limiited it to some "portable" solutions as I do access other computers from time to time (laptop from work, for example).
Edi, thanks for the link. pcmag wants me to pay ~ $6 before I try. However, google is our friend. I downloaded it and tried it out. It's like the other registry key map programs. It couldn't do what I described.
Marco, that's the best solution in the moment, many thanks. Although if I can find a system wide solution, that'll be ideal (either win32 or x-windows).
I'll insert this to my .emacs and check into my svn repo so I can update it on all the machines I have access to.
Thanks everyone for the interesting links and ideas. I'll see if my RSI symptom improves as a result of this change.
On 20 Jul 2005 03:25:34 -0700, "mac" <email...@gmail.com> wrote:
> Edi, thanks for the link. pcmag wants me to pay ~ $6 before I try.
Yes, a horrendous amout of money...
> However, google is our friend. I downloaded it and tried it > out. It's like the other registry key map programs. It couldn't do > what I described.
I noticed that they were talking about two different versions of the program and that Version 2 was supposed to be significantly different from Version 1. Are you sure your, er, Google workaround[1] found the right program /and/ the right version?
Cheers, Edi.
[1] BTW, if you were able to obtain the very program I was talking about this way then someone's obviously infringing their copyright.
--
Lisp is not dead, it just smells funny.
Real email: (replace (subseq "spamt...@agharta.de" 5) "edi")
> Are you sure your, er, Google workaround[1] found the > right program /and/ the right version? >From the 'help -> about' dialog, the version number and release date
match what they described in the article. So I think they are referring to the UI, that is radically different from version 1 (I have no way to tell).
>[1] BTW, if you were able to obtain the very program I was talking > about this way then someone's obviously infringing their > copyright.
I did not intend to "steal" it, just want to find out how it works. Having to pay upfront before I know whether it solves my problem is a bit silly...
> I'd gladly pay $300 for it if it works as described. > It's expected to hit the market in 2006. Finger crossed.
if you don't need the fancy LCD keys (you could just paint your layout on the keys), then it should be possible for < $20, for example there is a project for generating keycodes from customer hardware:
Combining this gives you a small device, which you can plug between the keyboard and the PC and which can be programmed from the PC for your custom layout. I wonder if someone did something like this already, it is so obvious.
I have some experience with PIC programming, see for example this simple project: http://www.frank-buss.de/blinker/ (I've done far more complex projects, but they belong to the companies for which I developed it), so let me know, if this is useful for you and I can develop it at weekend and publish it on my webpage (of course, you need someone who can program a PIC and solder some parts, or I can sell you the prototype, if you want).
>I'd gladly pay $300 for it if it works as described. >It's expected to hit the market in 2006. Finger crossed.
Doesn't look very ergonomic on finger muscles etc to me. Too flat, with not enough downward play and resistance. I'd recomend a high quality keyboard with real springloaded switches in the keys, not the rubber-mat stuff that is ubiquitous with nearly all keyboards. Examples are certain models from Cherry and the old IBM "steel-plate"-style keyboards (I don't know if IBM still ships them; I've recently found on the net the company which made them for IBM and they're still selling them but I forgot the URL). And of course if you're an Emacs user, remap capslock to control, if you're using a PC-style keyboard. Using the control in the lower left corner is begging for cramps and strained tendons.
> >OK. I'm already prone to RSI when I am programing C and Perl. > >However, when I spend more and more time in CL, the symptom is getting > >worst on my left forearm because of the extended use of the left shift > >key (to reach '(' & ')')
Perhaps consider using Emacs and meta-() which inserts parens in balanced pairs. Then you just have to hop forward/backward over them, but you have half as many to insert using a shifted operation.
The other option that's even better is, again with Emacs re-definability, swap [] and () so that you have parens for lowercase. The symbolics keyboard offered () on lowercase and a lot of people liked it. (It also had colon on a shift-insensitive key, which some people used.)
> Amazing.. I never use the left shift key for (), actually, I don't > use the left shift key much at all...
Heh. I use the left shift key exclusively. Weird world, huh? (I also just hold that left shift key when typing all-uppercase rather than hit shift-lock because it doesn't seem to bother me.)
> But I guess, you'd then get RSI in the right arm ;-(
Btw, I have a personal theory that the people who get this problem, or, at least, carpal tunnel problems, are primarily those who didn't do enough sports as a kid. That is, I've noticed some people are mostly immune to these effects and some people get them right away. I think that if you do things with your wrist (which might be heavy typing, but which can also include various "wrist sports") when you're a kid, you work the tunnel into a larger shape while you're still growing and it happens less. I type very fast for many, many hours a day and seem to have no effect, and I attribute it to my childhood sports: tetherball, bowling, competitive swimming, ping pong, volleyball, horizontal bar, etc. My pet theory is that many computer programmers start out as indoor folks who avoided sports and were pretty introverted, and that people who were outdoorsy/sports-oriented didn't get into computers as much. (I was the introverted part, but my dad pushed me to do some sports on some bizarre theory that I'd get a sports scholarship to college...) Like most things of this kind, there are probably exceptions, but I've been informally asking people who have RSI and carpal tunnel problems about their sports background for years and so far have found no reason to think this corrolation doesn't hold. I keep wishing someone would do a more scientific study. Meanwhile I still recommend that parents who want their kids to go into computers push them into these various sports I've enumerated and anything else that would build arm/wrist strength, including early learning to type at an early age while your bones are still growing/forming.
Stefan Ram wrote: > Recently I had a strange keyboard hardware failure, which > made the cursor-up key emit an additional scroll-lock key > event. It took me some time to find a program that was able > to disable the scroll-lock key ("Powerpro").
I would suggest to buy a new keyboard. On the other side, all my computer equipment changed since the last some 10 years, only my old Microsoft natural keyboard survived (the good old one, with the high quality keys and the normal cursor key arrangement, not in "+" form), so perhaps I would use such a program, too, if it would have such a failure :-)
Ingvar <ing...@hexapodia.net> writes: > I'm not entirely sure, I was doing an awful lot of martial arts as I > was growing up (kicking/punching, grappling, wresting, swords, rifles, > all sotrs, really) and quite a few of those have lots of wrist moment > and I certainly have RSI-similar things happening once in a while > these days.
It could be that some of these work the wrists more than others. Using it is not the same as working it. Of course, I'm being vague. But your datapoint is noted.
> I attribute it more to bad keyboards, though, since my > keyboard at home (an old IBM buckle=-spring keyboard) doesn't really > cause me any problems.
I've used a LOT of keyboards in my time and very bad posture and I seem immune on pretty much all of them.
> My keyboard at work does, though, but since I > do more "plain text" and "change/check configurations" at work and > more programming at home, it sort of balances out.
My theories are primarily about carpal tunnel, not RSI. Some people blur these, so I tend to raise them in both forums. But I believe there are in fact two phenomena, and the legit RSI thing is not caused by what I said. I have sometimes gotten RSI myself when I've done immensely repetitive things (e.g, when translating MACSYMA [~100,000 lines] from Maclisp/Zetalisp to Common Lisp, I spent a couple of DAYS in query-replace changing / to \ which was usually but not always the same operation and involved a lot of confirmation queries, and again when producing the ANSI CL standard, which began [with Steele's permission] from the CLTL sources and was slowly transformed into ANSI CL--he'd made some decision to use @foo[...] (like the typesetting language Scribe) rather than \foo{...} like normal TeX, but had made it inconsistently and I spent many days (weeks, I think), using complicated query-replace-like transformations to get it back into normal TeX notation uniformly... both of those drove my hands nuts for reasons I assume are RSI effects). But in my experience, and it sounds like yours agrees, it's generally helped by just varying the routine, stopping for breaks, etc. Works well if you have limited work or a lot of flexibility about your ordering. Not so good for poor people on assembly lines, etc.
Probably not, actually. But thanks for thinking of me. Mike McMahon did the original one with LET. Bob Kerns and I did one with LAMBDA "in response" but explicitly avoided use of backquote since McMahon had already played that game. I don't have it handy, but I'll dredge that conversation up sometime out of my files perhaps for one of the books I'm working on. It's pretty funny.
Kent M. Pitman wrote: > I have sometimes gotten RSI myself when I've done immensely repetitive > things (e.g, when translating MACSYMA [~100,000 lines] from > Maclisp/Zetalisp to Common Lisp, I spent a couple of DAYS in > query-replace changing / to \ which was usually but not always the > same operation and involved a lot of confirmation queries, and again > when producing the ANSI CL standard, which began [with Steele's > permission] from the CLTL sources and was slowly transformed into ANSI > CL--he'd made some decision to use @foo[...] (like the typesetting > language Scribe) rather than \foo{...} like normal TeX, but had made > it inconsistently and I spent many days (weeks, I think), using > complicated query-replace-like transformations to get it back into > normal TeX notation uniformly... both of those drove my hands nuts for > reasons I assume are RSI effects). But in my experience, and it sounds > like yours agrees, it's generally helped by just varying the routine, > stopping for breaks, etc. Works well if you have limited work or a lot > of flexibility about your ordering. Not so good for poor people on > assembly lines, etc.
I can recommend the use of Carbon Emacs plus a Griffin Powermate for these kinds of tasks (hitting the same key or key combinations over and over again).
I had some problems with my right arm (it hurt) which came from too much mouse usage. I got the TouchStream keyboard and after two weaks all problems were gone. You don't need force to press a key. Just a slight touch is enough. I personally got the DVORAK version of the keyboard which is (for me) much better than qwerty.
I simply lay down four fingers of the left hand in the row over the home row and press N with my right hand for "(" and a S for ")". It also has an emacs mode with many gestures plus you can extend it with your gestures. Your options are saved on the keyboard. So you could use it at home on your windows system and just plug it via usb into your linux box at work, or into the mac of your friend and it will simply work.
On Thu, 21 Jul 2005 00:57:49 +0200, André Thieme <address.good.until.2005.sep...@justmail.de> wrote: > You want to use the TouchStream LP: > http://www.fingerworks.com/
From that website:
Important note!
FingerWorks has ceased operations as a business.
Cheers, Edi.
--
Lisp is not dead, it just smells funny.
Real email: (replace (subseq "spamt...@agharta.de" 5) "edi")
>>I'd gladly pay $300 for it if it works as described. >>It's expected to hit the market in 2006. Finger crossed.
> Doesn't look very ergonomic on finger muscles etc to me. Too flat, > with not enough downward play and resistance. > I'd recomend a high quality keyboard with real springloaded switches > in the keys, not the rubber-mat stuff that is ubiquitous with nearly > all keyboards. Examples are certain models from Cherry and the old IBM > "steel-plate"-style keyboards (I don't know if IBM still ships them; > I've recently found on the net the company which made them for IBM and > they're still selling them but I forgot the URL). > And of course if you're an Emacs user, remap capslock to control, if > you're using a PC-style keyboard. Using the control in the lower left > corner is begging for cramps and strained tendons.
UniComp owns the rights to make these keyboards (bought from LexMark in the mid-90s). See http://www.pckeyboard.com. I have two of their "Customizer" keyboards (with Windows keys, in Norwegian layout), and they work just fine - the feel is not quite as good as my 1989 Model M, but it is possible that they will improve with age.
I have both a Kinesis ErgoElan and the TouchStream LP, and I don't like the TouchStream LP at all. I keep losing my position, and the problem I had got worse with the TouchStream. But then again I'm not much of a mouse user, so if your main problem is using the mouse, then it might help...
