December 2003 - Legacy - lists.gtalug.org

ZeroInstall; application installallation Nirvana ?
by legrady-bJEeYj9oJeDQT0dZR+AlfA＠public.gmane.org 13 Dec '03

13 Dec '03

It's either brilliant or there's something missing. I like the implications for an xterm-type machine. Have some important things happen on a central server, mundane stuff can run on the terminal, software which is used gets cached locally. David J Patrick wrote: > I just followed an osnews.com story and discovered ZeroInstall > (http://zero-install.sourceforge.net/) > I'm /floored/ by the implications. > Does this not look like THE most brilliant approach to software > availability ? > ..or do I need more sleep ? > djp > > -- > The Toronto Linux Users Group. Meetings: http://tlug.ss.org > TLUG requests: Linux topics, No HTML, wrap text below 80 columns > How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml > -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

ZeroInstall; application installallation Nirvana ?
by davidjpatrick-rieW9WUcm8FFJ04o6PK0Fg＠public.gmane.org 13 Dec '03

13 Dec '03

I just followed an osnews.com story and discovered ZeroInstall (http://zero-install.sourceforge.net/) I'm /floored/ by the implications. Does this not look like THE most brilliant approach to software availability ? ..or do I need more sleep ? djp -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by fraser-Txk5XLRqZ6CsTnJN9+BGXg＠public.gmane.org 13 Dec '03

13 Dec '03

On December 13, 2003 11:18 am, Peter Hiscocks wrote: > Further, I would assume this performance metric is 'speed'. It would be > interesting to test for 'effort', but that's a lot more difficult to > quantify. The other metric would be 'ease of learning', which would be > relatively easy to measure. I was mostly interested in dvorak because of it's supposed reduction in fatigue. I've just remapped my keyboard to the dvorak layout and while my typing speed is drammatically slower I'm not finding the adaptation too difficult. I've done about half of an online typing tutor and I'm finding it pretty easy. If I like it (or can stand it) for a few weeks perhaps I'll get up to speed. -- Fraser Campbell <fraser-Txk5XLRqZ6CsTnJN9+BGXg at public.gmane.org> http://www.wehave.net/ Georgetown, Ontario, Canada Debian GNU/Linux -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by henry-lqW1N6Cllo0sV2N9l4h3zg＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, 13 Dec 2003, Taavi Burns wrote: > > ...Almost all the "we should switch" > > arguments also support "switch to Maltron, not Dvorak". > > Indeed, I see that point. Given that Maltron uses the basic qwerty > letter layout... Uh, come again? It most certainly doesn't. The home row, for example, is ANISF DTHOR, with E on the left thumb and space on the right thumb. Some Maltron keyboards do come with a Qwerty mode as well, but that's explicitly a compatibility kludge. (Note, keyboards shaped vaguely like the Maltron keyboard are not necessarily Maltron keyboards.) > Question: who made the Maltron such that it's that much superior to > a Dvorak layout? Has anyone tried using the physical Maltron keyboard > with a Dvorak letter mapping? Why would you bother? Use the Maltron shape with the Maltron layout. > > the Dvorak layout; if you could shoehorn some approximation of it into a > > Qwerty-based physical layout (which I'm unsure of), the result is quite > > likely to be better than Dvorak. > > I'm saying that after looking at a picture of the Maltron keyboard, > I don't think there's any way to convert a standard keyboard short of > a hacksaw and soldering iron. The physical metaphor is wrong. Correct, but that's backwards from what I was suggesting, which is adapting the Maltron layout to the standard physical shape. I'm not sure it is reasonably possible, mind you, because of the thumb problem. > > Preferably a factor of ten. It has to be at least a factor of two to get > > people excited. 10% just isn't enough when there are major compatibility > > issues. > > Noting that the compatibility issue is a human one. The machine can > trivially remap the keyboard's logical layout. True, but so what? The human issue is a large one, actually a bigger barrier than a hardware issue would be. > On a slightly divergent note, I'm still quite perplexed as to why > having the most commonly typed letter _not_ on the home row is not > such a performance hit as one would expect. Note that the most commonly typed *key* is space, which has a whole thumb dedicated to it in either Qwerty or Dvorak, and is in the home position for that thumb in Maltron. That said, yes, E is the commonest letter and it's not in the home row, but only about one character in every ten is an E, so by itself it does not make a huge difference. And Qwerty does put it on a strong finger, and in a location where it's not a long reach. (Maltron's E on the left thumb is clearly better, mind you...) > Neverminding that > the studies you cite say that there's negligible improvement, I'm > quite curious as to why that is. Basically, because the Qwerty layout is not a bad one (for an electronic keyboard -- some of its problems, like the assignment of A and the shifts to the little fingers, loom larger on manual typewriters), and the improvements to be had by rearranging it simply aren't huge. Henry Spencer henry-lqW1N6Cllo0sV2N9l4h3zg at public.gmane.org -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by taavi-LbuTpDkqzNzXI80/IeQp7B2eb7JE58TQ＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, Dec 13, 2003 at 05:24:20PM -0500, Henry Spencer wrote: > We routinely live with *many* decisions that are known to be suboptimal > but for which it just isn't worth switching. For example, there is > general agreement that a somewhat wider railroad gauge (spacing between > the two rails) would be superior... but it's not going to happen, given > how many railyards, bridges, tunnels, etc. would have to be rebuilt. There's a funny story about why railroad tracks are as far apart as they are...it involved the width of a horse's ass... > have a limited working life anyway, Maltron keyboards would not be > expensive if made in really large numbers, and if we're going through the > pain of switching, we should go all the way to the optimal solution rather > than settling for a halfway point. Almost all the "we should switch" > arguments also support "switch to Maltron, not Dvorak". Indeed, I see that point. Given that Maltron uses the basic qwerty letter layout, it's probably easier for someone to switch . > > It's still a qwerty, though. I don't see how a logical remapping of > > a standard keyboard to some approximation of that would be of any use. > > If it's of no use, then Dvorak is of no use. The Maltron key layout was > designed with far more complete knowledge of typing ergonomics etc. than Question: who made the Maltron such that it's that much superior to a Dvorak layout? Has anyone tried using the physical Maltron keyboard with a Dvorak letter mapping? > the Dvorak layout; if you could shoehorn some approximation of it into a > Qwerty-based physical layout (which I'm unsure of), the result is quite > likely to be better than Dvorak. I'm saying that after looking at a picture of the Maltron keyboard, I don't think there's any way to convert a standard keyboard short of a hacksaw and soldering iron. The physical metaphor is wrong. > > You're saying that no switchover (to Dvoark) is acceptable. How much > > better does something have to be for it to be worth it? > > Preferably a factor of ten. It has to be at least a factor of two to get > people excited. 10% just isn't enough when there are major compatibility > issues. Noting that the compatibility issue is a human one. The machine can trivially remap the keyboard's logical layout. > > and buy anything special. Windows can be configured to switch between > > qwerty and Dvorak using ALT-LEFTSHIFT. Does it get any simpler than that? > > That's not "simple", not if you have to configure every machine in a > company to do that. There's a big difference between what's reasonable > for an organized conversion campaign, and what's reasonable for one person > with unusual needs/preferences. Most large organisations would be using some form of imaging anyway. It's trivial to ensure that the correct keyboard definition files exist, and are easily available for use. On a slightly divergent note, I'm still quite perplexed as to why having the most commonly typed letter _not_ on the home row is not such a performance hit as one would expect. Neverminding that the studies you cite say that there's negligible improvement, I'm quite curious as to why that is. -- taa /*eof*/ -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by henry-lqW1N6Cllo0sV2N9l4h3zg＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, 13 Dec 2003, Taavi Burns wrote: > > Mainly, the very high costs of conversion, including a lengthy period of > > operating both... > > Agreed. But at what point do the long-term benefits of that marginal > improvement overcome the initial cost of switching? Possibly never, if the long-term benefits are small, the assumptions they rest on are uncertain (will we still be using keyboards at all in 100 years?), and current users would get little benefit in return for the pain. We routinely live with *many* decisions that are known to be suboptimal but for which it just isn't worth switching. For example, there is general agreement that a somewhat wider railroad gauge (spacing between the two rails) would be superior... but it's not going to happen, given how many railyards, bridges, tunnels, etc. would have to be rebuilt. Note that if a keyboard switch *was* undertaken, something like the Maltron keyboard almost certainly would be the new keyboard of choice. Dvorak's only virtue, in comparison, is physical compatibility with Qwerty, and physical compatibility isn't that important in such a major change: retraining costs totally dominate equipment costs, most keyboards have a limited working life anyway, Maltron keyboards would not be expensive if made in really large numbers, and if we're going through the pain of switching, we should go all the way to the optimal solution rather than settling for a halfway point. Almost all the "we should switch" arguments also support "switch to Maltron, not Dvorak". > > You might perhaps be able to come up with a vaguely Maltron-ish remapping > > of the standard layout. The thumb areas would be tricky. > > It's still a qwerty, though. I don't see how a logical remapping of > a standard keyboard to some approximation of that would be of any use. If it's of no use, then Dvorak is of no use. The Maltron key layout was designed with far more complete knowledge of typing ergonomics etc. than the Dvorak layout; if you could shoehorn some approximation of it into a Qwerty-based physical layout (which I'm unsure of), the result is quite likely to be better than Dvorak. > > The cost of conversion, however, suggests that at most one switchover is > > acceptable. > > You're saying that no switchover (to Dvoark) is acceptable. How much > better does something have to be for it to be worth it? Preferably a factor of ten. It has to be at least a factor of two to get people excited. 10% just isn't enough when there are major compatibility issues. > The beautiful thing about the Dvorak is that you DON'T need to go out > and buy anything special. Windows can be configured to switch between > qwerty and Dvorak using ALT-LEFTSHIFT. Does it get any simpler than that? That's not "simple", not if you have to configure every machine in a company to do that. There's a big difference between what's reasonable for an organized conversion campaign, and what's reasonable for one person with unusual needs/preferences. Henry Spencer henry-lqW1N6Cllo0sV2N9l4h3zg at public.gmane.org -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by taavi-LbuTpDkqzNzXI80/IeQp7B2eb7JE58TQ＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, Dec 13, 2003 at 02:46:12PM -0500, Henry Spencer wrote: > Mainly, the very high costs of conversion, including a lengthy period of > operating both. As with a number of other things, having everybody do > things the same reasonably-good way is much more important than finding > the absolutely optimal way. Agreed. But at what point do the long-term benefits of that marginal improvement overcome the initial cost of switching? If we're still all using qwerty in 1000 years, might it not be better to deal with 10 years of confusion now, even if just for an extra 1%? Over 1000 years, 1% is a lot of work. > > what does a Maltron layout look like? Is it even physically similar to > > the keyboards we use today? > > Only very loosely. It's closer to the "natural" keyboards. On each end, > there is a sort of shallow bowl of finger keys. Below and inward from > that, tilted backward, on each side is a small array of thumb keys (which > includes assorted shifts and movement keys, the space key, and the E key). That sounds a lot like one of the Kinesis keyboards. > You might perhaps be able to come up with a vaguely Maltron-ish remapping > of the standard layout. The thumb areas would be tricky. It's still a qwerty, though. I don't see how a logical remapping of a standard keyboard to some approximation of that would be of any use. > The cost of conversion, however, suggests that at most one switchover is > acceptable. You're saying that no switchover (to Dvoark) is acceptable. How much better does something have to be for it to be worth it? > <http://www.keytools-ergonomics.co.uk/advice/onehanded.asp> concludes that > unless a one-handed typist was previously a fast two-handed typist, the > overwhelmingly best choice is one-handed typing on a standard Qwerty > keyboard, simply because it avoids needing special accommodations for that > one person. (And if he or she was a fast two-handed typist, there are > one-handed variants of Qwerty which may be preferred.) > > In any case, if you're buying a custom keyboard, there are one-handed > Maltron keyboards which are probably preferable over anything Dvorak. The beautiful thing about the Dvorak is that you DON'T need to go out and buy anything special. Windows can be configured to switch between qwerty and Dvorak using ALT-LEFTSHIFT. Does it get any simpler than that? I'm pretty sure that KDE can do things as simply as that, too, as should any other properly hotkeyed wm. Any public terminal will have the keys clearly labeled if it cannot be remapped. I do believe that someone else mentioned that he has few problems switching between using a Dvorak and qwerty layout, making the analogy to the different fingerings required by people who play multiple musical instruments. Anecdotal evidence for sure, but then I don't buy the reasons for not using a one-handed Dvorak in the case of a physical disability. Yes, I'm a geek and a musician and have an Engineering degree. I understand that I find joy in the journey at least as much as the destination, and that's not the case for most people who just want to get the job done. But if other people don't care, who will? Should I go install Windows? -- taa /*eof*/ -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

Exchange replacement, again...
by cbbrowne-HInyCGIudOg＠public.gmane.org 13 Dec '03

13 Dec '03

> At 01:04 AM 12/13/2003 -0500, Madison Kelly wrote: > > Can anyone recommend a method, how-to or general pointer to an > > application or collection of applications that would allow for providing > > similar services that Exchange does except on Linux? Obviously > > compatibility with MS Outlook would be nice however that is not a > > requirement is there is a windows client (Mozilla?) that can access these > > services. > > A Linux-based replacement for Exchange has been done. It was described in > an issue of Linux Journal within the past year. I don't remember the > specific issue off-hand. The article explains what was done and may have > links to what you need to set it up. The project site is thus: <http://kroupware.org> The server combines: OpenLDAP (to store authentication information) PostFIX (MTA) Cyrus IMAP (IMAP server) Apache (for WEBDAV, administration, and, soon, webmail) ProFTPd (for upload of busy/free info) SASL (to perform authentication) The client consists basically of KMail + Kontact; calendar management comes in somewhere; I'm not certain precisely where. (The authoritative documentation is in German, and my German is very weak...) While the current client software was written for KDE, nothing would stop someone from implementing code for the corresponding GNOME applications to have them "play well" with the pretty-standard Kroupware server components. Ditto for Mozilla. -- If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me http://cbbrowne.com/info/emacs.html Why don't sheep shrink when it rains? -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by taavi-LbuTpDkqzNzXI80/IeQp7B2eb7JE58TQ＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, Dec 13, 2003 at 04:06:25PM -0500, William Park wrote: > > > Also, Dvorak doesn't work for Vi users. Since I'm hardcore vi-guy (even > > > my shell commandline is vi-mode), I'm stuck with Qwerty. > > > > It works perfectly well. You just have to relearn the command keys. ;) > > That's what I mean. 'hjkl' keys are all over the place. Or you could probably remap them to be in the correct places. -- taa /*eof*/ -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

OT: keyboard layouts
by opengeometry-FFYn/CNdgSA＠public.gmane.org 13 Dec '03

13 Dec '03

On Sat, Dec 13, 2003 at 03:26:50AM -0500, Taavi Burns wrote: > On Sat, Dec 13, 2003 at 12:35:41AM -0500, William Park wrote: > > Typing English essay and typing computer program are a little different. > > Also, Dvorak doesn't work for Vi users. Since I'm hardcore vi-guy (even > > my shell commandline is vi-mode), I'm stuck with Qwerty. > > It works perfectly well. You just have to relearn the command keys. ;) That's what I mean. 'hjkl' keys are all over the place. -- William Park, Open Geometry Consulting, <opengeometry-FFYn/CNdgSA at public.gmane.org> Linux solution for data management and processing. -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0