In my case, they're all different generations of technology. Most of them wouldn't even be usable if it weren't for a skilled administrator hacking and patching them together.
IMHO, schools and the like would be better off with volunteers offering time to help set up and maintain these systems. And of course super cheap MS licenses (high-school educated people can't get jobs with KOffice alone)
The geeks can slam MS, go dumpster diving, and build 12 computer LANs running an assortment of OSes. There's no need to spoonfeed leading edge technical skills to those without initiative nor inspiration.
And what is wrong with writing, reading and mathematics? What can be taught on modern machines which can't be taught on ancient ones? Programming, keyboarding? or do kids need industry skills... isn't that what college is for?
One year you can be bleeding edge, but if you sit on your butt, you'll very quickly become a useless freeloader.
Security must be the most wicked example of this. Six months out of date means you're useless until you catch up.
If tech workers were unionized, a prized worker five years ago, given a guaranteed senior position through "seniority", allowed to remain stagnant would be a bafoon. Imagine taking direction from somebody who thinks of Java as some new experimental thing, UML does not exist, and fat clients are the norm.
Pilots, teachers and the like don't have these problems. They have different problems.
As for the "idiot construction worker" comment... Most tech people I've known fully respect trade workers. Who in their right mind would call a gainfully employed individual in a job with plenty of free time an idiot?!
That combined with the "you deserve more" comment has me wondering why you're trying to manipulate your audience.
You don't find it the least bit odd that the octane rating has little to do with the presence of octane?
Re:does this break the theory of relativity?
on
Stop, Light.
·
· Score: 2
I hate it when somebody says something irrefutable. And here I was trying to avoid somebody splitting hairs by telling me that pi was not a constant depending on your reference frame.
More precisely, the ratio of the circumference into the diameter of a circle varies in an accelerated reference frame regardless of the position, velocity or acceleration of the observer. That is an important distinction, I'm glad you pointed it out.
Re:does this break the theory of relativity?
on
Stop, Light.
·
· Score: 2
That makes no sense...
It was either Faraday or Maxwell... or somebody else around there which did some work which strongly indicated that c was a constant. Looking into that is on my list of things to do... this work prompted Michaelson and Morley to perform their experiments showing light moved at a constant speed in all directions.
And if you're interested in the "cosmic" speed, there were those pesky calculations involving the speed of light based on the orbits of the moons of Jupiter. This leveraged the width of the earth's orbit against Newtonian physics and the observed position of Jupiter's moons. Not horribly precise, but nothing is known to an infinate number of significant digits.
The constant 'C' is defined as the speed of light in a vacuum, as one poster here worded it so eloquently, the speed of light in different medium is due to absorption and retransmission.
But I'm only saying that 'C' is constant. I'm not saying what 'C' is. Just like PI is a constant, only known to a million or so significant digits (in an unaccelerated reference frame).
Stopping a photon is probably just some media spin.
It is a good alternate method to render data, but Microsoft (and I believe a few others recently) are the only ones who currently support it. It is almost there. I haven't done CSS development since 1.0, but I've followed the (major) developments of it on and off.. and the end results I see are analogous to the problems encountered doing pixel-perfect tables. Finding the lowest common denominator between all the browsers has more to do with hacking and experience than anything to do with standards.
But as for your response, you're telling me that it is all those pesky Netscape users who won't let you use CSS to render information... On one hand you say that standards are important, and that CSS provides clear rendering of information, but on the other hand, you're probably using tables to hack the web into a glossy magazine.
All these features provide zero improvements to the quality or readability of content... more often than not they detract from it. You're right though, CSS is a step in the right direction, but that doesn't mean that it should be embraced at the expense of giving one ill-reputed company a nearly complete stranglehold on the marketplace.
Sorry I've struck a nerve, I'm well aware of what CSS was supposed to do. The reality is that it hasn't made the web any less ugly... yet.
The end result is that people use CSS and make their documents unreadable to Netscape users. HTML worked fine... the only problem was that people couldn't stop trying to use it to do desktop publishing.
When people start adopting CSS, the same trash will start all over again, people will start adopting the equivalent of a Netscape blink tag, only it will come from Opera, Mozilla, Netscape, or IE. All to overcome some perceived problem rendering information which could just as easily be presented in plain text.
The only advantage is that then maybe people will be able to actually see the information in plain text, trimming off all this cruft introduced since HTML 2.0.
You're just ticked off that compatability issues and market forces prevented CSS from becoming what it was/is promised to be. I agree it is a shame, but ignoring the current state of the web and using CSS is going to cause readers nothing but headaches.
They work fine for small sites rendering graphics to relatively known platforms. As soon as you try to make the content accessible to a blind person, readable on a cell phone, indexable (and understandable) by a search engine, or in any other way try to decipher the content from the presentation, you run into problems.
The reason I say for small sites too is that by abstracting the content from the presentation, it makes dynamic updates very simple. You don't need a webmaster or HTML guru to put up a site which meets the corporate formatting guidelines.
Currently, clever dynamic page generation gets around that quite well, your backend can see the structure, to create indices and the like, but no machine in the rest of the world can understand what your page is trying to convey. Only humans who have a knack for deciphering magazine-like columns peppered with graphics and the like can read it.
The Java applet loads on the cooperative bank, but all the labels are screwed up... lots of buttons with "Unknown text id >10113" labels.
The Javascript on the Barclaycard site does cause problems. The instance of the browser locks up... it has to be closed. It doesn't appear to affect other sessions though. On the other hand, the site doesn't do Javascript preloads for its mouseovers... poor design.
Still, in an ideal world no script should take out a browser, and no application should be able to crash X. I've had Gnomehack crash X. It is one of my greatest complaints of "Linux stability". The OS is solid... the GUI falls on its face far too often.
I'm sure there is nothing on the page which couldn't be rendered using code which is compatible with all browsers. Web developers are obsessed with stupid things like Javascript, CSS and the like. That's where most of the gut-wrenching compatability issues come in.
The point is not so much that Microsoft has won, but that crafting a site for IE-only is nothing more than stroking one's ego by showing off cool widgets.
Turn off the crappy HTML extensions and let people read the contents of your page... assuming there is content at all.
The Quebec referrendum was just about as close, and if there were errors in counting the wishes of a separatist movement, the result could have been far worse.
IIRC it was somewhere around 49.99% vs 50.01%. The results were instantaneous. I do not know if they used similar procedures as the municipal or federal system.
A light Mozilla session vastly overshadows the memory utilization of NS4.x. Infact, it beats Lotus Notes, and other major bloatware. If you leave the process inactive for a while, and the memory utilizaiton hits 50MB or so... it is a real drag to click on an icon as everything very slowly returns from swap. I hope these are memory leaks... and if so, that they're correctable.
Does anybody who has the source code know what it taking up all that RAM?
OTOH, NS4.x and IE5 run on minimal hardware such as Windows 3.1 machines with 8M of ram (don't run Java unless you have 16 or so)
I don't see it. I see the media saying the sky is falling, and I see corporations failing to upgrade tens of thousands of machines (if not more) first because of y2k freezes, then because of hardware shortages. This year might show a turnaround... unless this silly pseudo-recession impacts the sales.
Nonetheless, I wasn't speaking specifically of PC sales. This is the strategy they've been using on the handheld platform too. Operating systems are still desparately required by corporations, as are perpetual system replacements, upgrades and license renewal.
Sony and Nintendo have oodles of power, but they have not locked in... the world?... to their plaform.
Unfortunately Microsoft has infinate resources to throw at this... so unlike Nintendo, Sega, or any other of the bread-and-butter console players, Microsoft can keep releasing new products again and again, failing until they eventually get it right. If they fail long enough, they will bundle it with free high speed Internet/WebTV, or simply a free console.
No matter how great the short term loss, long term success is all they care about.
I hesitate to recommend a book which I haven't actually read. There have been plenty of books which looked good, but when push came to shove, the detailed info was simply incorrect.
This is one of those Exam Prep ones. Big tacky and red. "General Linux 1", Dee-Ann LeBlanc from Coriolis. I haven't given it a read cover-to-cover so I can only recommend pulling it off a bookshelf and skimming it for now... but it looks quite solid. ISBN 1-57610-567-9.
I picked up one of the LPI books yesterday. It appears to be one of the most thorough books covering all the basic stuff you need to know to work with a Linux box. It does cover everything you said, and in a fair bit of depth. Things I've done and I can barely recall how to do, like configuring ppp to answer dial up lines. It also goes into depth on the protocols to do things like routing. User management goes into password strength, shadowing etc. The book delves into IP chains, SocksV5, etc. And this is for the the 101 exam. The last of 13 chapters is about configuring X.
Other topics include package managemnet, recompiling the kernel, shell scripting, regular expressions, bind, apache, wuftpd, sendmail, printing and managment of network print queues, disabling unused network services, quotas, logfiles, tracking security updates online, hardware configuration and modules.
Honestly the book is a course outline for all the stuff I've been meaning to solidify my knowledge on. I won't really know how thorough it is until I give it a read, but skimming over it, it looks impressive.
I've been a linux sysadmin before, and I'm ashamed to admit that I didn't know half this stuff, but that was some time ago. I have known other admins who wouldn't know a fraction of this stuff either. Most can figure it out on the job, and yes, that is a valuable skill, but if the certification holds up, it is a piece of paper saying that you've done just about everything on a linux box at least once before... and committed it to memory.
I gave it a very boring test on Word97... Using simple text, there are no problems.
I created a doc with the body of "This is confidential information" then the alphabet in lower case, then the same thing, except that it was not confidential information and the alphabet in upper case. I saved it, ran strings on it, and I saw all of my text (and the full name as registered in the product!)
I then selected the text with the mouse, hit delete, and checked strings.
All the information was gone... except that Word appeared to have taken the words "This is confidential information" as a title, and kept it in the document. The lower case alphabet was gone though. Of course my name was still there.
When I closed Word and reopened the second document using the run history, the undo buffer was empty.
I've been toying about with the idea of a standard API to open, read and write configuration files. Then people could use the architecture most suitable for thier particular situation, be it a memory resident registry-like system, or be it configuration files scattered all over the drive. The only difference is that people could make choices concerning performance/volatility tradeoffs.
A simple example of how this could be done would be something akin to a symbolic link which is really a dynamic file. You can open it, read from it, and hopefully even write to it in a limited capacity. The libary would translate these actions to and from the desired configuration changes... perhaps smiting you for syntax errors in the file layout (send the program an error that the file is read-only or something)
A bit like/proc only scattered all over the drive. They look like a bunch of text files, you can write to them, but they're not what they appear.
I've tried one of those simulators, I don't recall the name of the program, but I think it was written by Todd Lammle. Very poor. (To his credit, he writes better books.) It was not the kind of thing you could hack around with. I suppose my greatest complaint is that very simple broken configurations would actually work when they're not supposed to. Very simple problems, like setting up static routes in only one direction, and having reply packets make it through. Despite all these bugs, it costs roughly $300. I *ahem* evaluated the product before deciding that it didn't perform as advertised.
I was lucky because in Toronto we have a study group. They have a router lab available online for Toronto-area residents. It is not cost effective nor all that rewarding to allow people to book time on the routers globally... but you might want to search the web.. you might find something.
Your other option as somebody recommended is to build your own lab. Most people studying for their CCNE do this. Writing the exams is the only reason I could think that somebody would ask specifically for an IOS emulator.
I did a quick search and there does appear to be a Cisco IOS emulator for modifying ipchains rules using IOS commands. It looks pretty young. It didn't exist when I was studying for my CCNA
http://www.tarball.net/cish/
I think at some subconsious level, my brain has become hard-wired to place a 'U' after a 'Q'. I also have this problem where I ran out of sugar. Without sugar, I couldn't have coffee. Without coffee, I couldn't go out to get sugar. So I booted up my computer and logged onto Slashdot. I suppose that was my first mistake...
The first article, despite being very lengthy, doesn't appear to acually say anything. It goes into great detail about how QUERTY is the evolutionary survivor of nineteenth century typewriters... which in the Dvorak argument is completely meaningless. The Dvorak myth is that QUERTY is superior for manual typewriters.
I had to skim a few drawn out repeating paragraphs, but I couldn't find anything which actually stated why QUERTY was infact created, only that there was lots of evidence that Dvorak himself spread the rumour that his patented keyboard was better.
The author also does a horrid job trying to shoot down the lock-in argument using the DOS/Windows analogy. CP/M is source compatible with DOS 1.0, and almost all DOS apps run under modern versions of Windows. The whole failure of OS/2 was centred around Windows compatability and the market forces associated with promoting the competition's platform.
He does appear to enjoy shooting down lots of sources which people believe credible... moreso than he appeared to be interested in the merits and faults of the various keyboards. But he does have some interesting points.
From what the article claims, I think your asessment is accurate: "Mostly an urban myth popularized by Dvorak." But he doesn't do squat to actually refute the argument, only to refute its historical accuracy.
Everybody is looking for the next revolutionary design to replace WIMP. It is funny how all the text you're reading is on a two-dimensional plane, and the only other common means of communication is voice.
A boring unintrusive UI is very good. IMHO the ultimate UI is in dedicated physical devices. Rendering them in VR is just perpetuating the personal computer beyond what it really needs to be... in other words for the people trying to shatter the mould of thinking in WIMP, they're trapped in the mould of the PC.
Slashdot Mirror
In my case, they're all different generations of technology. Most of them wouldn't even be usable if it weren't for a skilled administrator hacking and patching them together.
IMHO, schools and the like would be better off with volunteers offering time to help set up and maintain these systems. And of course super cheap MS licenses (high-school educated people can't get jobs with KOffice alone)
The geeks can slam MS, go dumpster diving, and build 12 computer LANs running an assortment of OSes. There's no need to spoonfeed leading edge technical skills to those without initiative nor inspiration.
And what is wrong with writing, reading and mathematics? What can be taught on modern machines which can't be taught on ancient ones? Programming, keyboarding? or do kids need industry skills... isn't that what college is for?
That's a bargain. A QT development license is $1550. I wonder how much of a cut Borland is getting of that $999.
http://www.trolltech.com/products/purchase/pricing .html
One year you can be bleeding edge, but if you sit on your butt, you'll very quickly become a useless freeloader.
Security must be the most wicked example of this. Six months out of date means you're useless until you catch up.
If tech workers were unionized, a prized worker five years ago, given a guaranteed senior position through "seniority", allowed to remain stagnant would be a bafoon. Imagine taking direction from somebody who thinks of Java as some new experimental thing, UML does not exist, and fat clients are the norm.
Pilots, teachers and the like don't have these problems. They have different problems.
As for the "idiot construction worker" comment... Most tech people I've known fully respect trade workers. Who in their right mind would call a gainfully employed individual in a job with plenty of free time an idiot?!
That combined with the "you deserve more" comment has me wondering why you're trying to manipulate your audience.
You don't find it the least bit odd that the octane rating has little to do with the presence of octane?
I hate it when somebody says something irrefutable. And here I was trying to avoid somebody splitting hairs by telling me that pi was not a constant depending on your reference frame.
More precisely, the ratio of the circumference into the diameter of a circle varies in an accelerated reference frame regardless of the position, velocity or acceleration of the observer. That is an important distinction, I'm glad you pointed it out.
That makes no sense...
It was either Faraday or Maxwell... or somebody else around there which did some work which strongly indicated that c was a constant. Looking into that is on my list of things to do... this work prompted Michaelson and Morley to perform their experiments showing light moved at a constant speed in all directions.
And if you're interested in the "cosmic" speed, there were those pesky calculations involving the speed of light based on the orbits of the moons of Jupiter. This leveraged the width of the earth's orbit against Newtonian physics and the observed position of Jupiter's moons. Not horribly precise, but nothing is known to an infinate number of significant digits.
The constant 'C' is defined as the speed of light in a vacuum, as one poster here worded it so eloquently, the speed of light in different medium is due to absorption and retransmission.
But I'm only saying that 'C' is constant. I'm not saying what 'C' is. Just like PI is a constant, only known to a million or so significant digits (in an unaccelerated reference frame).
Stopping a photon is probably just some media spin.
It is a good alternate method to render data, but Microsoft (and I believe a few others recently) are the only ones who currently support it. It is almost there. I haven't done CSS development since 1.0, but I've followed the (major) developments of it on and off.. and the end results I see are analogous to the problems encountered doing pixel-perfect tables. Finding the lowest common denominator between all the browsers has more to do with hacking and experience than anything to do with standards.
But as for your response, you're telling me that it is all those pesky Netscape users who won't let you use CSS to render information... On one hand you say that standards are important, and that CSS provides clear rendering of information, but on the other hand, you're probably using tables to hack the web into a glossy magazine.
All these features provide zero improvements to the quality or readability of content... more often than not they detract from it. You're right though, CSS is a step in the right direction, but that doesn't mean that it should be embraced at the expense of giving one ill-reputed company a nearly complete stranglehold on the marketplace.
Sorry I've struck a nerve, I'm well aware of what CSS was supposed to do. The reality is that it hasn't made the web any less ugly... yet.
The end result is that people use CSS and make their documents unreadable to Netscape users. HTML worked fine... the only problem was that people couldn't stop trying to use it to do desktop publishing.
When people start adopting CSS, the same trash will start all over again, people will start adopting the equivalent of a Netscape blink tag, only it will come from Opera, Mozilla, Netscape, or IE. All to overcome some perceived problem rendering information which could just as easily be presented in plain text.
The only advantage is that then maybe people will be able to actually see the information in plain text, trimming off all this cruft introduced since HTML 2.0.
You're just ticked off that compatability issues and market forces prevented CSS from becoming what it was/is promised to be. I agree it is a shame, but ignoring the current state of the web and using CSS is going to cause readers nothing but headaches.
They work fine for small sites rendering graphics to relatively known platforms. As soon as you try to make the content accessible to a blind person, readable on a cell phone, indexable (and understandable) by a search engine, or in any other way try to decipher the content from the presentation, you run into problems.
The reason I say for small sites too is that by abstracting the content from the presentation, it makes dynamic updates very simple. You don't need a webmaster or HTML guru to put up a site which meets the corporate formatting guidelines.
Currently, clever dynamic page generation gets around that quite well, your backend can see the structure, to create indices and the like, but no machine in the rest of the world can understand what your page is trying to convey. Only humans who have a knack for deciphering magazine-like columns peppered with graphics and the like can read it.
I got some better results...
The Java applet loads on the cooperative bank, but all the labels are screwed up... lots of buttons with "Unknown text id >10113" labels.
The Javascript on the Barclaycard site does cause problems. The instance of the browser locks up... it has to be closed. It doesn't appear to affect other sessions though. On the other hand, the site doesn't do Javascript preloads for its mouseovers... poor design.
Still, in an ideal world no script should take out a browser, and no application should be able to crash X. I've had Gnomehack crash X. It is one of my greatest complaints of "Linux stability". The OS is solid... the GUI falls on its face far too often.
BTW, I'm using Win98 on this post... Mozilla 0.6
I'm sure there is nothing on the page which couldn't be rendered using code which is compatible with all browsers. Web developers are obsessed with stupid things like Javascript, CSS and the like. That's where most of the gut-wrenching compatability issues come in.
The point is not so much that Microsoft has won, but that crafting a site for IE-only is nothing more than stroking one's ego by showing off cool widgets.
Turn off the crappy HTML extensions and let people read the contents of your page... assuming there is content at all.
The Quebec referrendum was just about as close, and if there were errors in counting the wishes of a separatist movement, the result could have been far worse.
IIRC it was somewhere around 49.99% vs 50.01%. The results were instantaneous. I do not know if they used similar procedures as the municipal or federal system.
A light Mozilla session vastly overshadows the memory utilization of NS4.x. Infact, it beats Lotus Notes, and other major bloatware. If you leave the process inactive for a while, and the memory utilizaiton hits 50MB or so... it is a real drag to click on an icon as everything very slowly returns from swap. I hope these are memory leaks... and if so, that they're correctable.
Does anybody who has the source code know what it taking up all that RAM?
OTOH, NS4.x and IE5 run on minimal hardware such as Windows 3.1 machines with 8M of ram (don't run Java unless you have 16 or so)
muLinux was able to cram X on a floppy.
http://mulinux.nevalabs.org/
I don't see it. I see the media saying the sky is falling, and I see corporations failing to upgrade tens of thousands of machines (if not more) first because of y2k freezes, then because of hardware shortages. This year might show a turnaround... unless this silly pseudo-recession impacts the sales.
Nonetheless, I wasn't speaking specifically of PC sales. This is the strategy they've been using on the handheld platform too. Operating systems are still desparately required by corporations, as are perpetual system replacements, upgrades and license renewal.
Sony and Nintendo have oodles of power, but they have not locked in... the world?... to their plaform.
What do you mean by 'Culture of Economy?'
Unfortunately Microsoft has infinate resources to throw at this... so unlike Nintendo, Sega, or any other of the bread-and-butter console players, Microsoft can keep releasing new products again and again, failing until they eventually get it right. If they fail long enough, they will bundle it with free high speed Internet/WebTV, or simply a free console.
No matter how great the short term loss, long term success is all they care about.
I hesitate to recommend a book which I haven't actually read. There have been plenty of books which looked good, but when push came to shove, the detailed info was simply incorrect.
This is one of those Exam Prep ones. Big tacky and red. "General Linux 1", Dee-Ann LeBlanc from Coriolis. I haven't given it a read cover-to-cover so I can only recommend pulling it off a bookshelf and skimming it for now... but it looks quite solid. ISBN 1-57610-567-9.
I picked up one of the LPI books yesterday. It appears to be one of the most thorough books covering all the basic stuff you need to know to work with a Linux box. It does cover everything you said, and in a fair bit of depth. Things I've done and I can barely recall how to do, like configuring ppp to answer dial up lines. It also goes into depth on the protocols to do things like routing. User management goes into password strength, shadowing etc. The book delves into IP chains, SocksV5, etc. And this is for the the 101 exam. The last of 13 chapters is about configuring X.
Other topics include package managemnet, recompiling the kernel, shell scripting, regular expressions, bind, apache, wuftpd, sendmail, printing and managment of network print queues, disabling unused network services, quotas, logfiles, tracking security updates online, hardware configuration and modules.
Honestly the book is a course outline for all the stuff I've been meaning to solidify my knowledge on. I won't really know how thorough it is until I give it a read, but skimming over it, it looks impressive.
I've been a linux sysadmin before, and I'm ashamed to admit that I didn't know half this stuff, but that was some time ago. I have known other admins who wouldn't know a fraction of this stuff either. Most can figure it out on the job, and yes, that is a valuable skill, but if the certification holds up, it is a piece of paper saying that you've done just about everything on a linux box at least once before... and committed it to memory.
I gave it a very boring test on Word97... Using simple text, there are no problems.
I created a doc with the body of "This is confidential information" then the alphabet in lower case, then the same thing, except that it was not confidential information and the alphabet in upper case. I saved it, ran strings on it, and I saw all of my text (and the full name as registered in the product!)
I then selected the text with the mouse, hit delete, and checked strings.
All the information was gone... except that Word appeared to have taken the words "This is confidential information" as a title, and kept it in the document. The lower case alphabet was gone though. Of course my name was still there.
When I closed Word and reopened the second document using the run history, the undo buffer was empty.
I've been toying about with the idea of a standard API to open, read and write configuration files. Then people could use the architecture most suitable for thier particular situation, be it a memory resident registry-like system, or be it configuration files scattered all over the drive. The only difference is that people could make choices concerning performance/volatility tradeoffs.
A simple example of how this could be done would be something akin to a symbolic link which is really a dynamic file. You can open it, read from it, and hopefully even write to it in a limited capacity. The libary would translate these actions to and from the desired configuration changes... perhaps smiting you for syntax errors in the file layout (send the program an error that the file is read-only or something)
A bit like /proc only scattered all over the drive. They look like a bunch of text files, you can write to them, but they're not what they appear.
I've tried one of those simulators, I don't recall the name of the program, but I think it was written by Todd Lammle. Very poor. (To his credit, he writes better books.) It was not the kind of thing you could hack around with. I suppose my greatest complaint is that very simple broken configurations would actually work when they're not supposed to. Very simple problems, like setting up static routes in only one direction, and having reply packets make it through. Despite all these bugs, it costs roughly $300. I *ahem* evaluated the product before deciding that it didn't perform as advertised.
I was lucky because in Toronto we have a study group. They have a router lab available online for Toronto-area residents. It is not cost effective nor all that rewarding to allow people to book time on the routers globally... but you might want to search the web.. you might find something.
Your other option as somebody recommended is to build your own lab. Most people studying for their CCNE do this. Writing the exams is the only reason I could think that somebody would ask specifically for an IOS emulator.
I did a quick search and there does appear to be a Cisco IOS emulator for modifying ipchains rules using IOS commands. It looks pretty young. It didn't exist when I was studying for my CCNA http://www.tarball.net/cish/
I should probably give it a try some day.
I think at some subconsious level, my brain has become hard-wired to place a 'U' after a 'Q'. I also have this problem where I ran out of sugar. Without sugar, I couldn't have coffee. Without coffee, I couldn't go out to get sugar. So I booted up my computer and logged onto Slashdot. I suppose that was my first mistake...
The first article, despite being very lengthy, doesn't appear to acually say anything. It goes into great detail about how QUERTY is the evolutionary survivor of nineteenth century typewriters... which in the Dvorak argument is completely meaningless. The Dvorak myth is that QUERTY is superior for manual typewriters.
I had to skim a few drawn out repeating paragraphs, but I couldn't find anything which actually stated why QUERTY was infact created, only that there was lots of evidence that Dvorak himself spread the rumour that his patented keyboard was better.
The author also does a horrid job trying to shoot down the lock-in argument using the DOS/Windows analogy. CP/M is source compatible with DOS 1.0, and almost all DOS apps run under modern versions of Windows. The whole failure of OS/2 was centred around Windows compatability and the market forces associated with promoting the competition's platform.
He does appear to enjoy shooting down lots of sources which people believe credible... moreso than he appeared to be interested in the merits and faults of the various keyboards. But he does have some interesting points.
From what the article claims, I think your asessment is accurate: "Mostly an urban myth popularized by Dvorak." But he doesn't do squat to actually refute the argument, only to refute its historical accuracy.
Everybody is looking for the next revolutionary design to replace WIMP. It is funny how all the text you're reading is on a two-dimensional plane, and the only other common means of communication is voice.
A boring unintrusive UI is very good. IMHO the ultimate UI is in dedicated physical devices. Rendering them in VR is just perpetuating the personal computer beyond what it really needs to be... in other words for the people trying to shatter the mould of thinking in WIMP, they're trapped in the mould of the PC.
And X fonts will look good too...