I've always thought of my driving as being the first line of defense, and my seatbelt as being the second line, there in the case that the first fails. Ideally there shouldn't be any accidents (the software shouldn't self destruct when reading malformed data) but it doesn't hurt to wear your seatbelt, just in case (firewall).
I'm leaving airbags out of this entirely, because studies of them don't seem to show enough improvement to have this explosive device installed in the cabin of the vehicle. Especially since older models (such as the ones I have) can injure adults and kill children.
Getting really off topic - Has anybody ever thought of putting high performance seatbelts in passenger vehicles? You know, extra padding, double belts/belt webs for better weight distribution, etc, like the type used in racing cars?
XP definitely doesn't make shortcuts to the desktop. I'm pretty darned certain 98 didn't either; it just made shortcuts of executable files. _That_ was definitely annoying, but I'm perfectly fine with the filesystem logic.
I'm definitely running Debian Testing (Sarge, estimated release number 3.1), using the drivers in the kernel tree for 2.6. I use a KDE desktop too. This dual boots with XP using the nice little _text menu_ mode of LILO, as I have a strong dislike for graphical boot menus on a PC. The AGP GART stuff seems to work fine, but the nForce ethernet controller (really some sort of realtek 8100; realtek has a horrible reputation for NICs. Their leadership position is only due to the cheapness of their products) drivers that come with kernel 2.4 and 2.6 seem rather flakey.
Windows hasn't always had a predictable UI (and IMHO still hasn't). Case in point: drag n drop a file between 2 folders on the same disk: it moves the file, do the same between 2 disks: it copys the file, do the same to the desktop: it links the file. This is unpredictable in the extreem.
I've never seen Windows make a shortcut on your desktop with an ordinary data file. In the old days, the default action for moving an _executable_ file was to create a shortcut. And moving/copying files in that manner has been standard on the Amiga platform for 20 years and is consistent with file system logic. (Consider what Unix variants do with 'mv' - it's simply performing a relinking that's like a rename on the same filesystem, and a copy-and-delete (move) across different filesystems). Granted this might be a bit confusing to a novice/very casual computer user, but it shouldn't be too hard if one reads the fine manual.
Even if a user finds this confusing after RTFM, just have them drag with the right mousebutton. The context menu erases all doubt.
(For the record, I've not really used Windows before Windows 95, I found the 3.x series to be hideous and I was completely non-PC before 1990. I suspect Win3.x had a lot of GUI infelicities. None of my comments apply to pre-95 series Windows)
Oh, off on a tangent, one thing that has always annoyed me about post-Amiga OSes is that they do NOT seem to support what I call 'Type Behind'. Since most of these OSes either foreground a window when you click on it, or worse yet, when you hover your mouse over it, you must shuffle the windows on the screen if you want to type in one window while reading from another. Or click back and forth a lot. Or copy the information into the other window (often not possible with images/diagrams). AmigaOS doesn't automatically bring the selected window forward (unless you configure it to; see the ClickToFront commodity, I use that with the ALT key modifier to bring a window forwards), which lets you use touch typing on an input box / text editing area while viewing a document/etc over it.
PS. Running Debian Testing on this machine (2500XP/ATI 9600) is a sluggish and unresponsive GUI experience, compared to XP. Further, the driver for the nForce RealTek 8100 network adapter is rather crappy. Mind you, I find XP to be sort of sluggish and unresponsive too, just to a lesser extent. I like Linux-console-on-AthlonXP/P4 or AmigaOS on 060-with-video-accellerator type of responsiveness, myself.
If Enterprise is to be saved, the fans need to make the clear case to Paramount that they want it to continue, without simply "donating" money to the studio. They need to convince the execs that the fan base is out there and they will be able to make money from advertising on the show. But I sure as hell ain't gonna donate money to a company that's practically ripping its fans off at the DVD store after the show!
Isn't throwing lots of money at someone a really, really persuasive way of making a convincing case? I know that would work on me!;p Mind you, Paramount should have already responded with three mil being offered. How many other shows have that much donation? How many have ANY donation at all?
Um, isn't HyperTransport some sort of I/O bus, like PCI/AGP/PCI Express? You know, device-to-CPU/chip-to-chip type stuff? I don't think it's actually used for memory access, is it?
In any case, it's still going to be limited by the fact that even top grade DDR is still many, many times slower than what the CPU can handle. (In applications that are tied to memory performance, anyways)
Actually, the impact of memory latency and bandwidth on actual performance varies greatly with the type of application you're running.
Some of my software has been very efficient in terms of CPU cycles, but dealt with a vast amount of data and thus was tied to memory latency/bandwidth. Those projects would run at almost the same speed on a Pentium 166 MMX with SDRAM as on a P2-350. Some other projects were far more cycle-oriented but only used registers, and didn't care at all about memory performance. These projects showed a dramatic improvement on the P2.
Imagine, as an example, a function that increments or inverts each byte in a large array, versus one that iterates a variable or two. The former will rely on memory bandwidth, whereas the latter will end up with most or all of it's variables in register (let alone cache) and thus rely on the CPU's overall internal performance.
Actaully, the zero/NULL thing is just an abstraction in C.
When you say something like "char *ptr; if(ptr==NULL)..." or such, the compiler notices that you're comparing a pointer to the value zero, and will then generate machine code that checks it against the machine/OS specific invalid address value, rather than an actual comparison to zero.
Note - This physical null value happens to BE zero on most machines/OSes, but on some others it's weird stuff like $3F:FFFFFFFF.
Probably. But I'm using Debian, which is about to move to 2.4 once it's considered stable. Heh.
Mind you, that's a little extreme. I think they could have moved to 2.4 once it reached around.22 or so.
I still wouldn't move the main distro kernel to 2.6 yet though. That's great for an option, but I'd rather just stick with 2.4 for any server type stuff. (I do use 2.6 on my Linux-based workstations, but they're all behind a firewall with no forwarding and all my data files are stored on a 2.4-based file server)
But people that say that GTA doesn't cause violence are missing the bigger picture that it allows people to discover just how little coverage the police have in real life. That's probably not a good thing for anybody
Ahh, security through obscurity. That's great! Everybody doesn't know that the security doesn't really exist, so it works as if it does exist! Wonderful! Superb! Excellent!
- driving at triple digits on public roads is 'fun'
Huh? I do this all the time. You've written it like as if it's wrong or something.
In fact, the speed limit here IS three digits on some public roads. Not that ANYBODY actaully drives under or at the limit, anyways. The police are usually doing the limit+20km/h themselves when there's no emergency.
Oh, you must be using that legacy measuring system. Leagues per fortnight, hogsheads per rods, etc. Tisk tisk.
"Properly distributed" is a sort of whacky thing to say.
That could be applied to anything. Here's an example: Four inches of high quality sharpened steel, properly distributed (amongst the ribcages of the world), would kill EVERY LAST PERSON. Now imagine what the blade from a wind turbine could do.
Utter lunacy. Cassini probe protesters should probably try smoking -less- than eight blunts at a time.:p
I thought one of the big ideas of Perl was that it was more or less typeless.
It will automagically convert between types as needed.
(Disclaimer: I don't have much object-oriented Perl experience, and nor do I want to have it;p. The OO parts of perl might change this typelessness somewhat... )
Don't worry, we've extracted samples of your DNA, and tagged you with several redundant (classified) tracking devices that will track you anywhere in, under, or above the Earth. We've also implanted your brain with devices that can monitor your thoughts, as well as disable you should you become violent or make fun of the President.
We've installed keystroke loggers and patched your I/O stream functions, and replaced your copy of lynx, wget and links with an especially insecure version of IE running in text emulation mode.
We have all the private keys for all your encryption protocols, and have added special extra backdoor features to all protocols as added insurance. Most of these backdoors have the password or pass-phrase of 'password' and/or all-zero hex keys.
Our next plan will be to abduct you in the middle of the night, and place you in a simulated reality where you can do no harm. Once you commit your terrorist acts there, we can proceed to arrest, detain, question, and dissect you.
Yours Truly,
The Central Intelligence Agency.
PS. Don't give us any 'jurisdiction' crap. This is OUR planet. You are only here until we arrest you. And stop making 'the finger' at that camera. That's rude to our operators.
Gawd, what awful applications. I always hate when that happens.
My own code for Windows platforms is always pure Win32 (with DirectX or OpenGL for graphical things), and will run on Win95 through to XP SP2.
The trouble the other coders get into all starts when they start using auxillary components in Windows, like IE or the goddamned MFC library or such. (This is also where many of the security holes come from)
I'm pretty sure Debian runs a vanilla kernel as well.
I usually roll my own by hand on my Debian boxes, but there's a package that lets you build a Debian-kernel-binary-package from a vanilla kernel source tree. It never said anything about requiring special patches or whatnot.
Also, my self compiled, purely stock kernels don't seem to have any problems in Debian.
> If you didn't buy the Compaq, how do you know it still runs?
A friend got one. We had a supplier in the office who was selling them. He bought that, and I bought the Thudpad. The Compaq works fine, doesn't have any driver or IRQ issues, and is still running on the original stock battery. It's quite old now, too, been running for the past 5 years.
http://www3.sympatico.ca/milnoc/TP600Battery/
I find it funny that this comes up in the first page in a search for "Thinkpad 600". Not 'Thinkpad +"battery problem"'.. Just 600. Sort of funny that the laptop's PRIMARY SYSTEM DOES NOT WORK. (There was also a slashdot article at one point in the past about this very problem)
I mean, a laptop without a battery is just a low performance desktop in a small, badly designed package.
That's interesting, but it kills some of the advantage of the usual keys...
C / Copy. This one is obvious. X / Cut. It's right next to copy, and X is a symbol for deletion. V / Paste. V is the same shape as that mark used to insert text.
Those three keys are standard for a lot of OSes - Macs, as mentioned before, Windows, and Amigas, to name a few (Amigas use left-Amiga key, which is rougly where the 'windows' key is on a PC keyboard. Actually, the 104 key PC keyboard is much closer to the old Amiga keyboard than the 101 key is, but it's still missing the HELP button)
Try riding a motorbike - the clutch is hand-operated and the gearshifter operated by the foot. It works very well.
Actually, ditto for standard transmission automobiles - the clutch is operated by the foot, and the shifter by the hand.
These all have to be in proper order, where milliseconds do matter. I never have a problem. Naturally, different people have different capabilities - while most of my friends can shift properly, there's one who always manages to stall...
> Btw: How many "laptop problems" did you have with ThinkPads?!
Oh, I dunno.. it destroyed $300 batteries regularly, it had too many IRQs for all devices to be enabled at once, it had this ugly little cli--er, eraser thingy in the middle of the keyboard which was useless and required adjustment every few days, the BIOS setup looked like some 1983 CGA game for retarded five year olds, and was about as useful, it was twice the size and half the power as the equivalent Toshiba model of the time... Oh yah, and the drivers which when downloaded, had to be extracted onto a _floppy_ disk.
IBM Thinkpad 600E U2426. All the crap you never wanted, and then some. (Oh yah, the model number is obtuse and there are a trillion useless variants to this useless heap of garbage)
I should have bought that Compaq instead. It was the same price, and still runs to this day.
Oh well. At least IBM still offers downloads on legacy products like that, even if they are really stupid to install and highly broken.
The 'highly broken' part is a universal standard for the PC industry anyways though, which IBM created with that crappy machine they threw together in a drunken binge on a weekend in 1981 -- Unsharable interrupts? Static OS structures? Segmented, 20-bit memory? Processor-driven I/O? How could you do this with a motto of "Think"?? Or maybe that's a typo, maybe it's.. DRINK??
Remember kids, alcohol, like smoking, is _not cool_.
Well, you can apply the physics-2.6.11-st (Star Trek/Space Warp) patch, or the physics-2.6.11-sw (Star Wars/Hyperspace Jump) patch, or avoid stars entirely and apply the physics-2.6.11-b5 (Babylon 5/Hyperspace Gateway) patch, or a myriad of other alternative patches.
If you want to run a vanilla physics kernel, however, you are out of luck for now.
Some highlights of the major Star* patches: Star Trek: - Later versions of the patch allow you to make up particles to overcome problems and rewrite physics entirely! - Most versions of the patch allow you to adjust speed to a desirable setting. Need to go to the center of the galaxy in one movie, when it would otherwise take about thirty years in a series? Do "echo 9999.9 >/plotholes/speed/warp-factor-adjust" to increase speeds! - Early versions allow you to save the galaxy (well, the Federation) without huge armadas of ships - just get into a fistfight! (Warning: your Captain's shirt might get torn) - And much more!
Star Wars: - Allows you to convert distance into time and vice versa! (supplemental patch allows you to make up all sorts of weird explanations for it too; see first point for Star Trek patch) - Makes light a lot faster for ships, and slower for lasers! ("Jump to lightspeed!") - And much more!
If the format is open, then an open sourced solution could be developed to deal with any issues with the proprietary software.
If the file formats used are truly open (as in a decent standard that's well documented and actually works as it's described, which allows other applications to read and write in the same format without legal encumbrance), then the customer can take their data and have a new application or data converter written, allowing them to easily migrate to a new platform.
That addresses the following issues thusly:
>Other valid complaints are the per-seat costs, upgrade costs
These are limited by the fact that the customer can now walk away. Something they cannot easily do if they were locked into Word's *.doc format or such. If the proprietary vendor tried strong-arm tactics or charged too much, the customer could simply write, or have written, a replacement application (or a file converter to convert the data files to a format used by another application).
This effectively caps the maximum cost that the vendor can impose. Company X cannot charge a billion dollars for a given product if Company Y can offer the same thing for $59.95 that works on the same data, or if Open Source Programming Group Z offers a totally free version of said program.
Also, the bigger the vendor, and the more pervasive their product, the more likely some open source group is to pick up the challenge of making an open sourced solution which works with these same data files.
> limited effectiveness of outside support,
Well, only with regards to the application itself. The data, properly documented and open, could easily be handled by any competent programmer.
> architecture lock-in
What architecture lock in? Vendor screwing you over? Take your data and walk away. Have someone else write a replacement. Write it yourself! Use Y or Z's product! Suddenly Corp X decides that the 3.2 update will NOT cost a trillion dollars, afterall.
>and a slow, costly route to get bug-fixes and new features implemented.
Bug fixes to the original application would depend on the vendor. New features, however, could be implemented by utilties that work with the data created by the application. And if the original application was too buggy to work with at all, why was it purchased to begin with?
Open sourced solutions can be costly to bugfix as well. If your package is abandoned, you'd have to _hire_ people to fix it.
> And, of course, the threat of being left out-in-the-cold if the company stops offering the proprietary program (or if said company collapses).
How do you figure this? Take the data and _WALK AWAY_. This would be no different if an open sourced group collapsed or grew bored of a package.
---
Basically, having open file formats is like a foot in the door for open source software: If free/open source software can work with the same data, and be better and cheaper, then why wouldn't some conversion happen? It also takes away one of the tools in the arsenal of big corporate evils: the hostage-holding of your user-created data.
This could also revitalize a rather stagnant commercial software industry by forcing more interoperability.
What, can't open sourced software compete then? Is it like some big, free version of Microsoft? Forced to rely on legislation to get it's place in the world? I'd hope not!
I'd say that it's probably better to allow for free and proprietary to compete openly, than to have one side legislated over the other. It could easily go the other way, you know. "All government agencies must use Microsoft(TM)(R)(C)(Patent Pending)(TM)(C)(R) software AT ANY COST!".
How could this not be win-win?
This of course, all hinges on the definition of an 'open' format:
1. It must be readable and writable by other applications or utilities withou
Slashdot Mirror
I've always thought of my driving as being the first line of defense, and my seatbelt as being the second line, there in the case that the first fails. Ideally there shouldn't be any accidents (the software shouldn't self destruct when reading malformed data) but it doesn't hurt to wear your seatbelt, just in case (firewall).
I'm leaving airbags out of this entirely, because studies of them don't seem to show enough improvement to have this explosive device installed in the cabin of the vehicle. Especially since older models (such as the ones I have) can injure adults and kill children.
Getting really off topic - Has anybody ever thought of putting high performance seatbelts in passenger vehicles? You know, extra padding, double belts/belt webs for better weight distribution, etc, like the type used in racing cars?
Thank you, I'm definitely going to get this Nail utility. I've missed that feature for a long time now.
XP definitely doesn't make shortcuts to the desktop. I'm pretty darned certain 98 didn't either; it just made shortcuts of executable files. _That_ was definitely annoying, but I'm perfectly fine with the filesystem logic.
I'm definitely running Debian Testing (Sarge, estimated release number 3.1), using the drivers in the kernel tree for 2.6. I use a KDE desktop too. This dual boots with XP using the nice little _text menu_ mode of LILO, as I have a strong dislike for graphical boot menus on a PC. The AGP GART stuff seems to work fine, but the nForce ethernet controller (really some sort of realtek 8100; realtek has a horrible reputation for NICs. Their leadership position is only due to the cheapness of their products) drivers that come with kernel 2.4 and 2.6 seem rather flakey.
I've never seen Windows make a shortcut on your desktop with an ordinary data file. In the old days, the default action for moving an _executable_ file was to create a shortcut. And moving/copying files in that manner has been standard on the Amiga platform for 20 years and is consistent with file system logic. (Consider what Unix variants do with 'mv' - it's simply performing a relinking that's like a rename on the same filesystem, and a copy-and-delete (move) across different filesystems). Granted this might be a bit confusing to a novice/very casual computer user, but it shouldn't be too hard if one reads the fine manual.
Even if a user finds this confusing after RTFM, just have them drag with the right mousebutton. The context menu erases all doubt.
(For the record, I've not really used Windows before Windows 95, I found the 3.x series to be hideous and I was completely non-PC before 1990. I suspect Win3.x had a lot of GUI infelicities. None of my comments apply to pre-95 series Windows)
Oh, off on a tangent, one thing that has always annoyed me about post-Amiga OSes is that they do NOT seem to support what I call 'Type Behind'. Since most of these OSes either foreground a window when you click on it, or worse yet, when you hover your mouse over it, you must shuffle the windows on the screen if you want to type in one window while reading from another. Or click back and forth a lot. Or copy the information into the other window (often not possible with images/diagrams). AmigaOS doesn't automatically bring the selected window forward (unless you configure it to; see the ClickToFront commodity, I use that with the ALT key modifier to bring a window forwards), which lets you use touch typing on an input box / text editing area while viewing a document/etc over it.
PS. Running Debian Testing on this machine (2500XP/ATI 9600) is a sluggish and unresponsive GUI experience, compared to XP. Further, the driver for the nForce RealTek 8100 network adapter is rather crappy. Mind you, I find XP to be sort of sluggish and unresponsive too, just to a lesser extent. I like Linux-console-on-AthlonXP/P4 or AmigaOS on 060-with-video-accellerator type of responsiveness, myself.
Isn't throwing lots of money at someone a really, really persuasive way of making a convincing case? I know that would work on me! ;p Mind you, Paramount should have already responded with three mil being offered. How many other shows have that much donation? How many have ANY donation at all?
Um, isn't HyperTransport some sort of I/O bus, like PCI/AGP/PCI Express? You know, device-to-CPU/chip-to-chip type stuff? I don't think it's actually used for memory access, is it?
In any case, it's still going to be limited by the fact that even top grade DDR is still many, many times slower than what the CPU can handle. (In applications that are tied to memory performance, anyways)
Actually, the impact of memory latency and bandwidth on actual performance varies greatly with the type of application you're running.
Some of my software has been very efficient in terms of CPU cycles, but dealt with a vast amount of data and thus was tied to memory latency/bandwidth. Those projects would run at almost the same speed on a Pentium 166 MMX with SDRAM as on a P2-350. Some other projects were far more cycle-oriented but only used registers, and didn't care at all about memory performance. These projects showed a dramatic improvement on the P2.
Imagine, as an example, a function that increments or inverts each byte in a large array, versus one that iterates a variable or two. The former will rely on memory bandwidth, whereas the latter will end up with most or all of it's variables in register (let alone cache) and thus rely on the CPU's overall internal performance.
Actaully, the zero/NULL thing is just an abstraction in C.
..." or such, the compiler notices that you're comparing a pointer to the value zero, and will then generate machine code that checks it against the machine/OS specific invalid address value, rather than an actual comparison to zero.
When you say something like "char *ptr; if(ptr==NULL)
Note - This physical null value happens to BE zero on most machines/OSes, but on some others it's weird stuff like $3F:FFFFFFFF.
Probably. But I'm using Debian, which is about to move to 2.4 once it's considered stable. Heh.
.22 or so.
Mind you, that's a little extreme. I think they could have moved to 2.4 once it reached around
I still wouldn't move the main distro kernel to 2.6 yet though. That's great for an option, but I'd rather just stick with 2.4 for any server type stuff. (I do use 2.6 on my Linux-based workstations, but they're all behind a firewall with no forwarding and all my data files are stored on a 2.4-based file server)
But people that say that GTA doesn't cause violence are missing the bigger picture that it allows people to discover just how little coverage the police have in real life. That's probably not a good thing for anybody
Ahh, security through obscurity. That's great! Everybody doesn't know that the security doesn't really exist, so it works as if it does exist! Wonderful! Superb! Excellent!
Oh well.
- driving at triple digits on public roads is 'fun'
Huh? I do this all the time. You've written it like as if it's wrong or something.
In fact, the speed limit here IS three digits on some public roads. Not that ANYBODY actaully drives under or at the limit, anyways. The police are usually doing the limit+20km/h themselves when there's no emergency.
Oh, you must be using that legacy measuring system. Leagues per fortnight, hogsheads per rods, etc. Tisk tisk.
(For those who did not RTFA, it compares Redhat Enterprise Server to Windows Server 2003)
Redhat has always seemed to be a flashy, large distribution which favored new features and gadgets over stability and security.
I wonder how say, Debian (my personal favorite) might do in terms of security, or better yet, one of the security-centric distributions.
"Properly distributed" is a sort of whacky thing to say.
:p
That could be applied to anything. Here's an example: Four inches of high quality sharpened steel, properly distributed (amongst the ribcages of the world), would kill EVERY LAST PERSON. Now imagine what the blade from a wind turbine could do.
Utter lunacy. Cassini probe protesters should probably try smoking -less- than eight blunts at a time.
I thought one of the big ideas of Perl was that it was more or less typeless.
It will automagically convert between types as needed.
(Disclaimer: I don't have much object-oriented Perl experience, and nor do I want to have it ;p. The OO parts of perl might change this typelessness somewhat... )
Don't worry, we've extracted samples of your DNA, and tagged you with several redundant (classified) tracking devices that will track you anywhere in, under, or above the Earth. We've also implanted your brain with devices that can monitor your thoughts, as well as disable you should you become violent or make fun of the President.
We've installed keystroke loggers and patched your I/O stream functions, and replaced your copy of lynx, wget and links with an especially insecure version of IE running in text emulation mode.
We have all the private keys for all your encryption protocols, and have added special extra backdoor features to all protocols as added insurance. Most of these backdoors have the password or pass-phrase of 'password' and/or all-zero hex keys.
Our next plan will be to abduct you in the middle of the night, and place you in a simulated reality where you can do no harm. Once you commit your terrorist acts there, we can proceed to arrest, detain, question, and dissect you.
Yours Truly,
The Central Intelligence Agency.
PS. Don't give us any 'jurisdiction' crap. This is OUR planet. You are only here until we arrest you. And stop making 'the finger' at that camera. That's rude to our operators.
Gawd, what awful applications. I always hate when that happens.
My own code for Windows platforms is always pure Win32 (with DirectX or OpenGL for graphical things), and will run on Win95 through to XP SP2.
The trouble the other coders get into all starts when they start using auxillary components in Windows, like IE or the goddamned MFC library or such. (This is also where many of the security holes come from)
I'm pretty sure Debian runs a vanilla kernel as well.
I usually roll my own by hand on my Debian boxes, but there's a package that lets you build a Debian-kernel-binary-package from a vanilla kernel source tree. It never said anything about requiring special patches or whatnot.
Also, my self compiled, purely stock kernels don't seem to have any problems in Debian.
Or maybe they figure you're too crazy to sell anything to. ;p
> If you didn't buy the Compaq, how do you know it still runs?
.. Just 600. Sort of funny that the laptop's PRIMARY SYSTEM DOES NOT WORK. (There was also a slashdot article at one point in the past about this very problem)
A friend got one. We had a supplier in the office who was selling them. He bought that, and I bought the Thudpad. The Compaq works fine, doesn't have any driver or IRQ issues, and is still running on the original stock battery. It's quite old now, too, been running for the past 5 years.
http://www3.sympatico.ca/milnoc/TP600Battery/
I find it funny that this comes up in the first page in a search for "Thinkpad 600". Not 'Thinkpad +"battery problem"'
I mean, a laptop without a battery is just a low performance desktop in a small, badly designed package.
What the heck is it with this DLL happiness?
Fonts are DLLs, keyboard layouts are DLLs.. Gimme a break! Make it some sort of well-documented DATA file format!
Gah!
That's interesting, but it kills some of the advantage of the usual keys...
C / Copy. This one is obvious.
X / Cut. It's right next to copy, and X is a symbol for deletion.
V / Paste. V is the same shape as that mark used to insert text.
Those three keys are standard for a lot of OSes - Macs, as mentioned before, Windows, and Amigas, to name a few (Amigas use left-Amiga key, which is rougly where the 'windows' key is on a PC keyboard. Actually, the 104 key PC keyboard is much closer to the old Amiga keyboard than the 101 key is, but it's still missing the HELP button)
Try riding a motorbike - the clutch is hand-operated and the gearshifter operated by the foot. It works very well.
...
Actually, ditto for standard transmission automobiles - the clutch is operated by the foot, and the shifter by the hand.
These all have to be in proper order, where milliseconds do matter. I never have a problem. Naturally, different people have different capabilities - while most of my friends can shift properly, there's one who always manages to stall
> Btw: How many "laptop problems" did you have with ThinkPads?!
Oh, I dunno.. it destroyed $300 batteries regularly, it had too many IRQs for all devices to be enabled at once, it had this ugly little cli--er, eraser thingy in the middle of the keyboard which was useless and required adjustment every few days, the BIOS setup looked like some 1983 CGA game for retarded five year olds, and was about as useful, it was twice the size and half the power as the equivalent Toshiba model of the time... Oh yah, and the drivers which when downloaded, had to be extracted onto a _floppy_ disk.
IBM Thinkpad 600E U2426. All the crap you never wanted, and then some. (Oh yah, the model number is obtuse and there are a trillion useless variants to this useless heap of garbage)
I should have bought that Compaq instead. It was the same price, and still runs to this day.
Oh well. At least IBM still offers downloads on legacy products like that, even if they are really stupid to install and highly broken.
The 'highly broken' part is a universal standard for the PC industry anyways though, which IBM created with that crappy machine they threw together in a drunken binge on a weekend in 1981 -- Unsharable interrupts? Static OS structures? Segmented, 20-bit memory? Processor-driven I/O? How could you do this with a motto of "Think"?? Or maybe that's a typo, maybe it's.. DRINK??
Remember kids, alcohol, like smoking, is _not cool_.
Well, you can apply the physics-2.6.11-st (Star Trek/Space Warp) patch, or the physics-2.6.11-sw (Star Wars/Hyperspace Jump) patch, or avoid stars entirely and apply the physics-2.6.11-b5 (Babylon 5/Hyperspace Gateway) patch, or a myriad of other alternative patches.
/plotholes/speed/warp-factor-adjust" to increase speeds!
If you want to run a vanilla physics kernel, however, you are out of luck for now.
Some highlights of the major Star* patches:
Star Trek:
- Later versions of the patch allow you to make up particles to overcome problems and rewrite physics entirely!
- Most versions of the patch allow you to adjust speed to a desirable setting. Need to go to the center of the galaxy in one movie, when it would otherwise take about thirty years in a series? Do "echo 9999.9 >
- Early versions allow you to save the galaxy (well, the Federation) without huge armadas of ships - just get into a fistfight! (Warning: your Captain's shirt might get torn)
- And much more!
Star Wars:
- Allows you to convert distance into time and vice versa! (supplemental patch allows you to make up all sorts of weird explanations for it too; see first point for Star Trek patch)
- Makes light a lot faster for ships, and slower for lasers! ("Jump to lightspeed!")
- And much more!
Anyways, gotta run, tah tah~
If the format is open, then an open sourced solution could be developed to deal with any issues with the proprietary software.
If the file formats used are truly open (as in a decent standard that's well documented and actually works as it's described, which allows other applications to read and write in the same format without legal encumbrance), then the customer can take their data and have a new application or data converter written, allowing them to easily migrate to a new platform.
That addresses the following issues thusly:
>Other valid complaints are the per-seat costs, upgrade costs
These are limited by the fact that the customer can now walk away. Something they cannot easily do if they were locked into Word's *.doc format or such. If the proprietary vendor tried strong-arm tactics or charged too much, the customer could simply write, or have written, a replacement application (or a file converter to convert the data files to a format used by another application).
This effectively caps the maximum cost that the vendor can impose. Company X cannot charge a billion dollars for a given product if Company Y can offer the same thing for $59.95 that works on the same data, or if Open Source Programming Group Z offers a totally free version of said program.
Also, the bigger the vendor, and the more pervasive their product, the more likely some open source group is to pick up the challenge of making an open sourced solution which works with these same data files.
> limited effectiveness of outside support,
Well, only with regards to the application itself. The data, properly documented and open, could easily be handled by any competent programmer.
> architecture lock-in
What architecture lock in? Vendor screwing you over? Take your data and walk away. Have someone else write a replacement. Write it yourself! Use Y or Z's product! Suddenly Corp X decides that the 3.2 update will NOT cost a trillion dollars, afterall.
>and a slow, costly route to get bug-fixes and new features implemented.
Bug fixes to the original application would depend on the vendor. New features, however, could be implemented by utilties that work with the data created by the application. And if the original application was too buggy to work with at all, why was it purchased to begin with?
Open sourced solutions can be costly to bugfix as well. If your package is abandoned, you'd have to _hire_ people to fix it.
> And, of course, the threat of being left out-in-the-cold if the company stops offering the proprietary program (or if said company collapses).
How do you figure this? Take the data and _WALK AWAY_. This would be no different if an open sourced group collapsed or grew bored of a package.
---
Basically, having open file formats is like a foot in the door for open source software: If free/open source software can work with the same data, and be better and cheaper, then why wouldn't some conversion happen? It also takes away one of the tools in the arsenal of big corporate evils: the hostage-holding of your user-created data.
This could also revitalize a rather stagnant commercial software industry by forcing more interoperability.
What, can't open sourced software compete then? Is it like some big, free version of Microsoft? Forced to rely on legislation to get it's place in the world? I'd hope not!
I'd say that it's probably better to allow for free and proprietary to compete openly, than to have one side legislated over the other. It could easily go the other way, you know. "All government agencies must use Microsoft(TM)(R)(C)(Patent Pending)(TM)(C)(R) software AT ANY COST!".
How could this not be win-win?
This of course, all hinges on the definition of an 'open' format:
1. It must be readable and writable by other applications or utilities withou