Oh, man, congratulations. A fine troll, indeed. My hat is off to you.
Let me sum up:
"The Sims is not l33t because it does not require mad skillz. It would be better if it had human-like AI that could be generally used for everything. Come on! Doom 3 looks so good, why can't we pass the Turing test yet?"
The mad skillz, I'm not even going to comment on. The open avatar bullshit? Seriously, Sims forget to go to the fucking bathroom. They are not bright. The only way they function at all is by you constantly telling them what to do. Which may be what your complaint is actually about, except that what you propose is something that mankind has not figured out how to do, and likely won't for a long damn time.
I shake my head at moderators who see "Open Standards" and "mod community" and somehow are able to equate it with "fantastic modular AI bots which will bend to our every whim using a nice GTK slider interface." Grassroots "we can do it if we just try hard enough" AI work (also known as GOFAI, or "Good Old-Fashioned AI") hasn't gotten us anywhere in the past 50 years; I would wager that it's just not going to.
Kudos, though, for somehow making your psychotic rambling look well-reasoned and insightful.
By
functional-programming, I assume you mean structured programming?
No, I think he means functional programming. You know, that Lisp / Scheme ( / ML / Haskell / APL / Joy /... ) stuff that can't be used in a structured manner. Non-imperative programming.
The XGameStation, as far as I can tell, is the first computer in decades which is designed to not be magic.
So many developers these days, even after graduating with a computer science degree, don't have the faintest idea of what goes on inside the boxes on their desks which they are supposedly so obsessed with. It's black magic. Doesn't anyone else find this a touch distressing? How can you possibly call yourself a computer scientist without understanding how computers actually work?
People who are saying, "It's so much easier and cheaper to just buy a Dreamcast or a GBA or a modded XBox or an Atari 2600 and it'll perform better too!" are missing the point. The point is that you are taken, step by step, through how this hardware platform works. It's low-tech because it's interesting to learn how to interface with a television. It's *fun* to be a programmer with a soldering iron. You understand every layer of abstraction, you know precisely what is happening as your cube rotates on your TV, and it is *cool*.
It's important to see through layers of abstraction to what's actually going on, sometimes.
Repeat after me: Reverse engineering for the purposes of interoperability.
Compare the following conversations:
"Hey MPAA, let me play DVDs in Linux."
"No."
* hack *
"I told you not to do that! Now people can enjoy the media of their choice on the platform of their choice!"
"Hey Apple, let me play RealMedia on iPods."
"No."
* hack *
"I told you not to do that! Now people can enjoy the media of their choice on the platform of their choice!"
"It plays my DVDs out of the box" is not what will make Linux on the desktop work. What makes the desktop work is the antithesis of open-source and UNIX philosophy. The desktop is not about describing your task with small tools that do one thing well, it is about performing your tasks with large tools that are designed around performing related sets of tasks. Linux hackers are bored with this problem. They don't want to bother.
What Linux needs to succeed on the desktop is a thriving community of user interface hackers led by a Steve Jobs visionary-type. Linux has nothing to attract such people. Linux, in fact, has plenty to turn these people away, from a community that thinks the Gnome and KDE wars are good because it promotes choice, and that X is a good UI solution because you can download window manager themes with penguins and hot anime babe backgrounds. These people run screaming to their Macs. Their Macs understand them.
What is missing from the Linux desktop is not features. Linux does a tremendous job of having lots of features. What it does not have is any concept of the situations in which its users might use these features. It doesn't care; if you can do something, how can it be broken? You're just too lacking in hacker spirit to figure out how it works.
Uncle Grandma is never going to have enough hacker spirit to figure out how it works. If Free Software is to solve every problem in the world, it will recognize that. But -- here's a radical idea for you -- maybe Free Software and the Hacker Ethic aren't good at everything! Maybe it shouldn't solve every problem in the world! Perhaps some problems just don't fit will with the Open Source philosophy! Perhaps Linux will never catch on as a mainstream option for the desktop! Perhaps this isn't even a horrible, blasphemous thing!
The problem with that is that such things are not instantly consumable. With wikis and software, you're using it as you're contributing. Feedback is basically instant; you don't like the article, you change it. You find a bug in some software, you fix it.
With open-source hardware, you what? Spend hours and cash building something from a spec? Everyone contributing from all over the world is going to do this? No, what will happen is that someone will invent something, get 100 of them built, and sell them off. In the physical realm, that's what makes the most sense. It's way too expensive and inefficient to build one of something like a car.
When matter can be copied as cheaply as ideas, this will work.
"Interesting new technology! Wow, imagine what we could do with this!"
"No! Clearly the fact that you cannot immediately enumerate all of the possibilities of this technology means that it is garbage!"
"But what if--"
"NO!"
Re:You don't have the slightest idea...
on
QNX 6.3 Released
·
· Score: 1
Timesys's website is infuriating. I click on "Technical details" and get "Here's a list of drivers that it comes with." Well, are they semi-standard Linux kernel drivers somehow recompiled to run in user space? Are they custom drivers developed by TimeSys? When you claim you've turned a non-realtime OS into a hard-realtime one, there should be some mention of *how*, somewhere.
I *think* the drivers are custom-written. At which point, why are you bothering to call the thing Linux? What's the advantage of using a Linux kernel with a custom scheduler and no drivers?
The chief advantage to going with Linux for embedded, as I understand it, is the ridiculous heaps of driver support. Is that still there, or was that lost when the Linux kernel was mashed into a microkernel? If so, then, what, hard-realtime tasks can only talk to the user-space drivers, and the kernel is off-limits but needed significantly less? Is that the theory?
I'm downloading one of their free BSPs right now, so hopefully I'll be able to figure out for myself just how it works and stop ignorantly ranting at people on Slashdot.
Re:You don't have the slightest idea...
on
QNX 6.3 Released
·
· Score: 1
Okay, but have you somehow made the kernel hard realtime, or just "low latency"? I can pre-empt it, but I don't suppose I can re-enter it? Say the kernel is in a busy loop waiting on the floppy disk drive, processing a request from some lowly soft-realtime task. Do I have some guarantee that it's going to come out in time? Or should I just never touch the kernel in a hard-realtime task?
This is why a microkernel and user-space drivers make so much more sense for real-time. Your priorities are much finer-grained.
Re:QNX is the bad touch
on
QNX 6.3 Released
·
· Score: 2, Informative
Man, you're a troll. QNX2 came on 5.25" floppies in 1987. Here are some examples of some other things that happened in 1987:
IBM introduced the fantastic new "VGA" graphics standard, blowing everyone away with its 256 colours of fury.
Surely you'll concede that QNX 2 is superior to Windows 2?
Re:You don't have the slightest idea...
on
QNX 6.3 Released
·
· Score: 1
The webpage doesn't have much information about how this is actually achieved -- can you actually make kernel calls from a hard-realtime process, or is it just another hack to allow tasks run at a higher priority than the really-not-particularly-realtime kernel?
ObDisclaimer: I just finished spending the last 5 months of my undergraduate career fighting tooth and nail with RTAI for my thesis. Before that, I'd spent four years as a QNX consultant. Man, what a difference.
From the webpage itself, now that it's back up:
"Someone thought it'd be cute to link to me, which isn't a good idea with the 10 user version of IIS. Sorry for the 403.9's."
Want to DDOS someone? Just pretend to be them and submit their webpage to Slashdot!
Not quite -- imagine, if you will, if I rolled out of the lot with a hybrid manual/automatic car. The smarmy used car salesmen is smiling and waving as he says to me, "You don't want to worry about shifting gears? Just don't touch the clutch. No problem!"
The user, slightly wary, but trusting, cruises along in his slightly strange car. A song comes up on the radio that the user has a particularly passionate hatred for, so he fiddles with the dial to find a new station.
Suddely, the engine sputters out and stalls. The hapless user manages to make it to the side of the road before the car dies completely. Irate, he calls the dealer on his celphone. "Oh, were you listening to 98.3FM?", he says. "Yeah, that station doesn't support the automatic transmission; you have to switch to manual if you want to listen to it. Don't worry! You'll get used to it."
The problem? Because the scheme is so complicated, and because it got changed several times over the lifetime of X, developers haven't always properly supported it. And so, with some programs, Ctrl-C/Ctrl-V simply doesn't work. And then what does old Aunt Grandma User get told by her well-meaning Unix-expert grandnephewson? "Oh, that program works a little differently -- here, let me show you how to do it." She gets confused, forgets which program to use which technique in, tries to paste in a URL with the middle mouse button and all hell breaks loose. "This UNIX stuff is like reading Greek backwards while underwater!" she exclaims. And a little piece of Linux on the desktop dies.
Turn the middle-click off by default. Let the power-users enable it explicitly if they want it.
You know, Ctrl-C/Ctrl-V is the manual transmission of your story, really. The middle-click method makes an assumption about when you want to copy (switch gears) which are often less than optimal and that there's no real way to control. I mean, sure it's efficient, when it's right.
Now, pretending for a moment that the middle-click system wasn't so deficient that an entirely different mode of copy/paste is needed alongside it to make up for its shortcomings, are you saying the clueless user shouldn't be allowed to drive an automatic? That it's somehow a failing that he wants technology to solve his problems, rather than aggrivate them? That he should be strapped down and educated for his own good?
I mean, there needs to be a balance, here. I agree, the user shouldn't say, "This is bullshit, why can't the machine read my mind?", but he also shouldn't say, "This is a poorly thought out piece of garbage, but I'll put up with it anyway, and maybe even learn to like it one day." You need to look at it critically, see both its good and bad points. Then you can say whether it fits your needs or doesn't.
The needs for cut and paste: Efficient, easy to learn, hard to fuck up. There just aren't really very many good points about middle-clicking with those goals in mind. The efficiency gains from not having to explicity copy, in my mind, are negated by the much higher complexity of the mental model needed to support two modes of cut/paste at once. Especially when many applications are broken, only supporting one or the other, or maybe even combining them. And it's important to realize that the applications are broken precisely because the model was too complicated. The programmers couldn't be bothered to figure it out properly when it came time to implement cut/paste functionality. Hardly the mark of the Right Way To Do Things!
However, you've learned, probably after a fair bit of training, to use it properly. Great! I'm happy that you are able to navigate the minefield by heart and have become more productive for it. More power to the power users, I say! But it's still a fucking minefield. It's not something you want to force on everybody. And nobody is going to clean it up unless someone points out that it's there.
I think you took my comment exactly backwards from how I meant it. I'm saying that the fact that you've been working with Linux for 4+ years and still have trouble with copy and paste means that something is pretty fucking seriously wrong with copy and paste. The clipboard is a vital piece of functionality, used by almost all users many times a day, and it's totally, hopelessly broken. A complete and utter failure of user interface design, in almost every imaginable respect.
The fact that someone would blame themselves for not knowing how to work around such a blatant design flaw is troubling, to me.
Does anyone else find the phrase "I'm sure there is a reason, and I'm also sure it's my fault for not knowing it" with respect to basic, everyday user interface tasks troubling?
You'd be surprised how many things in business are done, not because they're the right way to do things, but because they can be easily measured. It's really rather frightening.
If you can tell me what "The Windows 9x kernel works by sending commands from the GUI, to
DOS, then to the kernel and back to DOS, and back to the GUI" is supposed to mean, you are a better man than I. The article seems to be suggesting that Windows NT was released around the same time as Windows 98..? And that it was called "MS-NTet"? (Which happens to be a googlewhack!)
If you're talking about the link between NT and VMS, this or this would be a much better read. If not, then... what the hell are you talking about?
Slashdot Mirror
And you got this information where?
Let me sum up:
"The Sims is not l33t because it does not require mad skillz. It would be better if it had human-like AI that could be generally used for everything. Come on! Doom 3 looks so good, why can't we pass the Turing test yet?"
The mad skillz, I'm not even going to comment on. The open avatar bullshit? Seriously, Sims forget to go to the fucking bathroom. They are not bright. The only way they function at all is by you constantly telling them what to do. Which may be what your complaint is actually about, except that what you propose is something that mankind has not figured out how to do, and likely won't for a long damn time.
I shake my head at moderators who see "Open Standards" and "mod community" and somehow are able to equate it with "fantastic modular AI bots which will bend to our every whim using a nice GTK slider interface." Grassroots "we can do it if we just try hard enough" AI work (also known as GOFAI, or "Good Old-Fashioned AI") hasn't gotten us anywhere in the past 50 years; I would wager that it's just not going to.
Kudos, though, for somehow making your psychotic rambling look well-reasoned and insightful.
You might be surprised. See "High Level Function Interception".
No, I think he means functional programming. You know, that Lisp / Scheme ( / ML / Haskell / APL / Joy / ... ) stuff that can't be used in a structured manner. Non-imperative programming.
It's kind of an interesting field.
So many developers these days, even after graduating with a computer science degree, don't have the faintest idea of what goes on inside the boxes on their desks which they are supposedly so obsessed with. It's black magic. Doesn't anyone else find this a touch distressing? How can you possibly call yourself a computer scientist without understanding how computers actually work?
People who are saying, "It's so much easier and cheaper to just buy a Dreamcast or a GBA or a modded XBox or an Atari 2600 and it'll perform better too!" are missing the point. The point is that you are taken, step by step, through how this hardware platform works. It's low-tech because it's interesting to learn how to interface with a television. It's *fun* to be a programmer with a soldering iron. You understand every layer of abstraction, you know precisely what is happening as your cube rotates on your TV, and it is *cool*.
It's important to see through layers of abstraction to what's actually going on, sometimes.
I'm sure they have every intention of starting something new and exciting, and will end up creating another, prettier, first-person shooter.
Maybe you'll be a pirate. That'd be pretty innovative.
Compare the following conversations:
"Hey MPAA, let me play DVDs in Linux."
"No."
* hack *
"I told you not to do that! Now people can enjoy the media of their choice on the platform of their choice!"
"Hey Apple, let me play RealMedia on iPods."
"No."
* hack *
"I told you not to do that! Now people can enjoy the media of their choice on the platform of their choice!"
"It plays my DVDs out of the box" is not what will make Linux on the desktop work. What makes the desktop work is the antithesis of open-source and UNIX philosophy. The desktop is not about describing your task with small tools that do one thing well, it is about performing your tasks with large tools that are designed around performing related sets of tasks. Linux hackers are bored with this problem. They don't want to bother.
What Linux needs to succeed on the desktop is a thriving community of user interface hackers led by a Steve Jobs visionary-type. Linux has nothing to attract such people. Linux, in fact, has plenty to turn these people away, from a community that thinks the Gnome and KDE wars are good because it promotes choice, and that X is a good UI solution because you can download window manager themes with penguins and hot anime babe backgrounds. These people run screaming to their Macs. Their Macs understand them.
What is missing from the Linux desktop is not features. Linux does a tremendous job of having lots of features. What it does not have is any concept of the situations in which its users might use these features. It doesn't care; if you can do something, how can it be broken? You're just too lacking in hacker spirit to figure out how it works.
Uncle Grandma is never going to have enough hacker spirit to figure out how it works. If Free Software is to solve every problem in the world, it will recognize that. But -- here's a radical idea for you -- maybe Free Software and the Hacker Ethic aren't good at everything! Maybe it shouldn't solve every problem in the world! Perhaps some problems just don't fit will with the Open Source philosophy! Perhaps Linux will never catch on as a mainstream option for the desktop! Perhaps this isn't even a horrible, blasphemous thing!
With open-source hardware, you what? Spend hours and cash building something from a spec? Everyone contributing from all over the world is going to do this? No, what will happen is that someone will invent something, get 100 of them built, and sell them off. In the physical realm, that's what makes the most sense. It's way too expensive and inefficient to build one of something like a car.
When matter can be copied as cheaply as ideas, this will work.
"Interesting new technology! Wow, imagine what we could do with this!"
"No! Clearly the fact that you cannot immediately enumerate all of the possibilities of this technology means that it is garbage!"
"But what if--"
"NO!"
I *think* the drivers are custom-written. At which point, why are you bothering to call the thing Linux? What's the advantage of using a Linux kernel with a custom scheduler and no drivers?
The chief advantage to going with Linux for embedded, as I understand it, is the ridiculous heaps of driver support. Is that still there, or was that lost when the Linux kernel was mashed into a microkernel? If so, then, what, hard-realtime tasks can only talk to the user-space drivers, and the kernel is off-limits but needed significantly less? Is that the theory?
I'm downloading one of their free BSPs right now, so hopefully I'll be able to figure out for myself just how it works and stop ignorantly ranting at people on Slashdot.
This is why a microkernel and user-space drivers make so much more sense for real-time. Your priorities are much finer-grained.
Surely you'll concede that QNX 2 is superior to Windows 2?
ObDisclaimer: I just finished spending the last 5 months of my undergraduate career fighting tooth and nail with RTAI for my thesis. Before that, I'd spent four years as a QNX consultant. Man, what a difference.
From the webpage itself, now that it's back up: "Someone thought it'd be cute to link to me, which isn't a good idea with the 10 user version of IIS. Sorry for the 403.9's." Want to DDOS someone? Just pretend to be them and submit their webpage to Slashdot!
Yes, but at least in Apple's terminal, it's Command-C that does the copying -- just like every other Apple app!
The user, slightly wary, but trusting, cruises along in his slightly strange car. A song comes up on the radio that the user has a particularly passionate hatred for, so he fiddles with the dial to find a new station.
Suddely, the engine sputters out and stalls. The hapless user manages to make it to the side of the road before the car dies completely. Irate, he calls the dealer on his celphone. "Oh, were you listening to 98.3FM?", he says. "Yeah, that station doesn't support the automatic transmission; you have to switch to manual if you want to listen to it. Don't worry! You'll get used to it."
The problem? Because the scheme is so complicated, and because it got changed several times over the lifetime of X, developers haven't always properly supported it. And so, with some programs, Ctrl-C/Ctrl-V simply doesn't work. And then what does old Aunt Grandma User get told by her well-meaning Unix-expert grandnephewson? "Oh, that program works a little differently -- here, let me show you how to do it." She gets confused, forgets which program to use which technique in, tries to paste in a URL with the middle mouse button and all hell breaks loose. "This UNIX stuff is like reading Greek backwards while underwater!" she exclaims. And a little piece of Linux on the desktop dies.
Turn the middle-click off by default. Let the power-users enable it explicitly if they want it.
Now, pretending for a moment that the middle-click system wasn't so deficient that an entirely different mode of copy/paste is needed alongside it to make up for its shortcomings, are you saying the clueless user shouldn't be allowed to drive an automatic? That it's somehow a failing that he wants technology to solve his problems, rather than aggrivate them? That he should be strapped down and educated for his own good?
I mean, there needs to be a balance, here. I agree, the user shouldn't say, "This is bullshit, why can't the machine read my mind?", but he also shouldn't say, "This is a poorly thought out piece of garbage, but I'll put up with it anyway, and maybe even learn to like it one day." You need to look at it critically, see both its good and bad points. Then you can say whether it fits your needs or doesn't.
The needs for cut and paste: Efficient, easy to learn, hard to fuck up. There just aren't really very many good points about middle-clicking with those goals in mind. The efficiency gains from not having to explicity copy, in my mind, are negated by the much higher complexity of the mental model needed to support two modes of cut/paste at once. Especially when many applications are broken, only supporting one or the other, or maybe even combining them. And it's important to realize that the applications are broken precisely because the model was too complicated. The programmers couldn't be bothered to figure it out properly when it came time to implement cut/paste functionality. Hardly the mark of the Right Way To Do Things!
However, you've learned, probably after a fair bit of training, to use it properly. Great! I'm happy that you are able to navigate the minefield by heart and have become more productive for it. More power to the power users, I say! But it's still a fucking minefield. It's not something you want to force on everybody. And nobody is going to clean it up unless someone points out that it's there.
The fact that someone would blame themselves for not knowing how to work around such a blatant design flaw is troubling, to me.
Because you really should.
Noise-cancelling headphones while listening to Primus? Doesn't that sort of defeat the purpose?
You'd be surprised how many things in business are done, not because they're the right way to do things, but because they can be easily measured. It's really rather frightening.
If you're talking about the link between NT and VMS, this or this would be a much better read. If not, then... what the hell are you talking about?
DOSBox and a shitload of power under the hood might do the trick.
Too bad it seems that nobody knows how to teach it.