Chassis - The car that wins is going to have very carefully measured and constructed alignment. Getting the vehicle to run 20 meters straight, and with the thrust exactly aligned (in whichever direction the particular design calls for) is key. You can't afford to waste any of those thrust newtons on trying to drive the car sideways, or having the car trade forward velocity for lateral movement.
Water/Air ratio - if you run out of air too quickly, you are hauling excess water all the way down the track. Similarly, if you design the vehicle such that air can escape freely once the water level falls below a certain point, you are wasting power.
"Gearing" - one has to wonder if a direct-drive design (air pushing water out the back) is the most efficient. If you used hydraulic principles, you could in effect "gear down" the high pressure, resulting in high-torque that could drive over-size wheels, similar to the way that a hydraulic lift works. The winning design is going to have to find a tradeoff between quick accelleration/coasting and continued power for the duration of the track. Could the careful design of a nozzle accomplish the same thing? (/me thinks back to the model rocket days)
Multiple pressure vessels - this is just fanciful thinking, but one of the techniques steam locomotives used was to use the high pressure air in one set of cylinders, and then re-use the resulting lower pressure to drive a second set of cylinders. Extrapolating from this, I wonder if having two pressure vessels, totalling the legal limit, and firing at different times (ie. one for acceleration, one for maintaining speed) might be feasible.
I guess, though, in the end, the simplest, lightest design will win. Having a good chassis is still key though.
This is only the second kernel I will have built (just installed Slackware 8.0 for the first time, and built 2.4.11)...Can I reuse the configuration file created by "make menuconfig" with 2.4.12, or should I try and re-select all the options I had previously?
"it's just that there are more and more applications that REQUIRE a GUI, or that are badly designed into using one "
- not really. More and more people are shoehorning applications into the WIMP model, whether they fit or not. That isn't the same as applications that "REQUIRE" a GUI. Badly designed applications are badly designed..If people choose to use them, what can you do? A better line of inquiry would be "why are most applications going to the WIMP model, appropriately or not?"...I'd hazard a guess that it's because more efficient or not, it's easier for the average user to understand and operate.
"A good example is the browser syndrome, where keyboard shortcuts are a no-go in most cases -- GUI systems like Motif, GTK, or even Win32, are well thought out. But remember, the original goal of GUIs was to facilitate _data representation_, i.e.: make it more easily assimilable by the person, NOT easier input"
"In the case of applications like CAD, graphics design, etc... a pointer-style input device makes sense -- but we NEED to solve the Repetitive Strain problem "
- Its not the mouse or the GUI that is causing RSI. If it was, why so many ergonomic keyboards? Why aren't there any ergonomic guitars? RSI is caused by bad ergonomics, nothing more, nothing less. If an application was poorly enough designed that it cause RSI, very few people would be able to use it for any length of time (which in fact makes me wonder about mouselook in FPS games, and what that is doing to a generation of computer users)
Unix is chock full of hard to reach keyboard symbols that have significant meanings (and that are needed to run certain commands) For all that geeks are supposed to cherish keystroke economy, there are a suprising number of conventions that require removing your hands from the home row, or using alt/ctrl/escape, etc to modify the default meaning of keystrokes.
Use the right tool for the right job. That's the UNIX/Linux philosophy. If you think that the WIMP interface is cramping your style with respect to whatever task you are trying to accomplish, then you are using the wrong tool.
With regards to the command line or WIMP interfaces being old, and not particularly forward looking, you are also missing a fundamental point: A graphical "pipe" isn't innovative either. You're simply shoehorning two paradigms together, and even worse, two totally incompatible paradigms at that. The pipe is a useful metaphor and operator for stream-oriented I/O. The WIMP is useful for (obviously) visually oriented information, and its designed for a completely different purpose than the pipe. The WIMP is designed to allow humans to manipulate data and abstract objects in a visual manner. The pipe is designed to allow users to allow the computer to do the same, without intervention.
If you want an innovative computing interface, worrying about streams, or visual representations of data is a waste of time. You're going to have to come up with something totally new. One good example is the use of sound to communicate the health and performance of networks or systems.
There's some significant insight contained in this post. I work for a company that, stats wise, generates 4x the volume of email of any other company in Canada, on a per-employee basis (we've been doing email on IBM mainframes since before most companies had computers).
Email is much more than just another form of messaging. I've seen email used as a form of decision records, as a primitive form of version control (the email thread contains each revision of the document in question), as discussion threads, and even as a form of middleware for some very significant applications (like train dispatching). In a company like mine, end-to-end delivery of email messages that exceeds 30 seconds is seen as a serious degradation of service (no kidding!)
Consitent, timely, and reliable delivery of email in large companies (outside, perhaps, of large dot.com online sales companies) is arguably more important than nearly any other form of networking.
The Afghanis maybe didn't shoot one of these down
on
Robots Go To War
·
· Score: 1
If you check the CNN article, it seems that Reuters is confirming that it was a helicopter belonging to the local opposition...
Why is this rated +4, funny? If the poster is describing an experience where he gamed and drank all night long, then left the LAN party the next morning, then he is likely describing driving drunk...If he is describing driving the day after (ie. assuming he got some sleep), then his description still fits that of impaired driving.
I don't think impaired driving is funny, particulary because of the terrible loss of life that can result.
There's no way Exchange2K could handle 50K users on a single box.
First, you've never obviously worked with Fibre Channel on the kind of scale that 50K users would require (ie. a big EMC box)...that many users pounding the same box will easily chew up 50% or more of your CPU power on I/O alone. Fibre is fast, but it is so fast that it can easily swamp Xeon CPU's. I know, because I did the benchmarking at my company.
Second, connection limitations in Win2K and Exchange alone mean that you are running very close to the theoretical maximum the OS supports..not a good idea.
Third, running that many users off of a single box is suicide. And if you've ever watched Exchange2K failover on a Win2K cluster, you'd know that it can take several minutes for everything to come up on the second node, if you've got a lot of users.
Finally, a 100GB array for 50K users results in a 2 megabyte mailbox..that's freaking ridiculous!
In short, you're either running a 50-user shop, or you have no idea what you are talking about.
Wrong. All Windows-based clients automatically get a CAL to utilize Terminal Services. You are only paying for the server software, in most cases.
If you are using dedicated thin-client hardware, it most likely ships with the CAL built right in.
Agreed...Not just art, but vision. The NeXT was a harbinger of things to come (that never did..alas), a bold vision of the future. I remember when there was a NeXT dealer in downtown Toronto. Us developers would go down on a hot, lazy, afternoon and gawk at the absolute beauty and precision of those machines. We were developing on generic 386's, running OS/2 1.3, using Smalltalk. Win95, NT, OS/2, and Linux were blips on the horizon, but there they were..black, powerful, and pure geek lust. They were the most futuristic looking, and most futuristically capable machines out there. They made all the high end offerings (like the RS/6000) look primitive, and made our 386's look just plain pathetic.
Now, everybody has machines 20x or more powerful, minus the grace and elegance (the iMac cube came close, but cutesy can't hold a candle to how the NeXT Cubes looked back then), and we still haven't achieved the panache, both visual and hands-on, that these things achieved.
Fortunately, here in Calgary, there is a certain oil company that still runs NextStep, although it is being phased out. Talking to the developers, to a person they nearly cry lamenting their phasing out.
Truly the passing of a legend. I'm not sure whether to be outraged that the folks in the article burnt one, or to be proud watching a Viking warrior go out in a burning effigy...
Which would the boxes themselves have wanted? I hope the latter...
Seriously people... Most, if not all, broadband providers prohibit running servers from home accounts (it's definitely that way for @Home users, even if they do generally turn a blind eye to small time web servers). They generally also have some sort of clause which basically doesn't guarantee unlimited or uncontrolled inbound or outbound access. For that matter, most broadband (and thinband) providers provide a clause which basically exempts them from any sort of service level agreement.
Signing on with a domestic oriented ISP means that you are essentially "users" on their network. Blocking inbound port 80 access is a good starting point for at least protecting their internal network segments. If you were running what is essentially a DHCP/DNS/proxy service for thousands of users, wouldn't you at least take this step to protect the integrity of your network?? (I admit it doesn't begin to solve all the problems, but...)
If you want to run your own "mini NOC", then pony up the cash and get ISDN, a T1, or something faster put into your basement. But if you are subscribing to a consumer grade ISP's offerings, don't be suprised when this happens. And especially don't start with the geek indignation, because consumer broadband is not meant, nor sold, under the pretense of running home servers.
It's a slim, expensive tome, but absolutely indispensable. It's almost impossible to not come into contact with C, especially if you are an Open Source user.
It's also practically a part of the geek heritage, both in the style in which it is written, and in the impact it has had on generations of coders. It is truly one of the underpinnings of a great part of Information Technology history.
From his description of how it works: "This method of optical feedback makes for very reproducible coffee strength, independent of the amount of coffee grinds used"
I must be missing something here. If I put in a small amount of coffee grinds, and dial in strong coffee, how in the hell is pouring more water over the grinds going to ever make the coffee sufficiently strong? Coffee makers only extract so much from the grinds. I know this, because I've tried reusing grinds in the past, when really desparate (and broke..). All that you will end up with is lots of weak, crappy tasting coffee. Even percolators, which continuously flow the partially made coffee back over the grinds can only achieve certain strength coffee based on a set amount of grinds.
What it boils down to (no pun intended) is that you _must_ put in enough grinds for whatever strength brew you are looking for.
That, my friends, is the law of conservation of coffee.
At least you understand what I'm saying...hell, at work , we throw together small HP servers (like the LPr) in about 1/2 hour...the point is that nobody should be whining about how hard/how much time it takes to put a machine together these days..I remember pushing roughly 32 individual RAM chips into my 386 motherboard, to get a grand total of 8 megabytes of ram..given that they were fragile, and it also cost me over $800, it took well over two hours:)
You don't have the time to build a system? Good grief...it takes about 20 minutes to get a mobo into a case, about another 20-30 minutes to install the drives (HD, CD, Floppy), and about 5 minutes to drop the cards in (video, NIC, sound). Wiring should take about 20 minutes, if you are doing it carefully (no backwards IDE cables on the first try), and neatly. That's a grand total of less than two hours.
Compare that with doing an install of Linux, coupled with the requisite kernel build, module installation, etc, and overall, the hardware build is most likely going to be the shortest part of the build. Please don't tell me you just slap on a default install of Redhat or whatever, because at that point, if you're looking for a pre-built machine and a default install of an OS, you might as well be buying a Dell with Windows ME pre-installed. Linux distros, out of the box, need a ton of work to become usable (as does Win2K, no bias implied)
One neat thing that this chipset supports is the so-called "twin-bank" technology. This interleaves access to the DDR DIMMS (ie. byte 1 is on DIMM1, byte #2 in on DIMM#2, byte #3 in on DIMM #3) to pump up the maximum memory bandwidth beyond what a single DDR DIMM would be capable of. I guess they had to do this since the GPU uses system memory (ugh), but it has a nice side-effect of really unleashing the performance potential of 1ghz+ CPU's, especially if you ditched the onboard video and stuck something decent in the AGP slot.
And to think, my 386 had interleaved memory back in the day..I was 'leet and didn't even know it;)
I don't understand..the goal of an audiophile is to attempt to reproduce the sound of live music, which, ironically, was recorded using audio equipment, and subsequently mixed in the artificial confines of a studio..let's not even begin to contemplate what exactly audiophiles are attempting to reproduce the experience of when playing stuff recorded in the studio, where the creation of music has basically nothing to do with playing live...
Slashdot Mirror
Chassis - The car that wins is going to have very carefully measured and constructed alignment. Getting the vehicle to run 20 meters straight, and with the thrust exactly aligned (in whichever direction the particular design calls for) is key. You can't afford to waste any of those thrust newtons on trying to drive the car sideways, or having the car trade forward velocity for lateral movement.
Water/Air ratio - if you run out of air too quickly, you are hauling excess water all the way down the track. Similarly, if you design the vehicle such that air can escape freely once the water level falls below a certain point, you are wasting power.
"Gearing" - one has to wonder if a direct-drive design (air pushing water out the back) is the most efficient. If you used hydraulic principles, you could in effect "gear down" the high pressure, resulting in high-torque that could drive over-size wheels, similar to the way that a hydraulic lift works. The winning design is going to have to find a tradeoff between quick accelleration/coasting and continued power for the duration of the track. Could the careful design of a nozzle accomplish the same thing? (/me thinks back to the model rocket days)
Multiple pressure vessels - this is just fanciful thinking, but one of the techniques steam locomotives used was to use the high pressure air in one set of cylinders, and then re-use the resulting lower pressure to drive a second set of cylinders. Extrapolating from this, I wonder if having two pressure vessels, totalling the legal limit, and firing at different times (ie. one for acceleration, one for maintaining speed) might be feasible.
I guess, though, in the end, the simplest, lightest design will win. Having a good chassis is still key though.
This is only the second kernel I will have built (just installed Slackware 8.0 for the first time, and built 2.4.11)...Can I reuse the configuration file created by "make menuconfig" with 2.4.12, or should I try and re-select all the options I had previously?
A few ideas:
"it's just that there are more and more applications that REQUIRE a GUI, or that are badly designed into using one "
- not really. More and more people are shoehorning applications into the WIMP model, whether they fit or not. That isn't the same as applications that "REQUIRE" a GUI. Badly designed applications are badly designed..If people choose to use them, what can you do? A better line of inquiry would be "why are most applications going to the WIMP model, appropriately or not?"...I'd hazard a guess that it's because more efficient or not, it's easier for the average user to understand and operate.
"A good example is the browser syndrome, where keyboard shortcuts are a no-go in most cases -- GUI systems like Motif, GTK, or even Win32, are well thought out. But remember, the original goal of GUIs was to facilitate _data representation_, i.e.: make it more easily assimilable by the person, NOT easier input"
"In the case of applications like CAD, graphics design, etc... a pointer-style input device makes sense -- but we NEED to solve the Repetitive Strain problem "
- Its not the mouse or the GUI that is causing RSI. If it was, why so many ergonomic keyboards? Why aren't there any ergonomic guitars? RSI is caused by bad ergonomics, nothing more, nothing less. If an application was poorly enough designed that it cause RSI, very few people would be able to use it for any length of time (which in fact makes me wonder about mouselook in FPS games, and what that is doing to a generation of computer users)
Unix is chock full of hard to reach keyboard symbols that have significant meanings (and that are needed to run certain commands) For all that geeks are supposed to cherish keystroke economy, there are a suprising number of conventions that require removing your hands from the home row, or using alt/ctrl/escape, etc to modify the default meaning of keystrokes.
Use the right tool for the right job. That's the UNIX/Linux philosophy. If you think that the WIMP interface is cramping your style with respect to whatever task you are trying to accomplish, then you are using the wrong tool.
With regards to the command line or WIMP interfaces being old, and not particularly forward looking, you are also missing a fundamental point: A graphical "pipe" isn't innovative either. You're simply shoehorning two paradigms together, and even worse, two totally incompatible paradigms at that. The pipe is a useful metaphor and operator for stream-oriented I/O. The WIMP is useful for (obviously) visually oriented information, and its designed for a completely different purpose than the pipe. The WIMP is designed to allow humans to manipulate data and abstract objects in a visual manner. The pipe is designed to allow users to allow the computer to do the same, without intervention.
If you want an innovative computing interface, worrying about streams, or visual representations of data is a waste of time. You're going to have to come up with something totally new. One good example is the use of sound to communicate the health and performance of networks or systems.
There's some significant insight contained in this post. I work for a company that, stats wise, generates 4x the volume of email of any other company in Canada, on a per-employee basis (we've been doing email on IBM mainframes since before most companies had computers).
Email is much more than just another form of messaging. I've seen email used as a form of decision records, as a primitive form of version control (the email thread contains each revision of the document in question), as discussion threads, and even as a form of middleware for some very significant applications (like train dispatching). In a company like mine, end-to-end delivery of email messages that exceeds 30 seconds is seen as a serious degradation of service (no kidding!)
Consitent, timely, and reliable delivery of email in large companies (outside, perhaps, of large dot.com online sales companies) is arguably more important than nearly any other form of networking.
If you check the CNN article, it seems that Reuters is confirming that it was a helicopter belonging to the local opposition...
9 /2 2/ret.afghan.plane/
http://www.cnn.com/2001/WORLD/asiapcf/central/0
Understood. I did find it funny at first, but it's hard not to be overly sensitized to stuff like this lately.
Thanks.
We were playing Dune 2 long before Warcraft2 was a dot on the map
Why is this rated +4, funny? If the poster is describing an experience where he gamed and drank all night long, then left the LAN party the next morning, then he is likely describing driving drunk...If he is describing driving the day after (ie. assuming he got some sleep), then his description still fits that of impaired driving.
I don't think impaired driving is funny, particulary because of the terrible loss of life that can result.
There's no way Exchange2K could handle 50K users on a single box.
First, you've never obviously worked with Fibre Channel on the kind of scale that 50K users would require (ie. a big EMC box)...that many users pounding the same box will easily chew up 50% or more of your CPU power on I/O alone. Fibre is fast, but it is so fast that it can easily swamp Xeon CPU's. I know, because I did the benchmarking at my company.
Second, connection limitations in Win2K and Exchange alone mean that you are running very close to the theoretical maximum the OS supports..not a good idea.
Third, running that many users off of a single box is suicide. And if you've ever watched Exchange2K failover on a Win2K cluster, you'd know that it can take several minutes for everything to come up on the second node, if you've got a lot of users.
Finally, a 100GB array for 50K users results in a 2 megabyte mailbox..that's freaking ridiculous!
In short, you're either running a 50-user shop, or you have no idea what you are talking about.
It logs keystrokes, so that they can later tell exactly what you were typing at the time!
wow..do you want some sort of award, or were you just hoping to karma troll a point or two??
Too late! The beowulf meme strikes again...
I'll make sure that it gets handled :)
Email me for the mailing address.
Wrong. All Windows-based clients automatically get a CAL to utilize Terminal Services. You are only paying for the server software, in most cases. If you are using dedicated thin-client hardware, it most likely ships with the CAL built right in.
Agreed...Not just art, but vision. The NeXT was a harbinger of things to come (that never did..alas), a bold vision of the future. I remember when there was a NeXT dealer in downtown Toronto. Us developers would go down on a hot, lazy, afternoon and gawk at the absolute beauty and precision of those machines. We were developing on generic 386's, running OS/2 1.3, using Smalltalk. Win95, NT, OS/2, and Linux were blips on the horizon, but there they were..black, powerful, and pure geek lust. They were the most futuristic looking, and most futuristically capable machines out there. They made all the high end offerings (like the RS/6000) look primitive, and made our 386's look just plain pathetic.
Now, everybody has machines 20x or more powerful, minus the grace and elegance (the iMac cube came close, but cutesy can't hold a candle to how the NeXT Cubes looked back then), and we still haven't achieved the panache, both visual and hands-on, that these things achieved.
Fortunately, here in Calgary, there is a certain oil company that still runs NextStep, although it is being phased out. Talking to the developers, to a person they nearly cry lamenting their phasing out.
Truly the passing of a legend. I'm not sure whether to be outraged that the folks in the article burnt one, or to be proud watching a Viking warrior go out in a burning effigy...
Which would the boxes themselves have wanted? I hope the latter...
Seriously people... Most, if not all, broadband providers prohibit running servers from home accounts (it's definitely that way for @Home users, even if they do generally turn a blind eye to small time web servers). They generally also have some sort of clause which basically doesn't guarantee unlimited or uncontrolled inbound or outbound access. For that matter, most broadband (and thinband) providers provide a clause which basically exempts them from any sort of service level agreement.
Signing on with a domestic oriented ISP means that you are essentially "users" on their network. Blocking inbound port 80 access is a good starting point for at least protecting their internal network segments. If you were running what is essentially a DHCP/DNS/proxy service for thousands of users, wouldn't you at least take this step to protect the integrity of your network?? (I admit it doesn't begin to solve all the problems, but...)
If you want to run your own "mini NOC", then pony up the cash and get ISDN, a T1, or something faster put into your basement. But if you are subscribing to a consumer grade ISP's offerings, don't be suprised when this happens. And especially don't start with the geek indignation, because consumer broadband is not meant, nor sold, under the pretense of running home servers.
It's a slim, expensive tome, but absolutely indispensable. It's almost impossible to not come into contact with C, especially if you are an Open Source user. It's also practically a part of the geek heritage, both in the style in which it is written, and in the impact it has had on generations of coders. It is truly one of the underpinnings of a great part of Information Technology history.
From his description of how it works: "This method of optical feedback makes for very reproducible coffee strength, independent of the amount of coffee grinds used"
I must be missing something here. If I put in a small amount of coffee grinds, and dial in strong coffee, how in the hell is pouring more water over the grinds going to ever make the coffee sufficiently strong? Coffee makers only extract so much from the grinds. I know this, because I've tried reusing grinds in the past, when really desparate (and broke..). All that you will end up with is lots of weak, crappy tasting coffee. Even percolators, which continuously flow the partially made coffee back over the grinds can only achieve certain strength coffee based on a set amount of grinds.
What it boils down to (no pun intended) is that you _must_ put in enough grinds for whatever strength brew you are looking for.
That, my friends, is the law of conservation of coffee.
At least you understand what I'm saying...hell, at work , we throw together small HP servers (like the LPr) in about 1/2 hour...the point is that nobody should be whining about how hard/how much time it takes to put a machine together these days..I remember pushing roughly 32 individual RAM chips into my 386 motherboard, to get a grand total of 8 megabytes of ram..given that they were fragile, and it also cost me over $800, it took well over two hours :)
You don't have the time to build a system? Good grief...it takes about 20 minutes to get a mobo into a case, about another 20-30 minutes to install the drives (HD, CD, Floppy), and about 5 minutes to drop the cards in (video, NIC, sound). Wiring should take about 20 minutes, if you are doing it carefully (no backwards IDE cables on the first try), and neatly. That's a grand total of less than two hours.
Compare that with doing an install of Linux, coupled with the requisite kernel build, module installation, etc, and overall, the hardware build is most likely going to be the shortest part of the build. Please don't tell me you just slap on a default install of Redhat or whatever, because at that point, if you're looking for a pre-built machine and a default install of an OS, you might as well be buying a Dell with Windows ME pre-installed. Linux distros, out of the box, need a ton of work to become usable (as does Win2K, no bias implied)
Either way, pre-installs suck
One neat thing that this chipset supports is the so-called "twin-bank" technology. This interleaves access to the DDR DIMMS (ie. byte 1 is on DIMM1, byte #2 in on DIMM#2, byte #3 in on DIMM #3) to pump up the maximum memory bandwidth beyond what a single DDR DIMM would be capable of. I guess they had to do this since the GPU uses system memory (ugh), but it has a nice side-effect of really unleashing the performance potential of 1ghz+ CPU's, especially if you ditched the onboard video and stuck something decent in the AGP slot.
;)
And to think, my 386 had interleaved memory back in the day..I was 'leet and didn't even know it
Major Havoc was great! Totally captivating at the time, compared to the other pablum that was around.
I don't understand..the goal of an audiophile is to attempt to reproduce the sound of live music, which, ironically, was recorded using audio equipment, and subsequently mixed in the artificial confines of a studio..let's not even begin to contemplate what exactly audiophiles are attempting to reproduce the experience of when playing stuff recorded in the studio, where the creation of music has basically nothing to do with playing live...
Truly amazing.
I guess a while(TRUE) would have been more appropriate...