For instance, Sympatico DSL here in Canada has chosen to use this awful PPP-over-Ethernet technology to share the lines. I'd prefer to use Sympatico over Rogers, cause I've mostly gotten better service, but the PPPoE is just too much hassle.
Hmmmm... I plugged in my linksys DSL router/hub, clicked the PPPoE radio button in the web setup for the router, entered my username and password and have since forgotten about the fact that my DSL is PPPoE. Hassle? Not in the least.
Earthlink's official response was that AT&T would purposely detect non-AT&T-DSL customers and downgrade their connection somehow. Of course I'm not sure I believe them, because the daily outages only seemed to be happening during peak hours.
SBC/Ameritech is the monopoly local phone provider in Chicago - not AT&T. AT&T has been planning to try to offer local phone service over it's cable lines in Chicago every since it bought the biggest cable company here but hasn't done much. So anyway, I doubt very much AT&T was responsible for the DSL problems you observed
The theory goes that anti-hyrdogen should have all the same observable physical properties that hydrogen does. If we can start to manufacturer and store non-trivial quantities of the stuff we can actually start to test whether or not this is true. We can see if it has the same obsorbtion spectrum as hydrogen, the same atomic weight, etc...
If there is a difference we might be able to use it to confirm or disprove our assumption that the entire universe is made of 'normal' matter. For example, if there is an observable difference between the absorbtion spectra of hydrogen and anti-hydrogen, we'd have a test to determine if a distant galaxy was made of anti-matter. If there is no difference, well, we've found a very expensive way to heat a small cup of coffee.
-josh
Why doesn't Sun adapt?
on
What is .NET?
·
· Score: 2
Why can't Sun see the writing on the wall and adapt, coopting some of the innovations Microsoft has surprised us with.
Take the JVM specification and broaden it. Add features that would more fully support the retargetting of other languages to the JVM. Provide better native libraries for GUI interfaces so we can produce GUI apps that don't require GHz machines to run well. And for christ sake, be prepared to sacrifice some of your freakin object oriented purity for performance every once in awhile.
The CLR is really just a more pragmatic JVM. It suggests best practices, and highly encourages them, but allows for backward compatibility and 'unsafe' operations if you think you know what you are doing, or have no other choice. It has more flexibility that the JVM and the java spec allow. This was always the crux of the disagreement MS and Sun had re: Java. MS wanted to allow the developers to chose, Sun thought they knew best.
-josh
Re:Something that isn't pointed out enough
on
SuSE 7.3 vs XP
·
· Score: 1
I meant multiple users with concurrent desktops. XP lets you 'log out' of your desktop, but keep all of your programs running. Then another user can log in. When they log off, you can go right back to where you left off.
I imagine this would be doable in X, perhaps with nested X servers, but noone has put it all together with a nice swapper interface.
-josh
Re:Something that isn't pointed out enough
on
SuSE 7.3 vs XP
·
· Score: 1
I LIKE this feature, and it's great in a multi-user/single computer situation, just tends to slow the machine down after switching. Seems the switcher could cooperate with the memory manager and force all the other user's processes to swap out. I imagine that would happen eventually as the new user worked, but in the mean time its slow.
-josh
Re:Something that isn't pointed out enough
on
SuSE 7.3 vs XP
·
· Score: 1
256MB seems to be the sweet spot for XP, but it wants more. I use it on a laptop I share with my GF. We use the multiple logins/switch user feature a lot (something I wish someone would figure out how to do in Linux). The machine can slow down considerably when I leave a ton of programs running and she logs in. Granted, windows 2000 does not have this feature, but with 256MB i've never been able to slow down w2k, no matter how much I keep open.
Unfortunately my Vaio doesn't support more than 256MB, otherwise it'd have 512MB.
Ok, computers used to be a great motivational tool, because they were a novelty. Kids would use them because they were new and cool. Well, wake up folks, its a new century and just about everyone who wants one can have one at home. Most kids (even poor kids) grow up with one now. It's nothing new, and just because you put your stupid flashcards on a computer doesn't mean Johnie is going to want to learn.
The novelty of computers has worn off, there is no magic bullet here. Teaching is all about the basics. Lets face it, some things are hard to learn, and even harder to teach, and no computer is going to take the place of a trained and creative human being.
School districts that waste tax dollar buying laptops for every student pain me no end. These are teaching tools, no more, no less, and there is no value in a 1-1 computer student ratio, anymore than there is value in a 1-1 blackboard to student ratio.
Certainly computer skills should be taught, just like reading skills, math skills and arts are taught. But there is no value to allowing computers to encroach on other subject matters, no value in allowing computers to be the delivery mechanism for all information. A learning and research tool, no doubt, but the end all and be of education they are not.
-josh
A Minor update to a development kernel?
on
2.5.4 Kernel Out
·
· Score: 1, Flamebait
Evolution has not stopped. Certainly the enviornment, or fitness plane, that human's inhabit has changed, and changed radically, but this merely changes the constraints that determine who dies and who lives.
For example. Modern science now allows women with unnaturally narrow hips to survive child birth (Cesaerian Section). This allows a new set of genetic material to be passed down through the generations - perhaps there are some other beneficial adaptations that are associated with narrow hips.
There are many other examples. Just because modern science allows some new sets of genes to replicate themselves - does not mean evolution has stopped - merely that different selective pressures can now come into play.
Think about it. Evolution (sorry to anthropomorphize here) is now free to play with a lot more vairables than it had before. For example. Since we can deliver almost any baby now, will there be a trend towards bigger babies, since the added drain on resources will no longer hurt the mother's chances of survival - even a 16lbs could be delivered via C section. Will bigger babies have a head start - start talking early, have bigger brains?
There are myriad other examples of this line of thinking.
I can reliable get 5-6 hours with my extended life battery on my vaio under heavy usage. Under typical usage it would go 10 hours.
If battery usage were really an issue with most laptop users, manufacturers could easily hit the 10 hour mark with more efficient/dimmer backlights and underclocked processors (no one needs 1GHz in a laptop anyway).
The problem is that its a rare laptop user that isn't far from an outlet. Sure, some people want to take a jaunt down to the beach to work on The Great American Novel for 10 hours - but those people are hardly enough to provide a strong market for fuel cells in laptops.
This brings to mind a serious question. What data, if any, do we throw away? With ever expanding storage capacities it's getting easier and easier to just keep stuff, than to sit down and figure out what you want to throw away.
In the past media degeneration and obscelecense over time have made the decisions for us. But going forward we will have massive distributed, redundant data stores, with geographically remote backups. The data isn't going to go away unless we tell it to.
Freenet addresses this problem by culling the less popular data (not actively, but as a end result of its caching policies) - but this has the unfortunate effect that important data can get lost. Not a desirable behavior for corporate data.
You can pick up an m100 real cheap, plenty of memory, and get some great graphing software relatively cheaply (check out powerOne graphing software for the Palm). There are also some great, free RPN calcs for the Palm, and I am sure many many more other programs than will every be available for the TI.
I dunno, but even on broadband connections I have never viewed an acceptable video stream based on a RealNetworks codec. I keep giving them a chance, downloading the new required viewer every other month, click 'No, cancel, no, exit' every time their viewer loads and prompts me to register. But their codecs suck.
On the exact same connection I can get near VHS quality streams (BBC online is a great example with their 300 Kbps feed) using windows media. I've tried many different Real feeds that claim the same bandwidth targets, and I've yet to see one that is watchable. I wish Real were better, but it not even in the same ballpark.
I think Real has done more to give streaming video a band name, than any other company out there.
Chicago already has many broadband options. Ameritech/SBC provides DSL (or another provider resells it), and there is a crazy company (RCN) that came through and re-wired thousands of buildings for their fiber to the telephone pole broadband/cable solution. AT&T "broadband" also might get around to providing data services to a meaningful population of it's Chicago subscribers sometime soon.
Most of these solutions do lack geographical reach. Good coverage is confined to the mostly affluent and gentrified portions of the city. I doubt even Ameritech, which already has a wire in most houses, has good DSL coverage in lower income areas, it just can't be a priority for them. This is where the city government can provide a much needed impetus. They can provide the motivation to provide a combination of services and locations that might be commercially marginal at best.
Of course the appearance of high speed internet cafes on the south side of Chicago might backfire on the liberal set, as they find that the locals they hoped to benefit are displaced by yuppies moving in to take advantage of all that cheap bandwidth.
Is there any reliable way to detect the characteristic activities of a keylogger? Rather than trusting a virus scanner, or trying to keep every possible back door fixed, I would like a utility that would look for suspicious activity indicative of such a key logging attack. I am assuming though that this would be relatively operating system dependent.
Beyond this, are there ways of making the operating system itself immune to keylogging? In windows this might be a custom keyboard driver. In Linux perhaps a kernel module.
No matter what you do they can always log at the hardware level (essentially bug your keyboard), but it'd be nice to make it as hard as possible for them.
So, you know those calls were your friend puts her phone in her pocket without locking the keypad, and accidentally calls you? Now you get to SEE the inside of her pocket, instead of just hearing it...
I really doubt that your method of numeric integration is going to be that critical an influence on the quality of game play. In fact I think that you ability to simulate physics without really doing the math would be more important. I doubt anyone is going to be timing a car's drop off of a cliff in Carmegeddon to determine what gravitational constant is used in the game.
I think it is more important to include as many effects as you can: gravity, linear momentum, angular momentum, elastic/inelastic collisions, friction (surface and wind), than it is to model the effects perfectly.
In fact one could argue that it is to the game designer's benefit to use an innaccurate and exaggerated physics model. Most real world collisons with the guard rail on a race course are relatively unspectacular (by design) - but that would be oh so boring in a racing game now wouldn't it?
With great reluctance, after failing in two other attempts to get other providers to come out and install DSL in my apartment, I decided to go with Ameritech DSL when I moved into a new place.
Other than the fact that they sent me a USB modem instead of the Ethernet version (quickly corrected) service has been excellent and bandwidth consistantly high and reliable.
I was rather surpised, until I realized it wasn't really the old Ameritech (which had managed to roll out DSL to several thousand subscribers in Michigan in 3 years), this is SBC, which has a much better record in the DSL marketplace.
It's good service and I know SBC is not going anywhere anytime soon Granted I don't have quite the speed you do, but maybe offering such speed is one of the reasons Covad went bankrupt in the first place.
Yeah, I don't ftp is so slow that anyone is going to pay $70k for their proprietary 'Transporter Fountains'. Seems like anyone with a little common sense and math ability could easily cobble together a software UDP based transfer protocol that has all of the properties described in the article.
The key is to build in redundancy without increasing the amount of data sent so much that you counteract the speed gains you get by using UDP.
I majored in physics and math and took many majors level course outside my major. My GPA was one of the best in my class.
What does this mean? Absolutely nothing. I do well at school, at test taking, and at figuring out what the hell the prof wants. I does not however mean I took the path of least resistance. Though I am sure you would love to believe that of all those who got a better GPA than yourself.
Many people who were smarting than myself got worse grade point averages, because they did not care to always do what it took to get the A. There were also few idiot who were right up there with me, who just worked so damned hard the profs took pity on them even though they understood nothing.
The article says that they observed the microlensing event several years back and then recently took another look at the area using Hubble. They found a faint red dwarf which probably cause the lensing of the nearby (arc second wise) main sequence star.
So, if we can observe the 'dark' matter as being a red dwarf, it's not exactly 'dark' is it? I would assume that objects like red dwarfs, if observable, would have already been counted in the total 'bright matter' column. If not, someone is just undercounting objects that are observable using normal astronomical methods, and needs to go back and make a better estimate of how many of them are out there.
"My theory is that this happens because it's harder to read code than to write it."
He couldn't be more right. I've recently been asked to port some code from another group in the company. Upon first reading it, I found global variables being referenced from everywhere, and it looked terrible.
I am usually the opposite, extremely leery of just starting over, no matter how bad the old code is.
The thing is once I have actually modify the code I realize it probably would have been better and faster to start from scratch.
Slashdot Mirror
For instance, Sympatico DSL here in Canada has chosen to use this awful PPP-over-Ethernet technology to share the lines. I'd prefer to use Sympatico over Rogers, cause I've mostly gotten better service, but the PPPoE is just too much hassle.
Hmmmm... I plugged in my linksys DSL router/hub, clicked the PPPoE radio button in the web setup for the router, entered my username and password and have since forgotten about the fact that my DSL is PPPoE. Hassle? Not in the least.
Earthlink's official response was that AT&T would purposely detect non-AT&T-DSL customers and downgrade their connection somehow. Of course I'm not sure I believe them, because the daily outages only seemed to be happening during peak hours.
SBC/Ameritech is the monopoly local phone provider in Chicago - not AT&T. AT&T has been planning to try to offer local phone service over it's cable lines in Chicago every since it bought the biggest cable company here but hasn't done much. So anyway, I doubt very much AT&T was responsible for the DSL problems you observed
The theory goes that anti-hyrdogen should have all the same observable physical properties that hydrogen does. If we can start to manufacturer and store non-trivial quantities of the stuff we can actually start to test whether or not this is true. We can see if it has the same obsorbtion spectrum as hydrogen, the same atomic weight, etc...
If there is a difference we might be able to use it to confirm or disprove our assumption that the entire universe is made of 'normal' matter. For example, if there is an observable difference between the absorbtion spectra of hydrogen and anti-hydrogen, we'd have a test to determine if a distant galaxy was made of anti-matter. If there is no difference, well, we've found a very expensive way to heat a small cup of coffee.
-josh
Why can't Sun see the writing on the wall and adapt, coopting some of the innovations Microsoft has surprised us with.
Take the JVM specification and broaden it. Add features that would more fully support the retargetting of other languages to the JVM. Provide better native libraries for GUI interfaces so we can produce GUI apps that don't require GHz machines to run well. And for christ sake, be prepared to sacrifice some of your freakin object oriented purity for performance every once in awhile.
The CLR is really just a more pragmatic JVM. It suggests best practices, and highly encourages them, but allows for backward compatibility and 'unsafe' operations if you think you know what you are doing, or have no other choice. It has more flexibility that the JVM and the java spec allow. This was always the crux of the disagreement MS and Sun had re: Java. MS wanted to allow the developers to chose, Sun thought they knew best.
-josh
I meant multiple users with concurrent desktops. XP lets you 'log out' of your desktop, but keep all of your programs running. Then another user can log in. When they log off, you can go right back to where you left off.
I imagine this would be doable in X, perhaps with nested X servers, but noone has put it all together with a nice swapper interface.
-josh
I LIKE this feature, and it's great in a multi-user/single computer situation, just tends to slow the machine down after switching. Seems the switcher could cooperate with the memory manager and force all the other user's processes to swap out. I imagine that would happen eventually as the new user worked, but in the mean time its slow.
-josh
256MB seems to be the sweet spot for XP, but it wants more. I use it on a laptop I share with my GF. We use the multiple logins/switch user feature a lot (something I wish someone would figure out how to do in Linux). The machine can slow down considerably when I leave a ton of programs running and she logs in. Granted, windows 2000 does not have this feature, but with 256MB i've never been able to slow down w2k, no matter how much I keep open.
Unfortunately my Vaio doesn't support more than 256MB, otherwise it'd have 512MB.
-josh
Ok, computers used to be a great motivational tool, because they were a novelty. Kids would use them because they were new and cool. Well, wake up folks, its a new century and just about everyone who wants one can have one at home. Most kids (even poor kids) grow up with one now. It's nothing new, and just because you put your stupid flashcards on a computer doesn't mean Johnie is going to want to learn.
The novelty of computers has worn off, there is no magic bullet here. Teaching is all about the basics. Lets face it, some things are hard to learn, and even harder to teach, and no computer is going to take the place of a trained and creative human being.
School districts that waste tax dollar buying laptops for every student pain me no end. These are teaching tools, no more, no less, and there is no value in a 1-1 computer student ratio, anymore than there is value in a 1-1 blackboard to student ratio.
Certainly computer skills should be taught, just like reading skills, math skills and arts are taught. But there is no value to allowing computers to encroach on other subject matters, no value in allowing computers to be the delivery mechanism for all information. A learning and research tool, no doubt, but the end all and be of education they are not.
-josh
This must be a slow geek news day...
Evolution has not stopped. Certainly the enviornment, or fitness plane, that human's inhabit has changed, and changed radically, but this merely changes the constraints that determine who dies and who lives.
For example. Modern science now allows women with unnaturally narrow hips to survive child birth (Cesaerian Section). This allows a new set of genetic material to be passed down through the generations - perhaps there are some other beneficial adaptations that are associated with narrow hips.
There are many other examples. Just because modern science allows some new sets of genes to replicate themselves - does not mean evolution has stopped - merely that different selective pressures can now come into play.
Think about it. Evolution (sorry to anthropomorphize here) is now free to play with a lot more vairables than it had before. For example. Since we can deliver almost any baby now, will there be a trend towards bigger babies, since the added drain on resources will no longer hurt the mother's chances of survival - even a 16lbs could be delivered via C section. Will bigger babies have a head start - start talking early, have bigger brains?
There are myriad other examples of this line of thinking.
Evolution CAN'T stop.
-josh
I can reliable get 5-6 hours with my extended life battery on my vaio under heavy usage. Under typical usage it would go 10 hours.
If battery usage were really an issue with most laptop users, manufacturers could easily hit the 10 hour mark with more efficient/dimmer backlights and underclocked processors (no one needs 1GHz in a laptop anyway).
The problem is that its a rare laptop user that isn't far from an outlet. Sure, some people want to take a jaunt down to the beach to work on The Great American Novel for 10 hours - but those people are hardly enough to provide a strong market for fuel cells in laptops.
-josh
This brings to mind a serious question. What data, if any, do we throw away? With ever expanding storage capacities it's getting easier and easier to just keep stuff, than to sit down and figure out what you want to throw away.
In the past media degeneration and obscelecense over time have made the decisions for us. But going forward we will have massive distributed, redundant data stores, with geographically remote backups. The data isn't going to go away unless we tell it to.
Freenet addresses this problem by culling the less popular data (not actively, but as a end result of its caching policies) - but this has the unfortunate effect that important data can get lost. Not a desirable behavior for corporate data.
-josh
You can pick up an m100 real cheap, plenty of memory, and get some great graphing software relatively cheaply (check out powerOne graphing software for the Palm). There are also some great, free RPN calcs for the Palm, and I am sure many many more other programs than will every be available for the TI.
-josh
I dunno, but even on broadband connections I have never viewed an acceptable video stream based on a RealNetworks codec. I keep giving them a chance, downloading the new required viewer every other month, click 'No, cancel, no, exit' every time their viewer loads and prompts me to register. But their codecs suck.
On the exact same connection I can get near VHS quality streams (BBC online is a great example with their 300 Kbps feed) using windows media. I've tried many different Real feeds that claim the same bandwidth targets, and I've yet to see one that is watchable. I wish Real were better, but it not even in the same ballpark.
I think Real has done more to give streaming video a band name, than any other company out there.
Perhaps TiVO can figure out what's wrong.
-josh
Chicago already has many broadband options. Ameritech/SBC provides DSL (or another provider resells it), and there is a crazy company (RCN) that came through and re-wired thousands of buildings for their fiber to the telephone pole broadband/cable solution. AT&T "broadband" also might get around to providing data services to a meaningful population of it's Chicago subscribers sometime soon.
Most of these solutions do lack geographical reach. Good coverage is confined to the mostly affluent and gentrified portions of the city. I doubt even Ameritech, which already has a wire in most houses, has good DSL coverage in lower income areas, it just can't be a priority for them. This is where the city government can provide a much needed impetus. They can provide the motivation to provide a combination of services and locations that might be commercially marginal at best.
Of course the appearance of high speed internet cafes on the south side of Chicago might backfire on the liberal set, as they find that the locals they hoped to benefit are displaced by yuppies moving in to take advantage of all that cheap bandwidth.
-josh
Is there any reliable way to detect the characteristic activities of a keylogger? Rather than trusting a virus scanner, or trying to keep every possible back door fixed, I would like a utility that would look for suspicious activity indicative of such a key logging attack. I am assuming though that this would be relatively operating system dependent.
Beyond this, are there ways of making the operating system itself immune to keylogging? In windows this might be a custom keyboard driver. In Linux perhaps a kernel module.
No matter what you do they can always log at the hardware level (essentially bug your keyboard), but it'd be nice to make it as hard as possible for them.
-josh
So, you know those calls were your friend puts her phone in her pocket without locking the keypad, and accidentally calls you? Now you get to SEE the inside of her pocket, instead of just hearing it...
-josh
Now do the math for television. I imagine the average in the US is at least an hour a day of television viewing. Truly a waste.
-josh
I really doubt that your method of numeric integration is going to be that critical an influence on the quality of game play. In fact I think that you ability to simulate physics without really doing the math would be more important. I doubt anyone is going to be timing a car's drop off of a cliff in Carmegeddon to determine what gravitational constant is used in the game.
I think it is more important to include as many effects as you can: gravity, linear momentum, angular momentum, elastic/inelastic collisions, friction (surface and wind), than it is to model the effects perfectly.
In fact one could argue that it is to the game designer's benefit to use an innaccurate and exaggerated physics model. Most real world collisons with the guard rail on a race course are relatively unspectacular (by design) - but that would be oh so boring in a racing game now wouldn't it?
-josh
With great reluctance, after failing in two other attempts to get other providers to come out and install DSL in my apartment, I decided to go with Ameritech DSL when I moved into a new place.
Other than the fact that they sent me a USB modem instead of the Ethernet version (quickly corrected) service has been excellent and bandwidth consistantly high and reliable.
I was rather surpised, until I realized it wasn't really the old Ameritech (which had managed to roll out DSL to several thousand subscribers in Michigan in 3 years), this is SBC, which has a much better record in the DSL marketplace.
It's good service and I know SBC is not going anywhere anytime soon Granted I don't have quite the speed you do, but maybe offering such speed is one of the reasons Covad went bankrupt in the first place.
-josh
Yeah, I don't ftp is so slow that anyone is going to pay $70k for their proprietary 'Transporter Fountains'. Seems like anyone with a little common sense and math ability could easily cobble together a software UDP based transfer protocol that has all of the properties described in the article.
The key is to build in redundancy without increasing the amount of data sent so much that you counteract the speed gains you get by using UDP.
-josh
I majored in physics and math and took many majors level course outside my major. My GPA was one of the best in my class.
What does this mean? Absolutely nothing. I do well at school, at test taking, and at figuring out what the hell the prof wants. I does not however mean I took the path of least resistance. Though I am sure you would love to believe that of all those who got a better GPA than yourself.
Many people who were smarting than myself got worse grade point averages, because they did not care to always do what it took to get the A. There were also few idiot who were right up there with me, who just worked so damned hard the profs took pity on them even though they understood nothing.
Back on-topic. I think KPMG hires the latter.
-josh
The article says that they observed the microlensing event several years back and then recently took another look at the area using Hubble. They found a faint red dwarf which probably cause the lensing of the nearby (arc second wise) main sequence star.
So, if we can observe the 'dark' matter as being a red dwarf, it's not exactly 'dark' is it? I would assume that objects like red dwarfs, if observable, would have already been counted in the total 'bright matter' column. If not, someone is just undercounting objects that are observable using normal astronomical methods, and needs to go back and make a better estimate of how many of them are out there.
-josh
What kind of word is that? Webcurity...
What next? Homelandcurity?
-josh
He says:
"My theory is that this happens because it's harder to read code than to write it."
He couldn't be more right. I've recently been asked to port some code from another group in the company. Upon first reading it, I found global variables being referenced from everywhere, and it looked terrible.
I am usually the opposite, extremely leery of just starting over, no matter how bad the old code is.
The thing is once I have actually modify the code I realize it probably would have been better and faster to start from scratch.
-josh