I have some code I whipped up a few years ago when I was playing with this stuff. I'll have to dig around for it, but if you email me, I can send it your way.
The article mentions that Irix and Linux will converge in the future. Is it just me, or is this a really good thing? Irix has a lot it can bring to Linux. Hopefully this starts catching with all the other major Unix vendors and they start looking to it as a common ground for technology.
Imagine Linux running 128 CPUs in the same box with a massive amount of memory.
Hopefully an end-user will hack this so we can display custom images as boot time. I kind of like the penguin image I see in the virtual console boot.:) Imagine seeing that as you boot to Windows!
Now, if only I could get the virtual console to work with 2.2.9 (vs. 2.3.6)...
I hate when articles criticize Linux for the nonexistance of 3rd party apps. What do they call Wordperfect? What about KOffice? Civ:CTP? There aren't many, but that doesn't mean there are *none*.
As long as employers don't discriminate unfairly against those without experience, this could be good. It's the same as getting volunteer experience before you apply for a job somewhere else. Contributing to Open Source projects shows that you know how to program and you're not afraid to have peers review your software.
How many of you have worked in a business where people's code was so bad that you'd never release it publically.:) Imagine hiring people whose code you could already see, and see that it's neat on top of that.
IBM is into the Open Source game now, but their code still has a propriety feel to it. If you grab any of the things of the IBM Alphaworks site, you'll find that a lot of them have available source, but the source you get is pretty cryptic.
As well it seems like a lot of the IBM stuff is only developed in-house--community patches don't get reintegrated back into the main source tree.
It's a great start, but a lot of these companies have a fair ways to go before they understand the power of the true Open Source model. Mozilla is probably the best example of how to do it right, IMHO.
Isn't Kaffe the brand name of the Java interpreter that comes with most RedHat distributions currently? As far as I know, it's 100% java compliant (except for a few Sun APIs).
Hopefully Nikon makes a Video for Linux driver for their camera. It's nice to have all the devices under a single device. This is the way Windows does it and (I'm sorry I have to admit this;)), I think that was one of the best aspects of Windows.
As well, it lets me run my QuickCam as a TV input device rather than requiring specific software to run the capture program.
This study isn't really for us, but for our children. We can harvest some of the gains of a life-long study like this, but to realize the entire potential, it needs to run for its full course.
I guess in 100 years our children will know whether the Internet really was a benefit to our society and can build on our successes and mistakes, while keeping in mind the effect that it has on the human part of the equation.
There are so many projects now that are designed to be finished in _our_ lifetime, that we forget the benefit of stuff that spans generations. Look at all the monumental pieces of architecture that took hundreds of years to complete.
"If I have to guarantee that my software will perform the way you think it's going to perform, that's going to be costly for me," Winpro's Harris says. "Software isn't like a piece of furniture -- there are many other variables."
Well... it looks like they're tired of fulfilling all of their promises. I guess they want a law to support vapourware.
Software better damned well perform as I expect it, as my expectations are based on your advertisements and demonstrations. If I see a demo that shows a petroleum accounting product doing a year-by-year budget comparison at a trade show, that feature had better be in the demonstrated version.
Imagine if your TV had "supports over 1,000 channels on the box" and only supported 100 when you actually tried to use it!
Please, please kill this law. The only good thing that could come out of this is giving OSS a huge boost, but there are a lot of industries that will be hurt by this no matter what.
Well, I think this is not only a good idea, but one which is easily possible. If you have ever scoped AC signals before, you'll know that there is a lot of noise at a lot of high frequencies.
It should be trivial to harness some of the higher bands for transmitting data. The only problem is that power cables used to transmit the data would probably have to be extensively shielded, as they would become transmitters and receivers, either losing or broadcasting the data signals at points. Normally, as the AC single is just a 60Hz signal, power cables are often just shielded to protect against the elements, not against transmission of signals.
Well, there's basically a tradeoff between the price of the storage medium and the quality/durability of it. Memory sticks and other solid-state storage devices are nearly invincible (except to various form of ESD/etc), but they become incredibly expensive in the range of sizes needed to store MP3 files (64-128MB). Mechanical magneto and optical storage devices are vulterable to such things as magnetic fields, physical wear, dust, scratching and contact with the user, but obtaining a 100-150MB of storage on one is trivial and relatively cheap (compared to memory).
I'm willing to risk the integrity of my data for a cheaper storage solution, as I will probably have it backed up somewhere. The question of whether Iomega will exist in the future, however, is one we will have to wait and see about.
BTW, the IBM compact-flash-sized drives might also be a good solution, but I have no idea of the cost.
If you want really nice performance under most platforms, take a look at the HotSpot JIT compiler. They say it's about 3-5x faster running stuff.
As well, there's a release candidate for the rewritten java compiler which simply kicks butt. It's orders of magnitude faster.
Get HotSpot from java.sun.com (link is on the front page) and join the Developers section (free) to get javac-ea (Early-Access compiler). It's worth the downloads...
The Village Voice seems convinced that oppression can be blamed on the oppressed (read the last paragraph). It's like saying, "You weren't very happy in that German death camp, so why didn't you just leave?"
I don't believe students always have someone to talk to, or they might not be sure of who to talk to. As well, by shifting the blame to the victims, it almost suggests that this abusive behaviour is acceptable, which it isn't.
Look at the men and women in abusive relationships. How many times have they been asked, "Why don't you just get out of there?" I think people need help to realize that there are more options than violence, and the people involved in the oppression should be taught respect for their fellow students/people/etc.
Hmm... I'm not totally sure that the RIAA has lost in this case. Watermarking CD's is a VERY effective way to superimpose information on top of a song without it being audible, and without compromising the information if the song is encoded into an MP3 or whatever.
I tried playing around with various methods to defeat JPEG watermarking, but even stuff like embossing a picture doesn't get rid of it. The only way to beat it would be to go back to the FFT of the image and attempt to restore it.
The audio watermarking is similar to this and won't be easy to defeat. Someone will have to write a program to actually go and remove the FFT components that they've stored the information in.
If you have a chance, use Corel Photopaint to load a JPG and use the watermarking feature and take a look at the FFT of the image before and after. You can see the spectral components of the information fairly easily.
"AbiSource software is available for all to use, free of charge. Like most Open Source companies, we make our money selling a variety of services and resources. While our software is free, the additional services and resources which we sell are not."
It's nice to have more people making money from the Open Source model, rather than the standard models. They create a baseline of quality for the rest of the industry to follow and provide a high-quality alternative to those who don't want to shell out for the full office suite.
I may as well explain (very basically) how CMOS technology works, so you can get an idea.
There are two types of CMOS transistors: NMOS and PMOS. The only difference between them is in how they are "doped" (impurity ions injected into the silicon where they lie), as one is the opposite of the other (where NMOS is doped with an electron acceptor, PMOS is doped with a electron donor and vice versa).
CMOS transistors have three pins. One of them is the "gate" and the other two are the source and drain, where current will either flow or not.
An NMOS transistor acts like a switch that turns "on" (closes the circuit, making a path from one switch terminal to the other), when you apply the Vcc (the voltage representing a digital "1") to the "gate", which is the top of the transistor.
A PMOS transistor works in the opposite way, conducting a current when there is a digital "0" (usually 0V) applied to the gate.
If this sparked your interest, go to a bookstore and pick up a basic digital electronics book. It'll take you through some of the related physics (which are pretty cool) and you'll learn too.
BTW, if you've ever wondered what.25 micron and other numbers mean, that's basically one of the dimensions of the transistors in the chip. Smaller transistors mean less heat and faster circuits.
Well, I guess the paper doesn't really explain how this can be useful. The difference between a standard transistor and a transistor I assume that would be made from this is that the logic would be controlled by unrestricted current flow, as opposed to voltage (well, *technically* standard CMOS logic relies on current, but it's essentially voltage based).
The setup for creating the standard logic gates would be similar to the CMOS version and probably wouldn't take anyone longer than an hour or so to flesh out.
The advantages: - *zero* power loss in the transistors: this means almost zero power consumption in the chip - low-to-zero capacitance in the transistors: computers that operate at the speed of light (electrons moving as fast as physically possible)
I don't know if the switch between conductor/insulator is infitesmal or requires a fairly large time to occur, however. I guess we'll see in a few months.
Exciting technology.. any more info, anyone?
Re:Terabits: Possible but Not Practical
on
Terabit Routers
·
· Score: 1
People like Gates could afford it, but no matter how rich you are, there's a point where the incremential benefit (a cost measured in dollars, but kind of an estimate) of adding another CPU outweighs the incremential cost of adding that CPU.
Looks like the Dense Wave Division Multiplexing technology is going to take over in a few years. I bet you'll see this technology appearing in digital cellphone networks, followed shortly by inclusion into the Internet backbones.
A couple of these routers stationed across the world should handle an enormous amount of bandwidth.
If you take a look at the whitepaper, you'll see that they're partnered with Nortel-- one of the leading telecommunications companies.
This is a really important issue we're dealing with. People are so caught up with every machine being hooked up to the Internet, they haven't realized that you can actually use either an ipmasq or a natd program to allow all your computers behind your firewall to access the internet w/o problems.
I've seen offers for 8 IP addresses with ADSL connections. It'd be great to have that, but it is really important? With a firewall, you get protection against attacks and the like as well. I'm starting to wonder if I really want any computer (other than a firewall) hooked directly up to the Internet.
Subnets are EASY to configure and work almost as well as nets connected directly to the Internet. Remember that the next time you set up a number of connected systems.:)
Slashdot Mirror
I haven't been able to find any mention of this periodical anywhere. Anyone ever actually hear about it, let alone see it?
I have some code I whipped up a few years ago when I was playing with this stuff. I'll have to dig around for it, but if you email me, I can send it your way.
The article mentions that Irix and Linux will converge in the future. Is it just me, or is this a really good thing? Irix has a lot it can bring to Linux. Hopefully this starts catching with all the other major Unix vendors and they start looking to it as a common ground for technology.
Imagine Linux running 128 CPUs in the same box with a massive amount of memory.
/me drools
Hopefully an end-user will hack this so we can display custom images as boot time. I kind of like the penguin image I see in the virtual console boot. :) Imagine seeing that as you boot to Windows!
Now, if only I could get the virtual console to work with 2.2.9 (vs. 2.3.6)...
I hate when articles criticize Linux for the nonexistance of 3rd party apps. What do they call Wordperfect? What about KOffice? Civ:CTP? There aren't many, but that doesn't mean there are *none*.
As long as employers don't discriminate unfairly against those without experience, this could be good. It's the same as getting volunteer experience before you apply for a job somewhere else. Contributing to Open Source projects shows that you know how to program and you're not afraid to have peers review your software.
:) Imagine hiring people whose code you could already see, and see that it's neat on top of that.
How many of you have worked in a business where people's code was so bad that you'd never release it publically.
IBM is into the Open Source game now, but their code still has a propriety feel to it. If you grab any of the things of the IBM Alphaworks site, you'll find that a lot of them have available source, but the source you get is pretty cryptic.
As well it seems like a lot of the IBM stuff is only developed in-house--community patches don't get reintegrated back into the main source tree.
It's a great start, but a lot of these companies have a fair ways to go before they understand the power of the true Open Source model. Mozilla is probably the best example of how to do it right, IMHO.
Isn't Kaffe the brand name of the Java interpreter that comes with most RedHat distributions currently? As far as I know, it's 100% java compliant (except for a few Sun APIs).
Hopefully Nikon makes a Video for Linux driver for their camera. It's nice to have all the devices under a single device. This is the way Windows does it and (I'm sorry I have to admit this ;)), I think that was one of the best aspects of Windows.
As well, it lets me run my QuickCam as a TV input device rather than requiring specific software to run the capture program.
This study isn't really for us, but for our children. We can harvest some of the gains of a life-long study like this, but to realize the entire potential, it needs to run for its full course.
I guess in 100 years our children will know whether the Internet really was a benefit to our society and can build on our successes and mistakes, while keeping in mind the effect that it has on the human part of the equation.
There are so many projects now that are designed to be finished in _our_ lifetime, that we forget the benefit of stuff that spans generations. Look at all the monumental pieces of architecture that took hundreds of years to complete.
"If I have to guarantee that my software will perform the way you think it's going to perform, that's going to be costly for me," Winpro's Harris says. "Software isn't like a piece of furniture -- there are many other variables."
Well... it looks like they're tired of fulfilling all of their promises. I guess they want a law to support vapourware.
Software better damned well perform as I expect it, as my expectations are based on your advertisements and demonstrations. If I see a demo that shows a petroleum accounting product doing a year-by-year budget comparison at a trade show, that feature had better be in the demonstrated version.
Imagine if your TV had "supports over 1,000 channels on the box" and only supported 100 when you actually tried to use it!
Please, please kill this law. The only good thing that could come out of this is giving OSS a huge boost, but there are a lot of industries that will be hurt by this no matter what.
Well, I think this is not only a good idea, but one which is easily possible. If you have ever scoped AC signals before, you'll know that there is a lot of noise at a lot of high frequencies.
It should be trivial to harness some of the higher bands for transmitting data. The only problem is that power cables used to transmit the data would probably have to be extensively shielded, as they would become transmitters and receivers, either losing or broadcasting the data signals at points. Normally, as the AC single is just a 60Hz signal, power cables are often just shielded to protect against the elements, not against transmission of signals.
For some examples of some *beautiful* panoramic art, check out:
http://www.hotspots.hawaii.com/wrinklehome.html
Well, there's basically a tradeoff between the price of the storage medium and the quality/durability of it. Memory sticks and other solid-state storage devices are nearly invincible (except to various form of ESD/etc), but they become incredibly expensive in the range of sizes needed to store MP3 files (64-128MB). Mechanical magneto and optical storage devices are vulterable to such things as magnetic fields, physical wear, dust, scratching and contact with the user, but obtaining a 100-150MB of storage on one is trivial and relatively cheap (compared to memory).
I'm willing to risk the integrity of my data for a cheaper storage solution, as I will probably have it backed up somewhere. The question of whether Iomega will exist in the future, however, is one we will have to wait and see about.
BTW, the IBM compact-flash-sized drives might also be a good solution, but I have no idea of the cost.
For the lazy:
t ml
HotSpot Performance Engine:
http://java.sun.com/products/hotspot/
Developer Connection:
http://developer.java.sun.com/developer/index.h
(you'll have to sign up)
If you want really nice performance under most platforms, take a look at the HotSpot JIT compiler. They say it's about 3-5x faster running stuff.
As well, there's a release candidate for the rewritten java compiler which simply kicks butt. It's orders of magnitude faster.
Get HotSpot from java.sun.com (link is on the front page) and join the Developers section (free) to get javac-ea (Early-Access compiler). It's worth the downloads...
Well, if Mozilla can act as a GTK+ container, I don't see any reason why you couldn't do that.
The Gecko inside IE screenshots were really cool... I wouldn't mind browsing that way, if it means I get the best engine available.
The Village Voice seems convinced that oppression can be blamed on the oppressed (read the last paragraph). It's like saying, "You weren't very happy in that German death camp, so why didn't you just leave?"
I don't believe students always have someone to talk to, or they might not be sure of who to talk to. As well, by shifting the blame to the victims, it almost suggests that this abusive behaviour is acceptable, which it isn't.
Look at the men and women in abusive relationships. How many times have they been asked, "Why don't you just get out of there?" I think people need help to realize that there are more options than violence, and the people involved in the oppression should be taught respect for their fellow students/people/etc.
Why not attack this problem from both sides?
Hmm... I'm not totally sure that the RIAA has lost in this case. Watermarking CD's is a VERY effective way to superimpose information on top of a song without it being audible, and without compromising the information if the song is encoded into an MP3 or whatever.
I tried playing around with various methods to defeat JPEG watermarking, but even stuff like embossing a picture doesn't get rid of it. The only way to beat it would be to go back to the FFT of the image and attempt to restore it.
The audio watermarking is similar to this and won't be easy to defeat. Someone will have to write a program to actually go and remove the FFT components that they've stored the information in.
If you have a chance, use Corel Photopaint to load a JPG and use the watermarking feature and take a look at the FFT of the image before and after. You can see the spectral components of the information fairly easily.
"AbiSource software is available for all to use, free of charge. Like most Open Source companies, we make our money selling a variety of services and resources. While our software is free, the additional services and resources which we sell are not."
It's nice to have more people making money from the Open Source model, rather than the standard models. They create a baseline of quality for the rest of the industry to follow and provide a high-quality alternative to those who don't want to shell out for the full office suite.
I may as well explain (very basically) how CMOS technology works, so you can get an idea.
;))
.25 micron and other numbers mean, that's basically one of the dimensions of the transistors in the chip. Smaller transistors mean less heat and faster circuits.
There are two types of CMOS transistors: NMOS and PMOS. The only difference between them is in how they are "doped" (impurity ions injected into the silicon where they lie), as one is the opposite of the other (where NMOS is doped with an electron acceptor, PMOS is doped with a electron donor and vice versa).
NMOS = n-channel metal-oxide semiconductor
PMOS = p-channel " " "
pronounced "enn-moss" and "pee-moss" (duh
Here's an easy way to think of these:
CMOS transistors have three pins. One of them is the "gate" and the other two are the source and drain, where current will either flow or not.
An NMOS transistor acts like a switch that turns "on" (closes the circuit, making a path from one switch terminal to the other), when you apply the Vcc (the voltage representing a digital "1") to the "gate", which is the top of the transistor.
A PMOS transistor works in the opposite way, conducting a current when there is a digital "0" (usually 0V) applied to the gate.
If this sparked your interest, go to a bookstore and pick up a basic digital electronics book. It'll take you through some of the related physics (which are pretty cool) and you'll learn too.
BTW, if you've ever wondered what
Well, I guess the paper doesn't really explain how this can be useful. The difference between a standard transistor and a transistor I assume that would be made from this is that the logic would be controlled by unrestricted current flow, as opposed to voltage (well, *technically* standard CMOS logic relies on current, but it's essentially voltage based).
The setup for creating the standard logic gates would be similar to the CMOS version and probably wouldn't take anyone longer than an hour or so to flesh out.
The advantages:
- *zero* power loss in the transistors: this means almost zero power consumption in the chip
- low-to-zero capacitance in the transistors: computers that operate at the speed of light (electrons moving as fast as physically possible)
I don't know if the switch between conductor/insulator is infitesmal or requires a fairly large time to occur, however. I guess we'll see in a few months.
Exciting technology.. any more info, anyone?
People like Gates could afford it, but no matter how rich you are, there's a point where the incremential benefit (a cost measured in dollars, but kind of an estimate) of adding another CPU outweighs the incremential cost of adding that CPU.
Looks like the Dense Wave Division Multiplexing technology is going to take over in a few years. I bet you'll see this technology appearing in digital cellphone networks, followed shortly by inclusion into the Internet backbones.
A couple of these routers stationed across the world should handle an enormous amount of bandwidth.
If you take a look at the whitepaper, you'll see that they're partnered with Nortel-- one of the leading telecommunications companies.
What an amazing product.
This is a really important issue we're dealing with. People are so caught up with every machine being hooked up to the Internet, they haven't realized that you can actually use either an ipmasq or a natd program to allow all your computers behind your firewall to access the internet w/o problems.
:)
I've seen offers for 8 IP addresses with ADSL connections. It'd be great to have that, but it is really important? With a firewall, you get protection against attacks and the like as well. I'm starting to wonder if I really want any computer (other than a firewall) hooked directly up to the Internet.
Subnets are EASY to configure and work almost as well as nets connected directly to the Internet. Remember that the next time you set up a number of connected systems.