I can't believe how many people have posted to dis this paper!
Did any of you read it?
It's well done research, it carefully categorizes what can and can't be reconstructed and it specifies with some detail the methods used to recreate the transmitted data.
you can't write your own class BetterString which fixes this unicode problem and still works like a String.
No shit. If Sting was non-final it would be a massive security hole...
All you'd have to do is create a String subclass that changes its value between the time you, say, check a file for permissions and the time you actually open the file.
What's the big deal with implementing a parallel class?
public class MyString extends Object...
Since you're so obviously put off by using "broken" UTF-16, I'm sure you'll have no problem re-implementing every single String handling function in the class library.
People like Gosling and Stroustrup are virtually professional language designers. That's a whole different ball game from what the average software designer is doing for a living.
If I were to design a language from the ground up, I wouldn't do it.
The whole idea is to use something standard so that it's simple to bring new people up to speed and apply techniques that other developers have discovered.
It would be akin to building a bridge using a completely new, untested building material. Not my idea of a good engineering decision.
It sounds more like some sort of ethernet inetrface or some other high-level "web-tone" interface that would provide direct tcp/ip connectivity from the end user's point-of-view.
The big drag with using cell phones for internet connectivity now is that you need a separate ISP to dial up to... blah. This sounds much cooler.
The article seems to be written from the point of view that anything that the cable companies could charge you should be charged for.
The truth of that matter is that you're paying for bandwidth, not the number of PCs connected. Bandwidth is easily controlled by existing head-end routing hardware. The incremental costs in providing service are running the connection to a house and then providing the bandwidth. Extra PCs all sharing the same amount of bandwidth have zero additional cost. Does the author understand the difference between packet switching and circuit switching?
I think the comparison with early cable "theft" is spurious. What exactly is being stolen when using a gateway router? Nothing.
Besides, saying stuff like:
(The Internet, mostly a bulletin board at the time, topped out at 9600 baud back then.)
doesn't exactly give you a lot of credibility. Packet versus circuit switching is probably a bit beyond this person.
If was using DSL mostly to commute and I left my job and had less cash laying around, I'd
probably cancel the DSL too.
No kidding. If I had a car that I needed for work and I lost my job, I'd drop the car like a hot potato. Duh. Job-related items for a job you no longer have are kinda redundant.
In other news, most Candian broadband carriers have dropped lower intro rates for cable modems and DSL. I guess they're not that worried about attracting new customers anymore.
Bluetooth doesn't stand a chance. Why? Because it interferes with 802.11 802.11 throughput drops like a
stone when a Bluetooth piconet is active. Many corporations have banned Bluetooth devices (before they
were even available) to avoid this.
Yeah, that's why they banned microwaves in the lunch room too. Nothing kills throughput than someone with the munchies makin' up a bag of popcorn and running the world's biggest 2.4GHz jammer at the same time.
Too bad they can't use spread spectrum in the 2.4GHz band... oh, what's that? All 2.4GHz devices are mandated by law to be spread spectrum to avoid this kind of nonsense? Wow. Clever.
Did this guy do any research? Or did he just start rattling whatever popped into his head that he remembered from the good ol' days as a grad student? Why ignore flight simulators - they had simplistic graphics, sure, but they first-person perspective and did 3D graphics. What about 3D first-person space sims, like Elite? And, of course, Wolf3D, which he completely omits.
This guy is a complete moron when it comes to the history of 3D FPS games. What's his PhD in, geology?
Not bunding Java reduces Java support to the same level as these other no-one-would-ever-download-such-a-huge-plugin plugins:
Flash
Acrobat
Quicktime
etc. All it means is that you have to use the Java Plug-in tag syntax instead of the tag and make sure your applet is good enough to make people want to download and install the plugin. It's not a really big deal.
For example, visit one of the Java plug-in demos and see how easy it is to install the silly thing.
[Hm, having done just that, it's not quite as easy as Flash, but oh well. Maybe Sun will wise up.]
And, of course, there's no need to mention that fact that this has been available in Japan and Europe for quite a while now. Is Slashdot the new vanguard boldly proclaiming America's technological backwardness to the world?
look what you've got
on 2.4 GHz already...
Oh yeah I forgot - Bluetooth too. Though maybe tha's less dangerous, as no one is building Bluetooth equipment yet.
You know, with the huge use of the 2.4 GHz band, I gotta wonder if these networks will really achieve any decent throughput. Sure, I know, they use spread sprectrum to get over all that unpleasant "noise" stuff, but look what you've got on 2.4 GHz already...
802.11b. And maybe lots of it in some places. You can only have so many overlapping networks before performance degrades.
2.4Ghz cordless phones. By FCC regs, these can transmit as much power as an 802.11b card, so it only takes a few of them in your neighbourhood to start running over your wireless LAN.
Cordless anything. Keyboards, mice, gamepads...and funny, what are these usually right next to... your 802.11b card!
Microwaves. Which can pump out a shit-load of power and noise into the 2.4GHz band
Other industrial ISM-band equipment
While you can boost performance by using a high-gain directional antenna and putting it up high (like the top of your house) the legal limits on trnasmitted power are still really low - like 1 W RMS (according to the article - I thought it was 1mW RMS, but my memory is bad). I toyed briefly with the idea of setting up an 802.11b antenna on the top of my new house as I work fairly close to where I line and it would just be, you know, cool. But it's expensive and the only way to really get decent range is to use a number of cells (and I don't have the capital to build out a city-wide network) or crank the power, which is fine by me, but technically illegal.
Anyway, I don't really have enough spare time to hand-roll and antenna like these freenet guys, so I was thinking about buying one from HyperLink Technologies, but then I'm too cheap to do that.
Anyway, sounds like fun. Anyone building one in Toronto?
heard about this thing in my weekend paper (Toronto Star) as well - I guess their PR people are earning their pay this
month.
My bad. I read it in SiliconValley.com via AvantGo. Which is the same parent paper as the Contra Costa Times, thus my extreme deja vu while reading the article.
I heard about this thing in my weekend paper (Toronto Star) as well - I guess their PR people are earning their pay this month.
There are still a few reasons not to give up your good ol' POTS line just yet though:
Checked the network availability stats on your cable/DSL modem versus your local telco lately? Ever picked up your phone and received a "server timeout" error? Obviously these net-to-phone gizmos are primarily for saving money on long distance calls, but you'll still need a land line for ordering pizza or calling 911 after your double-cheese-and-bacon-pizza induced heart attack.
These things still require a subscription to the order of about $10 to $20 a month. That's on top of your $40 (minimum) DSL subscription. And the article I read said it needed a home gateway box ($100 or so). Plus the cost of the net-to-phone device itself ($100-$200). You're going to have to make a lot of long distance phone calls to offset all that capital & ongoing expense.
You pay extra to get an incoming phone number.
The amortized average cost per minute for a LD call with one of these things is still a few cents . I pay.10 CDN for long distance, most US residents can get LD for 7 or 5 cents a minute, of peak. Again, unless you make a lot of calls or mostly on-peak calls you're getting a fairly marginal savings.
Overall, I'm not convinced that it's really economical. Neat, maybe, but not all that economical.
No, what they did is train a computer to replicate the 'brush-stroke' effect without the computer having any a priori or internal knowledge of how to make a brush stroke.
It probably uses a neural net or something similar. The paper itself hasn't been published yet, so who knows.
Slashdot Mirror
I can't believe how many people have posted to dis this paper!
Did any of you read it?
It's well done research, it carefully categorizes what can and can't be reconstructed and it specifies with some detail the methods used to recreate the transmitted data.
It's god damn good research.
It took me a few minutes to realize that UWB has absolutely nothing to do with data communications at all. Sheesh
you can't write your own class BetterString which fixes this unicode problem and still works like a String.
No shit. If Sting was non-final it would be a massive security hole...
All you'd have to do is create a String subclass that changes its value between the time you, say, check a file for permissions and the time you actually open the file.
What's the big deal with implementing a parallel class?
public class MyString extends Object...Since you're so obviously put off by using "broken" UTF-16, I'm sure you'll have no problem re-implementing every single String handling function in the class library.
Hey, I didn't say no one should do it.
I said _I_ wouldn't do it.
People like Gosling and Stroustrup are virtually professional language designers. That's a whole different ball game from what the average software designer is doing for a living.
If I were to design a language from the ground up, I wouldn't do it.
The whole idea is to use something standard so that it's simple to bring new people up to speed and apply techniques that other developers have discovered.
It would be akin to building a bridge using a completely new, untested building material. Not my idea of a good engineering decision.
Are they giant robots?
Do they fly on super-rockets?
Where can I enroll?
So was I.
It sounds more like some sort of ethernet inetrface or some other high-level "web-tone" interface that would provide direct tcp/ip connectivity from the end user's point-of-view.
The big drag with using cell phones for internet connectivity now is that you need a separate ISP to dial up to... blah. This sounds much cooler.
I modified my flatbed scanner so that it has a window.
You mean, as opposed to the existing large window on your scanner called "The scanning surface"?
They describe how it's an example of how this could have been a big scam.
It's not Swing - it's IFC.
Anyway, it was a mediocre profiler at best.
This is why most Slashdot readers live so well without Windows... they don't run one of those 10 apps (Word, Powerpoint, Quicken, etc).
Did you check out the catalog page?
What I want is that McDonald's game! Awesome!
The article seems to be written from the point of view that anything that the cable companies could charge you should be charged for.
The truth of that matter is that you're paying for bandwidth, not the number of PCs connected. Bandwidth is easily controlled by existing head-end routing hardware. The incremental costs in providing service are running the connection to a house and then providing the bandwidth. Extra PCs all sharing the same amount of bandwidth have zero additional cost. Does the author understand the difference between packet switching and circuit switching?
I think the comparison with early cable "theft" is spurious. What exactly is being stolen when using a gateway router? Nothing.
Besides, saying stuff like:
doesn't exactly give you a lot of credibility. Packet versus circuit switching is probably a bit beyond this person.
If was using DSL mostly to commute and I left my job and had less cash laying around, I'd probably cancel the DSL too.
No kidding. If I had a car that I needed for work and I lost my job, I'd drop the car like a hot potato. Duh. Job-related items for a job you no longer have are kinda redundant.
In other news, most Candian broadband carriers have dropped lower intro rates for cable modems and DSL. I guess they're not that worried about attracting new customers anymore.
Bluetooth doesn't stand a chance. Why? Because it interferes with 802.11 802.11 throughput drops like a stone when a Bluetooth piconet is active. Many corporations have banned Bluetooth devices (before they were even available) to avoid this.
Yeah, that's why they banned microwaves in the lunch room too. Nothing kills throughput than someone with the munchies makin' up a bag of popcorn and running the world's biggest 2.4GHz jammer at the same time.
Too bad they can't use spread spectrum in the 2.4GHz band... oh, what's that? All 2.4GHz devices are mandated by law to be spread spectrum to avoid this kind of nonsense? Wow. Clever.
Did this guy do any research? Or did he just start rattling whatever popped into his head that he remembered from the good ol' days as a grad student? Why ignore flight simulators - they had simplistic graphics, sure, but they first-person perspective and did 3D graphics. What about 3D first-person space sims, like Elite? And, of course, Wolf3D, which he completely omits.
This guy is a complete moron when it comes to the history of 3D FPS games. What's his PhD in, geology?
- Flash
- Acrobat
- Quicktime
etc. All it means is that you have to use the Java Plug-in tag syntax instead of the tag and make sure your applet is good enough to make people want to download and install the plugin. It's not a really big deal.For example, visit one of the Java plug-in demos and see how easy it is to install the silly thing.
[Hm, having done just that, it's not quite as easy as Flash, but oh well. Maybe Sun will wise up.]
Microcell has had 2.5G GPRS running for at least a few weeks now. Expensive, yeah, but they're not going to go out of business at least.
http://www.fido.ca/NASApp/info/HomeFrame/Promotion 01.jsp?lang=en for the marketing junk.
MICROCELL FIRST TO DELIVER COMMERCIAL AVAILABILITY OF 2.5G WIRELESS DATA SERVICES ACROSS CANADA AND THE UNITED STATES
And, of course, there's no need to mention that fact that this has been available in Japan and Europe for quite a while now. Is Slashdot the new vanguard boldly proclaiming America's technological backwardness to the world?
look what you've got on 2.4 GHz already... Oh yeah I forgot - Bluetooth too. Though maybe tha's less dangerous, as no one is building Bluetooth equipment yet.
- 802.11b. And maybe lots of it in some places. You can only have so many overlapping networks before performance degrades.
- 2.4Ghz cordless phones. By FCC regs, these can transmit as much power as an 802.11b card, so it only takes a few of them in your neighbourhood to start running over your wireless LAN.
- Cordless anything. Keyboards, mice, gamepads...and funny, what are these usually right next to... your 802.11b card!
- Microwaves. Which can pump out a shit-load of power and noise into the 2.4GHz band
- Other industrial ISM-band equipment
While you can boost performance by using a high-gain directional antenna and putting it up high (like the top of your house) the legal limits on trnasmitted power are still really low - like 1 W RMS (according to the article - I thought it was 1mW RMS, but my memory is bad). I toyed briefly with the idea of setting up an 802.11b antenna on the top of my new house as I work fairly close to where I line and it would just be, you know, cool. But it's expensive and the only way to really get decent range is to use a number of cells (and I don't have the capital to build out a city-wide network) or crank the power, which is fine by me, but technically illegal.Anyway, I don't really have enough spare time to hand-roll and antenna like these freenet guys, so I was thinking about buying one from HyperLink Technologies, but then I'm too cheap to do that.
Anyway, sounds like fun. Anyone building one in Toronto?
They also sell pre-paid cell phone cards at post offices. Does Canada Post run any cell phone networks? NO!
Damn, you people are stupid. There's a post office in the Shopper's Drug Mart at King and Yonge... does Canada Post sell shampoo now too?
My bad. I read it in SiliconValley.com via AvantGo. Which is the same parent paper as the Contra Costa Times, thus my extreme deja vu while reading the article.
There are still a few reasons not to give up your good ol' POTS line just yet though:
- Checked the network availability stats on your cable/DSL modem versus your local telco lately? Ever picked up your phone and received a "server timeout" error? Obviously these net-to-phone gizmos are primarily for saving money on long distance calls, but you'll still need a land line for ordering pizza or calling 911 after your double-cheese-and-bacon-pizza induced heart attack.
- These things still require a subscription to the order of about $10 to $20 a month. That's on top of your $40 (minimum) DSL subscription. And the article I read said it needed a home gateway box ($100 or so). Plus the cost of the net-to-phone device itself ($100-$200). You're going to have to make a lot of long distance phone calls to offset all that capital & ongoing expense.
- You pay extra to get an incoming phone number.
- The amortized average cost per minute for a LD call with one of these things is still a few cents . I pay
.10 CDN for long distance, most US residents can get LD for 7 or 5 cents a minute, of peak. Again, unless you make a lot of calls or mostly on-peak calls you're getting a fairly marginal savings.
Overall, I'm not convinced that it's really economical. Neat, maybe, but not all that economical.It probably uses a neural net or something similar. The paper itself hasn't been published yet, so who knows.