> There is precious little data to suggest tethering users actually use more data.
There is data, though no carrier would publish those.
You wouldn't fire up HD videos on the small screen of the handset, even if the bandwidth would be available. With a large screen and a real keyboard you have more opportunities using up the available bandwidth. Also there is little motivation downloading ISOs (legitimate or otherwise) on your handset. Even with OTA phones usually don't get 100 MB of OS-patches a week (there are even features to throttle the bandwidth to the Windows and Apple update servers during peak-hours, I kid you not).
Also "flat"-rates which are only intended to be used with handsets are made cheaper so that selling a new smartphone along with a new contract is made easier. As I once saw in a presentation from a large Performance Enhancement Proxy vendor: "Bandwidth is a gas". Give the users the possibility to use it, and they will certainly use it until all capacity is maxed out. Especially in times where carriers are investing buckets full of money to set up 4G networks they are afraid of one thing: That subscribers will use the extra capacity to go to online movie rental shops who're going to make the profit - and extra capacity doesn't necessarily lead to new subscribers... That latter will most probably happen, I saw it happen a few times after large mobile network expansions. (Adding more GGSN to the network, therefore distributing the subscriber's traffic over more internet handovers points than before)
Future tariffs will be * different bandwidths (256k, 2M, 5M, something along those lines) * different caps
The whole anti-tethering thing is not greed, but Angst. Sure, mobile phone carriers have a problem with their business models. But a mobile phone carrier ain't just another ISP: They have to run a mobile network only to be ALSO an ISP. And that's expensive. Really, really expensive.
I wonder how a small firm like TomorrowNow with 400 service-contracts - and a net profit of ~50 mio. USD in the time frame in question - can make a such a damage. A, maybe it's because Oracle bought the companies who made the software TomorrowNow was offering services for...
Who cares about advanced gaming capabilities if the current line of smartphones if they can't even provide a sufficient battery-life so that one can actually *use* those gaming features for longer than a few hours?
Generally if a software is released under the terms of the GPL, the following freedoms apply. The FSF says:
"Your software shall grant you these freedoms: * to run for any purpose * to study and adapt to your needs * to redistribute, so you can help others * to release improvements, so everyone benefits."
That's the "free" in free software (contrary to open source).
Also the GPL doesn't allow discrimination; a "3rd party contract" can't change the terms of the GPL. A company can't restrict the terms of the GPL just because it doesn't fit to their business-model. If they do so, they fail to comply with the GPL. If they want to restrict rights, they shouldn't have chosen the GPL in the first place.
I'd say: Forget about this restriction, it's invalid. You should be contacting the FSF or, specifically, the Freedom Task Force [1] about this case. They're from the european branch of the FSF, but I'm sure they can help you out anyway.
I don't agree, Tor ain't dead. It's just a bad idea to run a Tor exit-node in Germany. You can still run a middleman-node. And other countries aren't (yet) that aggressive. Just have a look at the Tor node-statistic[1] and see: It's not dead. Tor is still one of the best tools for privacy on the internet we have.
First: Trifish is right. After all, I was treated fairly and only little violence was used (means, I was cuffed and my civil rights were violated). And as trifish correctly pointed out, I was released shortly after the interrogation. They even drove me home. Nonetheless, it was an experience I didn't wanted to make.
Secondly: I don't get the focus of all the discussions up here on/.
It's not about Nazis, Fascism or whatever, but merely about the incompetence of the police and the lawyer of the state. All those discussion are bizarrely drifting away in a mix of Kafka and Godwin.
The whole point is: 1) If you run an exit-node in Germany, think about the consequences. 2) Don't count on the competence of the authorities. 3) Prepare your family if you run a node. 4) Get a lawyer in case something goes horribly wrong. 5) You have the right to remain silent. Use it. (Unlike me, who just came back from a pub-crawl and was in the mood for extensive talking. Gnaaa.)
There's no great master plan of the evil German government to destroy Tor.
I haven't looked into that "Killer-NIC" (not interested anyway) but i know a bit about latency; people building cluster are investing a great deal of time reducing the latency the operating system's TCP/IP stack introduces. This can make up a difference of ~100-microseconds. So chosing the right network technology to deliver your data from one host to another one takes a great deal of time and lot's of effort is put into the research of new networking-technologies.
See Infiniband, or the various TCP-offloading features, take Level 5's Etherfabric or projects like MPI/GAMMA who circumvent the whole IP-shebang directly and let the programm build up it's own Ethernet-frames.
So i wouldn't necessarily denouce the whole "Killer-NOC" thing as bullshit; but on the other i must admit: The network in between (speak: Internet) is the problem for gaming, NOT necessarily the IP-stack or NIC.
I'm not a gamer so i can't tell if a difference in latency of, let's say, 100 us really brings an advantage (he, folks, that's 0.01 milliseconds!) - but well, let's see what kinda specs the company is going to release. Alex.
The LSC (LIGO Scientific Collaboration) thinks a little bit different about that. In their document "First report on the S3 analysis" (http://einstein.phys.uwm.edu/PartialS3Results) which is based on the Einstein@Home community efforts they say:
"However, the numbers of sources and their distances from us are uncertain, and in their first few years of operation it is quite possible that the LIGO and GEO instruments may not detect anything." "So far, we have not seen any evidence for pulsar signals in the S3 data. As described earlier, this is not surprising, because LIGO is not sensitive enough to guarantee that we will see one or more pulsars."
LIGO is going to be upgraded ("Advanced LIGO"), which will improve the detection of events by the factor 100-1000. Maybe the theory of grav-waves is even wrong, who knows...:-)
Nothing new. Same with glasblowers, who get an eye cataract from their job. As tgibbs said: Take any confined and insulated tissue, pump energy in it, eventually the cells will be damaged. Sooner or later.
IEEE should declare itself non-profit. Volunteers, corporate and non-corporate members can join. Corporate standard-warriors can submit standards, participate in gremi and discuss. Volunteering private persons could submit standards as well.
The leading-board could be funded by voluntary donations. They could adopt an early-subscription model; the whom who pays, gets the standard a couple of weeks before it's publicly available. That's fair.
See how the IETF does it. That could be a model for the IEEE.
Standards should be FREE as in free beer, free of patents and available to anyone anyway. With no charge.
Maybe the IEEE will dissolve itself in the IETF. Maybe ITU and ETSI join them...
For those of you using ECN be reminded that the website is a ECN blackhole[1] and will not load if you don't disable ECN or strip off the ECN bits from the header.
For you Linux-users, it can easily be done with "iptables -t mangle -A POSTROUTING -p tcp -d 208.50.55.160 -j ECN --ecn-tcp-remove".
Why shouldn't it? First, it's not iptables running the firewall, but the Netfilter-Framework which resides in the kernel. Iptables is just the userspace program to administrate the firewall functions the Netfilter-Framework offers.
Netfilter has several hooks where you can injected or gather data at several points of the firewalling process. It should be possible to migrate certain parts of the framework to dedicated hardware.
However, it has to be shown that you get an actual performance advantage - pushing data on the chip, wait for interupt, get data from chip, continue with processing - all that stuff takes some time and normal CPUs are fast enough nowadays to filter lot's of data per second (i personally filter 5*100MBit + 1*2Mbit/s on a dual Xeon-3 750).
Anyway, as long as there are no more specific-specs available one can speculate as much as you want.
Slashdot Mirror
> There is precious little data to suggest tethering users actually use more data.
There is data, though no carrier would publish those.
You wouldn't fire up HD videos on the small screen of the handset, even if the bandwidth would be available. With a large screen and a real keyboard you have more opportunities using up the available bandwidth. Also there is little motivation downloading ISOs (legitimate or otherwise) on your handset. Even with OTA phones usually don't get 100 MB of OS-patches a week (there are even features to throttle the bandwidth to the Windows and Apple update servers during peak-hours, I kid you not).
Also "flat"-rates which are only intended to be used with handsets are made cheaper so that selling a new smartphone along with a new contract is made easier. As I once saw in a presentation from a large Performance Enhancement Proxy vendor: "Bandwidth is a gas". Give the users the possibility to use it, and they will certainly use it until all capacity is maxed out. Especially in times where carriers are investing buckets full of money to set up 4G networks they are afraid of one thing: That subscribers will use the extra capacity to go to online movie rental shops who're going to make the profit - and extra capacity doesn't necessarily lead to new subscribers... That latter will most probably happen, I saw it happen a few times after large mobile network expansions. (Adding more GGSN to the network, therefore distributing the subscriber's traffic over more internet handovers points than before)
Future tariffs will be
* different bandwidths (256k, 2M, 5M, something along those lines)
* different caps
The whole anti-tethering thing is not greed, but Angst. Sure, mobile phone carriers have a problem with their business models. But a mobile phone carrier ain't just another ISP: They have to run a mobile network only to be ALSO an ISP. And that's expensive. Really, really expensive.
Alex.
So anything except the official clients and some selected 3rd party clients will be banned?
Meh.
Hope they won't use the Galaxy S' proprietary 3.5 mm headphone jack anymore. That one is ridiculous.
I wonder how a small firm like TomorrowNow with 400 service-contracts - and a net profit of ~50 mio. USD in the time frame in question - can make a such a damage.
A, maybe it's because Oracle bought the companies who made the software TomorrowNow was offering services for...
head.bang->desk();
Who cares about advanced gaming capabilities if the current line of smartphones if they can't even provide a sufficient battery-life so that one can actually *use* those gaming features for longer than a few hours?
Generally if a software is released under the terms of the GPL, the following freedoms apply. The FSF says:
"Your software shall grant you these freedoms:
* to run for any purpose
* to study and adapt to your needs
* to redistribute, so you can help others
* to release improvements, so everyone benefits."
That's the "free" in free software (contrary to open source).
Also the GPL doesn't allow discrimination; a "3rd party contract" can't change the terms of the GPL.
A company can't restrict the terms of the GPL just because it doesn't fit to their business-model. If they do so, they fail to comply with the GPL. If they want to restrict rights, they shouldn't have chosen the GPL in the first place.
I'd say: Forget about this restriction, it's invalid. You should be contacting the FSF or, specifically, the Freedom Task Force [1] about this case. They're from the european branch of the FSF, but I'm sure they can help you out anyway.
Alex.
[1] http://www.fsfeurope.org/projects/ftf/ftf.en.html
I don't agree, Tor ain't dead. It's just a bad idea to run a Tor exit-node in Germany. You can still run a middleman-node. And other countries aren't (yet) that aggressive. Just have a look at the Tor node-statistic[1] and see: It's not dead. Tor is still one of the best tools for privacy on the internet we have.
Alex.
[1] https://torstat.xenobite.eu/showstatistics.php
First: Trifish is right. After all, I was treated fairly and only little violence was used (means, I was cuffed and my civil rights were violated). And as trifish correctly pointed out, I was released shortly after the interrogation. They even drove me home. Nonetheless, it was an experience I didn't wanted to make.
/.
Secondly: I don't get the focus of all the discussions up here on
It's not about Nazis, Fascism or whatever, but merely about the incompetence of the police and the lawyer of the state. All those discussion are bizarrely drifting away in a mix of Kafka and Godwin.
The whole point is:
1) If you run an exit-node in Germany, think about the consequences.
2) Don't count on the competence of the authorities.
3) Prepare your family if you run a node.
4) Get a lawyer in case something goes horribly wrong.
5) You have the right to remain silent. Use it. (Unlike me, who just came back from a pub-crawl and was in the mood for extensive talking. Gnaaa.)
There's no great master plan of the evil German government to destroy Tor.
Alex.
Or stealing the wires from the railroad system as they do in Germany:/ kupferdiebstahl/index.jhtml
http://www.wdr.de/themen/panorama/kriminalitaet09
The copper-thieves are partly responsible for train-delays because it happens so often.
The german railroad-company lost 20 mio. Euros because of stolen wires.
Now that's a lot. Copper ain't cheap.
Even if it would be text-only, you still could use uuencode/uudecode to transmit binary data.
http://en.wikipedia.org/wiki/Uuencode
The whole point about TOR is that it's working transparently and is rather easy to use.
And who am I that i start censoring myself? A deputy?
Alex.
Why should jc react at all? The only one who'll draw attention will be John Gruber himself.
I'm saying over and over again: Never post before you had your first coffee:
;)
> he, folks, that's 0.01 milliseconds!
It's 0.1 ms, yeah, i suck, i know.
Alex.
I haven't looked into that "Killer-NIC" (not interested anyway) but i know a bit about latency; people building cluster are investing a great deal of time reducing the latency the operating system's TCP/IP stack introduces. This can make up a difference of ~100-microseconds. So chosing the right network technology to deliver your data from one host to another one takes a great deal of time and lot's of effort is put into the research of new networking-technologies.
See Infiniband, or the various TCP-offloading features, take Level 5's Etherfabric or projects like MPI/GAMMA who circumvent the whole IP-shebang directly and let the programm build up it's own Ethernet-frames.
So i wouldn't necessarily denouce the whole "Killer-NOC" thing as bullshit; but on the other i must admit: The network in between (speak: Internet) is the problem for gaming, NOT necessarily the IP-stack or NIC.
I'm not a gamer so i can't tell if a difference in latency of, let's say, 100 us really brings an advantage (he, folks, that's 0.01 milliseconds!) - but well, let's see what kinda specs the company is going to release.
Alex.
P.S.: The Cluster Monkey had quite an extensive comparison of different low-latency high-bandwdith network-technologies: http://www.clustermonkey.net/content/view/124/34/
I compiled some quick facts which compare those three supercomputers and added pointers to other resources for your convenience:
http://www.bloglines.com/blog/ITnomad?id=126
Cheers, Alex.
The LSC (LIGO Scientific Collaboration) thinks a little bit different about that.
:-)
In their document "First report on the S3 analysis" (http://einstein.phys.uwm.edu/PartialS3Results) which is based on the Einstein@Home community efforts they say:
"However, the numbers of sources and their distances from us are uncertain, and in their first few years of operation it is quite possible that the LIGO and GEO instruments may not detect anything."
"So far, we have not seen any evidence for pulsar signals in the S3 data. As described earlier, this is not surprising, because LIGO is not sensitive enough to guarantee that we will see one or more pulsars."
LIGO is going to be upgraded ("Advanced LIGO"), which will improve the detection of events by the factor 100-1000.
Maybe the theory of grav-waves is even wrong, who knows...
...to see the date when the image was taken.
Alex.
Nothing new. Same with glasblowers, who get an eye cataract from their job.
As tgibbs said: Take any confined and insulated tissue, pump energy in it, eventually the cells will be damaged. Sooner or later.
Yeah, and while the MPAA is at it, tell the NRA that guns kill people...
scnr, Yalla.
IEEE should declare itself non-profit. Volunteers, corporate and non-corporate members can join.
Corporate standard-warriors can submit standards, participate in gremi and discuss.
Volunteering private persons could submit standards as well.
The leading-board could be funded by voluntary donations. They could adopt an early-subscription model; the whom who pays, gets the standard a couple of weeks before it's publicly available. That's fair.
See how the IETF does it. That could be a model for the IEEE.
Standards should be FREE as in free beer, free of patents and available to anyone anyway. With no charge.
Maybe the IEEE will dissolve itself in the IETF. Maybe ITU and ETSI join them...
Just my 2c.
Alex.
Any advice for the poor schmuck who's going to get the blame?
Cuba is nice at this time of the year...
For those of you using ECN be reminded that the website is a ECN blackhole[1] and will not load if you don't disable ECN or strip off the ECN bits from the header.
For you Linux-users, it can easily be done with "iptables -t mangle -A POSTROUTING -p tcp -d 208.50.55.160 -j ECN --ecn-tcp-remove".
Alex.
[1] http://urchin.earth.li/ecn/
Have you ever stood in front of a SUN Fire 6800 cabinet? They have a big sucker of a SUN logo illuminated by those blue LEDs...
:-)
So shiny... Happy happy happy... Must drool and watch... can't resist...
I'm still waiting for the Octalus-like big mouth with needle-sharp fangs coming out of the cabinet snatching for my head
Yalla.
Why shouldn't it? First, it's not iptables running the firewall, but the Netfilter-Framework which resides in the kernel. Iptables is just the userspace program to administrate the firewall functions the Netfilter-Framework offers.
Netfilter has several hooks where you can injected or gather data at several points of the firewalling process. It should be possible to migrate certain parts of the framework to dedicated hardware.
However, it has to be shown that you get an actual performance advantage - pushing data on the chip, wait for interupt, get data from chip, continue with processing - all that stuff takes some time and normal CPUs are fast enough nowadays to filter lot's of data per second (i personally filter 5*100MBit + 1*2Mbit/s on a dual Xeon-3 750).
Anyway, as long as there are no more specific-specs available one can speculate as much as you want.
Cheers, Alex.
I like to be up-to-date...
Are you sure those where binary transmissions? :-)