BitTorrent Ponders Releasing World ISP P2P Speed Report 156
Mark.JUK writes "The San Francisco-based inventor of the hugely popular peer-to-peer (P2P) internet file sharing protocol BitTorrent has revealed that it is considering whether or not to release the broadband performance (speed) data for more than 9,000 ISPs around the world. The technology company claims that the data forms part of its new project, which is sadly still in the very early stages of development, but could one day give consumers a near real-time perspective of how their ISP is performing. It wouldn't just cover P2P traffic either, with BitTorrent also tracking general HTTP transfers too. BitTorrent claims that its service can, for example, display that most UK ISPs 'aggressively throttle BitTorrent traffic after 6 p.m. at night,' with speeds suddenly going 'off a cliff.' Suffice to say that such information could prove to be very useful for consumers and advocates of Net Neutrality."
Considering? (Score:5, Insightful)
The debate is important. (Score:1)
I am speculating, but this information could tell certain copyright cartels where to target their legislative action. Considering that risk, should these data be made public?
Re: (Score:2)
it is when you're shaking someone down
Re: (Score:3)
Re: (Score:2)
It can be, depending on who is making it, and the decision that's being made.
Re: (Score:3)
Consideration of performing an action is news now?
It shouldn't be, but in this case it seems that's the case. They're just trying to gain some attention so everyone is listening when they do release the report. I don't really see why they wouldn't release this information.
Re: (Score:2)
Oh, I'm sure they can get sued one way or another to prevent its release.
In the modern American justice system, this is how I see it going down:
1) Party with an interest in not seeing the data get out (ISP, record label, whatever) sues. Yeah, the case probably won't win, but that's not the point; the point is to get a legal injunction stopping its release while the case goes to trial. Their lawyers will do everything they can to drag it out - release thousands of documents, line up dozens of "experts" for t
Re: (Score:3)
Not really, but the existence of the data itself is.
Staging the release for benefit of the press? (Score:2)
Really -- why doesn't he just release it? Guess it needs 'staging' so it gets the proper attention paid to it in the press....?
why? (Score:2)
i doubt there is someone in the U.S. that believes there is an ISP that doesn't treat their customers like cornholio. i guess this intended for other nations.
Re: (Score:2)
You can be the hole, or you can be the corn.
Re:why? (Score:4, Informative)
I don't understand why everyone always says "the US sucks" and "other countries are better" (or words to that effect). Is this a case of thinking the Grass is Greener on the other side of the fence?
Because it isn't true. Here is how the US compares to other continent-spanning nations/federations. Maybe I'm biased but I don't think second place is a bad place to be:
Mbit/s
12.3 Russian Federation
10.3 US
10.0 EU
9.3 Canada
8.0 Australia
5.7 Saudi Arabia
4.8 Brazil
3.8 China
3.4 Mexico
Mbit/s (EU versus US member states):
29 Lithuania
26 Latvia
24 Romania
23 Netherlands, Sweden
18 Portugal
17 Germany
16 Bulgaria, Denmark
15 DE, Belgium
14 Luxembourg, MA, RI, VA, WA, Hungary, MD, France
13 NY, Finland, NJ
12 NH, MN, Estonia
11 Austria
10 Slovakia, Czech, UK, Spain
8 Slovenia, Malta
7 Poland
6 Ireland, Georgia, Greece, Turkey
5 Cyprus
4 Italy
3 Greenland
Re: (Score:1)
cool story bro.
now throw in this little thing called "price".
Re: (Score:2)
The GP is probably basing his judgement on factors not directly related to speed, like usage caps and price per month. But then, that list looks like it's been made based on the advertised rate which, as TFA shows, can be not entirely accurate for normal usage scenarios.
Re: (Score:2)
>>>that list looks like it's been made based on the advertised rate
No it came from speedtest.net, which collects aggregate data from actual file transfers across the wires. It is the most accurate database available.
Re: (Score:2)
Re: (Score:2)
14 mbits in Washington State? Bullshit.
Maybe in the tiny selected areas that get FIOS, maybe in the center of Seattle, but the state on average? Not even half that. No way in hell.
I question your data.
Re: (Score:2)
Re: (Score:2)
The list is not meant to be an exhaustive list of all 75 US and EU states (I don't feel like typing that many names). What matters is this. We're number 2 in comparison to other continent-spanning countries/federations.
12.3 Russian Federation
10.3 US
10.0 EU
9.3 Canada
8.0 Australia
5.7 Saudi Arabia
4.8 Brazil
3.8 China
3.4 Mexico
Re: (Score:2)
We're number 2 in comparison to other continent-spanning countries/federations.
That's little consolation for those of us who, thanks to government-granted monopolies, get 4Mb/s. Also, who cares about "continent-spanning countries"? All of South Korea gets faster internet than any city in the U.S.
Re: (Score:2)
>>>All of South Korea gets faster internet than any US city
"People may think you are stupid, but there's no point opening your mouth and proving it." - Abraham Lincoln. In other words I'm saying your statement is flat fucking false:
S. Korea == 38.5 Mbit/s
Pembroke MA == 38.5 Mbit/s
Bridgeport WA == 39.55 Mbit/s
Vincentown NJ == 47.2 Mbit/s
Orange NJ == 45.5 Mbit/s
Swedesboro NJ == 43.0 Mbit/s
Holmdel NJ == 39.6 Mbit/s
Kingston RI == 53.6 Mbit/s
Olean NY == 36.5 Mbit/s
Lakeview OR == 40.8 Mbit/s
La Jolla CA
Re: (Score:2)
There will always be people living too far away for decent service. Trying to bring top end service to those people would quickly become a "bridge to nowhere" type issue. I live in Kentucky and have 50mb service. But 10 miles (16km) away from town, the best available is 10mb. Population density allows for more expensive infrastructure.. there is nothing wrong with that. But the problems come in with competition, which you pointed out. If your local ISPs aren't investing in infrastructure and both offe
Re: (Score:2)
Re: (Score:2)
i doubt there is someone in the U.S. that believes there is an ISP that doesn't treat their customers like cornholio. i guess this intended for other nations.
So for I have had zero problems with Verizon FiOS. They treat me pretty well right now, even 6 months into the contract nothing has changed.
Re: (Score:2)
You mean like this one? (Score:3)
Purportedly from Netflix:
http://2.bp.blogspot.com/_gC6nMAI6mu8/TUHG6jsQq-I/AAAAAAAAADE/Bwe1fkAUxzA/s1600/isp_usa.png [blogspot.com]
Re: (Score:2)
Innnnteresting.
The lower band is cellphones and DSL, and the upper band is cable providers.
Verizon is sort of by itself. I'm guessing that Netflix is averaging cellphones and home-fiber customers, because I would surely expect the FiOS folks to be getting the best speeds of all.
Most intersting fact is that Clearwire is totally shitty. AT&T is kicking their ass, but then, they do brag about being the fastest, they just don't mention how little coverage they have for any sort of broadband-speed service.
Re: (Score:2)
I would surely expect the FiOS folks to be getting the best speeds of all.
You would be correct. At least in my case. I don't know what the speed is when I'm watching netflix, but I've never had to buffer a video.
Re: (Score:2)
Re: (Score:1)
What amazes me about that is that Comcast, Charter and Cox are so obviously in the lead. Did Netflix pay them anything for this?
Re: (Score:2)
Looks like the average is somewhere around 2000 kbit/s.
Good to know my 1000k line is slower but not significantly slower than typical.
Re: (Score:2)
Does the Verizon and Frontier line include both DSL and FIOS? Because that makes the data for those two providers pretty damned useless.
Forget advocates how about consumers in general (Score:5, Insightful)
Perhaps people would like to know tha the 10Mb/sec speed advertised by their provider is only available from 4am to 6:30am on weekdays.
These actual usable bandwidth numbers should be general public knowledge. It would enable consumers to make valid choices and perhaps make providers do some real provisioning to support their advertised bandwidths.
Re: (Score:2)
Re: (Score:3)
Or it could also be interesting that your 10 Mb/s speed is available never. How would that go over?
Reminds me of when I call places like my ISP and other companies. I always get the message "We're experiencing an unusually high volume of calls at the moment, please bear with us."
Really, unusually high? Why does it *always* seem to be "unusually" high? I would love to set up some automated system to call these companies and see how often I get that message. I bet that even if you called every hour for weeks, you would always get that message. Seems to me that you can't call something unusual if its always
Re: (Score:2)
I bet that even if you called every hour for weeks, you would always get that message. Seems to me that you can't call something unusual if its always the case.
Actually, they're usually only open from 9 to 5. Outside of these hours, you don't get the "unusually busy" message; you get the "please call back between 9 and 5" message. Since 9 to 5 is only 1/3 of a day, the volume they experience during this period is "unusually busy" compared to when they don't accept calls.
I'm still amazed at how many places do this; it should be extremely easy in this day and age to at least record and sort messages during off-hours instead of saying "please call back during worki
Re: (Score:2)
I think everyone is cognizant of the "up to" in the brochure.
Re: (Score:2)
*** mean bandwidth 512/340 Kbps, 99th percentile 1Mbps/512Kbps.
Re: (Score:1, Flamebait)
Or it could also be interesting that your 10 Mb/s speed is available never. How would that go over?
I'm sorry if the marketing jargon used by ISPs frustrates you. Unfortunately, you aren't paying for guaranteed service. This was in the fine print of the paper you signed. Did you read it?
Don't feel too bad, though - you can call always call your ISP and ask for a 99.999 uptime SLA and a CIR of 10 Mbps, but they'll probably just laugh at you, considering most Department of Defense bases don't even have that level of service, and I can guarantee you can't afford it. It doesn't matter which ISP you're with -
Re: (Score:2)
Mod ^^^^this dude^^^^ up please.
Seriously, all the kiddies here who complain about throttling, not getting 100% of advertised speeds, etc. have never dealt with real Internet transit costs. If you want a specific CIR [wikipedia.org] and uptime SLA, you have to pay, and pay dearly. Note that FiOS, "business class" service from most residential ISPs, the cheap 100 Mbps circuits from Cogent, and the "2 TB/month" from your cheap hosting provider might have at least some form of SLA for uptime if you're lucky. But the CIR for s
Re: (Score:2)
So why aren't we requiring the providers to at least make available what their uptime and average data rate per customer is? They know this number, they know how much data they can move in any area. Yet they protect this information like it is a trade secret, that the average user could not possibly understand that they are offering 20MBps to the entire county while only having 200MBps backbone, and have only had an uptime over the past year of 98%.
To the GP's auto analogy, if I went to a car dealership and
Re: (Score:2)
So why aren't we requiring the providers to at least make available what their uptime and average data rate per customer is? They know this number, they know how much data they can move in any area. Yet they protect this information like it is a trade secret, that the average user could not possibly understand that they are offering 20MBps to the entire county while only having 200MBps backbone, and have only had an uptime over the past year of 98%.
Such information could legitimately be considered trade secrets. Oversubscription ratio is a major component of network design, and increasing that ratio has a significant benefit to the bottom line. If a particular topology, peering design, or technology allows a network to increase oversubscription while providing similar QoS to customers, that is a huge competitive advantage.
As for requiring disclosure of specific metrics, that is something that regulation could mandate. But the reality is a lot more com
Re: (Score:2)
Re: (Score:2)
And then what?
My other choice is DSL through Frontier. The first thing Frontier did after taking over Verizon's landlines was stop the FIOS rollout, so their speed is 3 mbits and will be for the foreseeable future. (The exact same speed it has been since 1998...) Clear's closing down shop, at least direct to residential customers, and they couldn't even hit the 3 mbits where I am anyway.
Knowing how bad my ISP service is would just be rubbing in how awful the lack of competition is.
Dumb comment (Score:4, Insightful)
Suffice to say that such information could prove to be very useful for consumers and advocates of Net Neutrality.
What a stupid thing to say. It doesn't offer any insight as to why bandwidth may have gone off a cliff. Net Neutrality is not the same thing as responsible QoS! Get that through you heads!
After 6pm, Internet traffic for most ISPs goes through the roof. With it, latency and available bandwidth are typically negatively affected. With a responsible QoS, which is still fully Net Neutral, its easily possible to explain services such at BT "falling off a cliff." After all, if you give it a low priority, which reasonably it should, other users may simply be driving it "off the cliff."
Me, like most every reasonable person in the world, certainly does not want to have You Tube, general web browsing, email, IRC, streaming music, game playing, or any of a number of other services negatively affected because Joe down the street is downloading his fifth illegal movie for the day, especially when he's likely to watch it later, or getting his next WoW update. Some things require an interactive level of performance - some others do not. BT, by definition, is a service which should receive a low priority in any QoS infrastructure.
Net Neutrality is about ensuring company X doesn't get premier service at the expense of its competition. Its not about ensuring reasonable QoS to ISP customers. Please stop conflating the two.
Now having said all that, there may be other things are work here, but there is nothing in the article which suggests there is anything controversial going on. As is, things are reasonably explainable with traditional usage trends and a reasonable desire to maintain a reasonable QoS to customers.
Re: (Score:1)
Re: (Score:2)
While all true, I think you give the ISP's too much credit. If you really believe that tripe, I know this guy in Nigeria, he's a prince and he'd like to meet you.
So far, the only thing you've validated is you are unethical and have no idea how networking works, which makes us concluded you're unqualified to refute my statement.
Giving too much credit? Actually, statements like that make you look extremely stupid. As a matter of fact, every major ISP is going to have a QoS infrastructure in place. They may not all agree in exact service prioritization but its there. Furthermore, its an extremely reasonable assumption BT traffic has been given one of the lowest, if not
Re: (Score:2)
Not sure why you're having trouble grasping that.
Re: (Score:3)
as Net Neutrality has to do with discriminating with types of data or origin of data,
ORIGIN, NOT PROTOCOL.
Sorry for yelling, but this is getting old. GP just got done giving a very lucid explanation of why throttling BT more than HTTP is NOT a Net Neutrality issue, and you promptly responded with (paraphrased)
Yes but BT is being throttled more than HTTP, so its a neutrality issue
Get this through your head-- throttling by protocol is called QoS. It is legitimate. It is not a net neutrality issue. Responsible ISPs do this.
Throttling by source or destination address or domain is not QoS, is illegitimate, and IS a net neutrality issue.
Finally, if you cant be
Re: (Score:3)
Net neutrality isn't particularly about origin, other than "origin outside of the ISP's own moneymaking sphere."
Net neutrality is entirely about fairness of access in the face of pressure to monetize that access.... increasing the effective cost of access to content that the network provider isn't already making money on to drive traffic to its own revenue-providing offerings.
So, origin is a big part of it. But ignoring protocol is misguided. QoS is explicitly about protocol, but it's also a convenient end-
Re: (Score:2)
Re: (Score:2)
Net Neutrality is a common term, not a technical
Right, but it has a very technical meaning to those who understand the technical merit and implications. You're simply attempting to ignore the technical merit and ride the political bullshit speak so as to attempt grasp at justification to support your obtuse line of reasoning.
Factually, politicians are obfuscating the technical merit in an effort confuse those, like you, who don't understand the technical details. In doing, they hope to come out on top from the confusion of the ignorant masses. Basically,
Re: (Score:2)
You fucking work with computers, you're not fucking Merriam-Webster.
And you apparently dont understand what the issue is. Boiling it down to "throttling different types of traffic differently" is just plain wrong, and not in some stupid merriam webster sense of it. There are legitimate uses for throttling traffic; the whole point of net neutrality is to prevent abuses by ISPs in the name of extorting money out of various content providers. The point ISNT to cripple legitimate traffic shaping techniques, and trying to shoehorn in anti-QoS clauses is precisely why techies
Re: (Score:2)
Both are NN issues. Because what is already happening is tunneling things like BT in HTTP, where everything defaults to Port 80 for data traffic between two points. The problem is that people are violating the very nature of the system to route around perma-broken links. You're just shoveling the problem up the Stack.
QoS only works for temporary / bursty problems. When you have someone not having enough bandwidth to handle NORMAL traffic, all the time, then no amount of QoS can fix the problem. And how prov
Re: (Score:2)
treat **all content**, sites, and platforms equally
So if you bump all HTTP traffic down to "bulk", and all SIP (VOIP) traffic up to "premium", that IS NOT NET NEUTRALITY. Youre still treating all content in a protocol equally. It can be abused, but it is a legitimate technique. Net neutrality issues would be giving Skype VOIP traffic priority over other traffic in the same protocol.
Dont understand why thats important? Consider this-- if an HTTP traffic is delayed 1500 ms, it is still useful. If the packet has to be resent, it can. VOIP traffic is usel
Re: (Score:2)
There is a difference between filtering and shaping. Filtering prevents the delivery. Shaping simply delays the delivery. The first is bad. The second, if used correctly, as the article outlines, is very, very, very, very, very, very good.
So yes, filter is a NN issue but in this case, there is absolutely no reason to presume filter is the least bit topical. As such, attempting to bring in NN into the discussion is nothing but a red herring.
Re: (Score:2)
Net Neutrality is indeed different than QoS... and if their data shows that the QoS for BT drops off by 80% while HTTP drops off by 40%, then you ALSO have a Net Neutrality issue, as Net Neutrality has to do with discriminating with types of data or origin of data, instead of capacity.
So you're saying that my theory of QoS is likely correct. HTTP has a higher QoS than BT - hardly surprising.
Not sure why you're having trouble grasping that.
Agreed. Not sure why you're having trouble grasping that.
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Now your just putting words in my mouth.
*facepalm*
There are no quotes. I didn't say you literally said that. I know you didn't literally say that. That was for illustrative purposes to show how rediculas your position is. Your position is literally, completely without merit - in this regard.
Obviously you don't trust them. I don't either. I don't know anyone who does. But that's not really germane to the topic at hand. Not one bit. Not in the least. Why? Because its *ALL* readily explained by network engineering best practices and well established
Re: (Score:2)
Feel free to actually say something which isn't completely stupidity. Please explain why anything I said is invalid and I'll continue to engage your lunacy.
Face it, ignorance, stupidity, and completely unjustified paranoia is not the basis for a rational discussion - and yet that's the entire basis of your debate.
Seriously, if you have anything to offer, let's see. If its more baseless stupidity, just shut the fuck up and listen to the people who actually know what the fuck we're talking about.
Re: (Score:2)
Have a life? Fuck you troll! Pathetic.
Re: (Score:2)
For the last time, this has absolutely nothing to do with "truth." My facts have absolutely nothing to do with "trust". Period. It has everything to do with QoS and the fact they use it. Period.
Your response is emotionally charged. I've read those same articles you have. I understand the ISPs are not my best friend and they are out for themselves. They are a company after all. But none of that has anything to do with the topic at hand. Again, trust has nothing to do with it.
Re: (Score:2)
Natural usage will swell around 6pm, but it will do so organically with a bit of variance. I think they are talking about a specific stop watch accurate literal cliff at a specific time.
Re: (Score:2)
Okay... so why would they do that? Because they don't have enough bandwidth to go around. What's the alternative? Leave TCP congestion control to do it.
So... how do you tell the difference between TCP throttling BitTorrent usage due to exponential backoff triggered by packet loss, and traffic shaping?
Re: (Score:2)
So... how do you tell the difference between TCP throttling BitTorrent usage due to exponential backoff triggered by packet loss, and traffic shaping?
One slows down HTTP traffic, the other does not. I don't mind the network doing some traffic shaping, but I'm not sure what their justification is for letting customer A use 10Mb/s for HTTP, but only allowing customer B to use 5MB/s for BitTorrent (or some other protocol). Of course, this has to apply bidirectionally - BitTorrent uses a lot more upstream, but if you're letting customers saturate their uplink with HTTP uploads or outgoing SMTP, but not with BitTorrent then you've also got some explaining t
Re: (Score:2)
One slows down HTTP traffic, the other does not.
Wrong. Slashdot has even covered several articles which talk about the gory details of how you're wrong just on the surface, even without getting into the fact that congestion controls work because it slows traffic down. Holy shit, where do you get you're information. How the fuck do you think congestion is avoided if it doesn't slow things down? That's a rhetorical question. I know the answer - obviously you don't.
Next.
Re: (Score:3, Insightful)
I had a nice factual rebuttal, but this:
"Me, like most every reasonable person in the world"
Tells me you are so emotionally caught up in a perceived problem you stopped actually thinking about it.
Re: (Score:1)
Re: (Score:2)
"Me, like most every reasonable person in the world, certainly does not want to have You Tube, general web browsing, email, IRC, streaming music, game playing, or any of a number of other services negatively affected because Joe down the street is downloading his fifth illegal movie for the day,"
The problem with this stance is that ISP's naturally oversold bandwidth in the past, and this argument is tenuous today given technological advances and the huge decrease in bandwidth costs that have not been passed
Re:Dumb comment (Score:5, Informative)
After 6pm, Internet traffic for most ISPs goes through the roof. With it, latency and available bandwidth are typically negatively affected.
Internet traffic tends to look like a perfect curve that starts an upward trend around 700-800 for a given timezone and increases in a consistent patterned manor until 12 noon. There is a slight dip between noon and 1300 a second peak from 1300 till 1400 and a steady decline until 2300 to midnight. The decline from midday until midnight is slower but from all my experience in web traffic I don't see an increase in traffic after 1800 compared to the rest of the day.
Re: (Score:2)
The problem is people asserting they know what the Internet is for, and other rogue uses should be throttled so they can enjoy the Internet for it's int
Re: (Score:2)
You're conflating issues. I did not attempt to specifically validate their QoS scheme. All I said is its not only reasonable but all but impossible for them not to be pushing BT on the bottom of the QoS pile.
In fact, in other posts, I even offer QoS prioritization likely differs from ISP to ISP. And of course, the quality of the QoS configuration is likely to differ greatly. Regardless, none of that should be conflated with Net Neutrality issues nor the general desire to place BT at extremely low priorities
Re: (Score:2)
What if I'm Joe down the street, waiting on my linux iso to download, or waiting on my game to update? You are saying your ability to wa
Re: (Score:2)
Re: (Score:2)
Enforcing arbitrary QoS based on traffic type is retarded. The correct thing to do would be to allow the customers to set what traffic they want to prioritize. Either way to insist that your youtube is more important then someone else's bittorrent makes no sense. You pay the same amount and expect the same service. Just because you feel your traffic is more important (and large commercial interests think http > all other traffic because it can serve ads). If you arbitrarily pick a protocol and restr
Re: (Score:2)
> Why should I be throttled to 25kbs or less like my ISP tries to??
So that others can get a decent portion of the network?
I'm a little disappointed with BT, who appear to have stopped publishing contention ratios. I'm going to use Namesco (who run over BT's lines) as my example instead; http://www.names.co.uk/adsl.html [names.co.uk] . With at ratio of 50:1, you're sharing your upstream connection from the exchange to your ISP, with 49 other people. 20:1, 19 other people. Either way, the contract with the ISP is fairly
Release it! (Score:2)
Why are they worried about ISP reactions? They're just (hopefully) releasing data. It isn't biased or skewed, it just is. If the data is embarrassing to an ISP, that's the ISP's problem.
Freedom Box (Score:2)
Eben Moglen http://lastonk.blogspot.com/2011/02/freedom-box.html [blogspot.com] is trying to make a box that makes it damn hard to track people using this sort of stuff. I'm not tech savvy enough to know if these things would help in this situation, but I strongly suspect it would.
If nothing else... you could use them to create ad hoc darknets capable of distributing p-p without ever going through an ISP at all... I'm thinking this is the right kind of forum to find people with the skills to help Eben out.
Re: (Score:2)
UK ISPs are at the mercy of BT's infrastructure (Score:2)
display that most UK ISPs 'aggressively throttle BitTorrent traffic after 6 p.m. at night,' with speeds suddenly going 'off a cliff.'
No, that's quite normal for some areas. It's not just BitTorrent but everything, due to oversubscription on BT's infrastructure. Right down latency, like a 12ms ping turning into a 50ms epic journey.
Re: (Score:2)
Re: (Score:2)
Depends. You should see an undulating curve as people "log on" over time. If there's a sharp point on the bandwidth curve then it's generally a good indicator that something artificial is happening.
Could it be possible that at 6 PM the Bittorrent QoS level gets adjusted, and for the rest of the day it's at the same level as http and rtsp? Definitely artificial, but not necessarily nefarious.
Legal BitTorrent (Score:2)
Re: (Score:1)
So, in order to distribute something I made using BitTorrent I would need to get a cert?
And how much is this going to cost me?
Seriously, just cause I made it myself doesn't make it illegal; requiring certs is just illigitimizing all non-corporate stuff.
Re: (Score:1)
I can imagine a couple reasons why BitTorrent might be throttled. First, I'd bet BitTorrent users are sending and receiving a lot more traffic than the average user. ... If BitTorrent users want faster speeds, I'd recommend finding ways to make the BitTorrent landscape less populated by illegal warez. ... But, until that happens, I really don't care if ISP throttle or cut BitTorrent traffic entirely. ...
Or perhaps ISPs should actually provide the bandwidth that they advertise (not to mention "unlimited" data transfer)?
We Need To Stop This Madman, Now! (Score:2)
Are you... (Score:2)
...pondering what I'm pondering, BitTorrent?
Some throttling is needed when it come to BT (Score:2)
BitTorrent claims that its service can, for example, display that most UK ISPs 'aggressively throttle BitTorrent traffic after 6 p.m. at night,' with speeds suddenly going 'off a cliff.' Suffice to say that such information could prove to be very useful for consumers and advocates of Net Neutrality."
And a jolly good thing this is too. I need my ping to be as fast as possible in order to play online first person shooters and at times when bandwidth is short I would rather they throttled stuff that would not be adversely affected by a bit of a delay and prioritised my traffic that needs to get to its destination more quickly for it to be of any use.
I also understand them prioritising web browsing over P2P as well as P2P traffic is generally far more constant over a 24 hour period. As an P2P user as well
Re: (Score:2)
Re: (Score:2)
I have a 1Mb down pipe at my house. I have my P2P client throttled down to 25KB/s which I think is quite reasonable (1/5 of my theoretical maximum rate rate). There are often times that I can't even do basic web browsing because my ISP has throttled my bandwidth. I've done some minimal work to make sure I wasn't saturating my home router and such. It's the pipe. Pause my P2P client and about 5 minutes later I get back to my normal bandwidth amount.
Usually the problem is upload speed rather than download speed. I do not throttle my download speed at all but I throttle my P2P upload speed to be half of that available. This ensures that the packets that I am sending out requesting a web page always get through. The important thing to remember is that even though is that regardless of what you are doing, it requires bandwidth in both directions for the packets acknowledging the packet in the opposite direction got through.
Even though you have a 1Mb down
Re: (Score:2)
Something I haven't heard ISPs doing, but which makes some sense to me...
Instead of throttling services, why not throttle based on number of ports?
Using this method the first 40 or so ports opened to the gateway will all be at a standard QoS. Ports opened after that will be throttled.
Sure, this means that if you have bt running and then try to play an FPS, it's your FPS's ports that are going to be throttled at first, but after the current round of BT segments complete, the FPS connections will move up the
Nike Slogan (Score:1)
Just do it!
That's a good trick (Score:2)
As I read this I thought... "Why announce something and not release it? Everyone will bug you until you do, now you have to do it"
Then I realized... ahah! This creates a buzz and demand for the information that wasn't there before. If he'd simply released it how many would have noticed? Yet mentioning it first creates controversy and gets people insisting on their right to the information guaranteeing instant popularity and followup discussion.
Yeah I should have figured this out months ago in the lead-up
WoW Issues recently (Score:2)
Blizzard implemented a change recently. You can go google it in recent posts, but the net effect was a big drop in latency for most WoW players. The details aren't known, but the effect was to basically triple the amount of game data being sent up and down stream. WoW does not transmit a ludicrous number of bytes per second, so tripling it doesn't break anyone's bandwidth limits or anything- but suddenly, some of our raid (always the same people) started disconnecting hard. They could barely be online-
Do it! (Score:2)
Or give it to wikileaks whilst Julian Assange hasn't been assassinated by some upset government yet!
A surprise in the UK? Doubtful. (Score:2)
I don't think people who are paying £7.50 ( $13 ) or less per month for their Internet connection should be particularly surprised if they are subject to throttling and blocking. Do they really think that is a realistic price?
Mainstream broadband in the UK has been commoditised to such a degree that it is now "cheaper" than a month's worth of tabloid papers. That should illustrate how many people here think.
If one looks at other UK ISPs that do not throttle, block or cap they usually cost upwards of
Re: (Score:2)
Oh, I thought that it was Al Gore!
(And since he's such a polarising figure, I'll explain for mods on the wrong side of the political aisle from this joke: IT'S A JOKE.)
Re: (Score:2)
Re: (Score:2)
you are so dumb, if you don't like it, use an independent providers that support MLPPP, shutting up and paying for download is retarded