ISP did not throttle YouTube traffic at peak hours, telecom regulator finds

Allegations that an ISP throttled traffic from YouTube and other Google properties at peak hours were unfounded, according to the French telecommunications regulator.

Any problems with access to YouTube and other sites at peak hours were simply due to the overall volume of data flowing between the ISP, Free, and the rest of the Internet, said the regulator, Arcep.

Following a complaint from the French consumer association UFC-Que Choisir last September that Free appeared to be throttling traffic from YouTube, Arcep began an investigation, questioning the companies involved and their transit providers, and studying traffic flows.

Arcep spent six months studying traffic flows at Free, finding that Free and Google exchange traffic directly, via peering, and indirectly through a number of international transit providers, and that both routes are congested during peak hours, as they are at other ISPs.

Furthermore, there were no signs that Free was employing traffic management techniques on its network to discriminate against traffic based on the content, origin, destination or protocol used. In other words, Arcep said, “No practices contrary to principles of net neutrality were observed.”

At the time of the complaint, Free was known to be in dispute with Google over who should pay for additional interconnect bandwidth between the two. In January it briefly added a function to the DSL routers of millions of its customers to block ads served by Google and other major ad networks. The ad blocker, enabled by default although Free’s customers could choose to disable it, cut into the revenue the ad networks received from French surfers. Within days, the French government announced that such action was against the principles of net neutrality, and with the next router software update, the feature disappeared. There was no change in the number of complaints from Free customers on user forums and elsewhere about difficulties viewing YouTube videos at peak hours, though.

Neither Google nor Free immediately responded to requests for comments on the investigation’s findings.

Having recognized the importance to customers of the way in which ISPs and Internet companies balance their use interconnection and peering, the regulator now plans to conduct six-monthly reviews of operators’ practises and the effect this has on customers.

ISP did not throttle YouTube traffic at peak hours, telecom regulator finds

Allegations that an ISP throttled traffic from YouTube and other Google properties at peak hours were unfounded, according to the French telecommunications regulator.

Any problems with access to YouTube and other sites at peak hours were simply due to the overall volume of data flowing between the ISP, Free, and the rest of the Internet, said the regulator, Arcep.

Following a complaint from the French consumer association UFC-Que Choisir last September that Free appeared to be throttling traffic from YouTube, Arcep began an investigation, questioning the companies involved and their transit providers, and studying traffic flows.

Arcep spent six months studying traffic flows at Free, finding that Free and Google exchange traffic directly, via peering, and indirectly through a number of international transit providers, and that both routes are congested during peak hours, as they are at other ISPs.

Furthermore, there were no signs that Free was employing traffic management techniques on its network to discriminate against traffic based on the content, origin, destination or protocol used. In other words, Arcep said, "No practices contrary to principles of net neutrality were observed."

At the time of the complaint, Free was known to be in dispute with Google over who should pay for additional interconnect bandwidth between the two. In January it briefly added a function to the DSL routers of millions of its customers to block ads served by Google and other major ad networks. The ad blocker, enabled by default although Free's customers could choose to disable it, cut into the revenue the ad networks received from French surfers. Within days, the French government announced that such action was against the principles of net neutrality, and with the next router software update, the feature disappeared. There was no change in the number of complaints from Free customers on user forums and elsewhere about difficulties viewing YouTube videos at peak hours, though.

Neither Google nor Free immediately responded to requests for comments on the investigation's findings.

Having recognized the importance to customers of the way in which ISPs and Internet companies balance their use interconnection and peering, the regulator now plans to conduct six-monthly reviews of operators' practises and the effect this has on customers.

The Secret Behind a Mysterious Traffic Code? It’s Made Up

Jessica Sierra It’s a head-scratcher all right. The signs say “No Standing,” but search the official city directory and its state and federal counterparts and try to find what AWM stands for.

No Standing signs, mostly in Lower Manhattan, prohibit standing at a few locations on weekdays except for authorized vehicles. The only clue to exactly who is authorized are the letters AWM.

Turns out, AWM doesn’t stand for anything at all. The letters were a figment of the imagination of Samuel I. Schwartz, a former traffic commissioner, now a private traffic engineer known as Gridlock Sam.

In the 1980s, he was asked to reserve spaces for certain secretive federal agencies, which did not want to advertise their presence or have their unmarked vehicles identified.

The request was forwarded to the city after a particularly embarrassing intergovernmental episode: F.B.I. agents went inside a building to arrest someone. When they came out with the suspect, their car had been towed.

So the city installed dozens of signs, mostly downtown near the offices of the F.B.I. and the Secret Service, but also in Midtown for agents of the Postal Police and the Drug Enforcement Administration.

“The people who you shouldn’t know who they are,” Mr. Schwartz said.

“I came up with those letters because I thought it meant nothing and nobody would figure it out and it was in the days before you could Google anything,” he said. (Doing an Internet search for the letters today doesn’t help much. The first result in a Google search is for the Association for Women in Mathematics.)

So out of thin air, Mr. Schwartz made up the AWM designation, which the agencies and, presumably, traffic enforcement agents, all wink at. Robert Cassar, the president of Local 1182 of the Communications Workers of America, which represents traffic enforcement agents, said he doubted most people would park in those spots simply because they are mostly adjacent to government buildings.

“I would bet if you look at the summonses we issue you would find very few for people parking in those spots,” Mr. Cassar said.

Mr. Schwartz said he was unaware of any other fictional agency initials. A number of people have tried to figure out what the letters stand for, he said, but nobody has come even close.
“I just randomly picked the letters and issued AWM permits,” he recalled.

Internet traffic jams, meet your robot nemesis

On an 80-core computer at the Massachusetts Institute of Technology, scientists have built a tool that might make networks significantly faster just by coming up with better algorithms.

The system, called Remy, generates its own algorithms for implementing TCP (Transmission Control Protocol), the framework used to prevent congestion on most networks. The algorithms are different from anything human developers have written, and so far they seem to work much better, according to the researchers. On one simulated network, they doubled the throughput.

Remy is not designed to run on individual PCs and servers, but someday it may be used to develop better algorithms to run on those systems, said Hari Balakrishnan, the Fujitsu professor in Electrical Engineering and Computer Science at MIT. For now, it's churning out millions of possible algorithms and testing them against simulated networks to find the best possible one for a given objective.

IP (Internet Protocol) networks don't dictate how fast each attached computer sends out packets or whether they keep transmitting after the network has become congested. Instead, each system makes its own decisions using some implementation of the TCP framework. Each version of TCP uses its own algorithm to determine how best to act in different conditions.

These implementations of TCP have been refined many times over the past 30 years and sometimes fine-tuned for particular networks and applications. For example, a Web browser may put a priority on moving bits across the network quickly, while a VoIP application may call for less delay. Today, there are 30 to 50 "plausibly good" TCP schemes and five to eight that are commonly used, Balakrishnan said.

But up to now, those algorithms have all been developed by human engineers, he said. Remy could change that.

"The problem, on the face of it, is actually intractably hard for computers," Balakrishnan said. Because there are so many variables involved and network conditions constantly change, coming up with the most efficient algorithm requires more than "naive" brute-force computing, he said.

Figuring out how to share a network requires strategic choices not unlike those that cyclists have to make in bike races, such as whether to race ahead and take the lead or cooperate with another racer, said Balakrishnan's colleague, graduate student Keith Winstein.

"There's a lot of different computers, and they all want to let their users browse the Web, and yet they have to cooperate to share the network," Winstein said.

However, Remy can do things that human algorithm developers haven't been able to achieve, Balakrishnan said. For one thing, current TCP algorithms use only a handful of rules for how a computer should respond to performance issues. Those might include things like slowing the transmission rate when the percentage of dropped packets passes some threshold. Remy can create algorithms with more than 150 rules, according to the researchers.

To create a new algorithm using Remy, Balakrishnan and Winstein put in a set of requirements and then let Remy create candidates and try them against software that simulates a wide range of network conditions. The system uses elements of machine learning to determine which potential algorithm best does the job. As it tests the algorithms, Remy focuses on situations where a small change in network conditions can lead to a big change in performance, rather than on situations where the network is more predictable.

After about four to 12 hours, Remy delivers the best algorithm it's found. The results have been impressive, according to the researchers. In a test that simulated a fast wired network with consistent transmission rates across physical links, Remy's algorithms produced about twice the throughput of the most commonly used versions of TCP and cut delay by two-thirds, they said. On a simulation of Verizon's mobile data network, Remy's algorithms gave 20 percent to 30 percent better throughput and 25 percent to 40 percent lower delay.

But don't expect blazing page loads just yet. Balakrishnan and Winstein cautioned that Remy is still an academic research project.

For one thing, it hasn't yet been tested on the actual Internet. Though Remy's algorithms would probably work fairly well out there, it's hard to be sure because they were developed in simulations that didn't include all of the Internet's unknown variables, according to Balakrishnan. For example, it may be hard to tell how many people are active on a particular part of the Internet, he said.

If machine-developed TCP algorithms do end up going live, it will probably happen first on private networks. For example, companies such as Google already fine-tune TCP for the requirements of their own data centers, Balakrishnan said. Those kinds of companies might turn to a system like Remy to develop better ones.

But even if Remy never makes it into the real world, it may have a lot to teach the engineers who write TCP algorithms, Balakrishnan said. For example, Remy uses a different way of thinking about whether there is congestion than TCP does today, he said.

Though the researchers understand certain tools that Remy uses, they still want to figure out how that combination of tools can create such good algorithms.

"Empirically, they work very well," Winstein said. "We get higher throughput, lower delay and more fairness than in all the human designs to date. But we cannot explain why they work."

"At this point, our real aspiration is that other researchers and engineers pick it up and start using it as a tool," Balakrishnan said. "Even if the ultimate impact of our work is ... more about changing the way people think about the problem, for us, that is a huge win."

Where Car Is King, Smartphones May Cut Traffic

But now licensed cabdrivers in this city where the car remains king are facing their greatest competition in half a century, from new ride-sharing programs that use smartphone apps to connect drivers and passengers.

At the end of a night of drinking recently, rather than hailing a cab, Trisha DiFazio tapped an app called Lyft on her phone to summon a ride. Minutes later, a graduate student moonlighting as a driver pulled up in a Toyota S.U.V. with Lyft’s signature pink mustache affixed to its grille.

“This is so much cheaper than a cab, and so much easier,” said Ms. DiFazio, 31. “I absolutely think my friends drink and drive less because of this.”

Transit experts say these new services, which appeal to younger riders, could play a crucial role in ending the reign of single-occupant cars (and unending traffic) in Los Angeles, and many young residents have embraced them as a cheaper, more reliable and, well, more fun way to get around the city. But some of the city’s licensed cabdrivers have another name for ride-sharing services: illegal bandit taxis.

Last month, city regulators sided with the cab companies and sent cease-and-desist orders to Lyft and two other companies offering ride-sharing services, Uber and Sidecar, ordering them to shut down immediately. “It’s the Wild West with these operators,” said Tom Drischler, the taxicab administrator for the city’s Department of Transportation. “You have folks driving private cars and picking up strangers. Public safety is our assignment, and I don’t think it’s safe for the public.”

All three companies have refused to leave town, asserting that agreements they made with state regulators allow them to operate anywhere in the state.

Smartphone-based ride-hailing services have faced opposition from taxi companies and city regulators almost everywhere they have appeared, from Las Vegas to Cambridge, Mass. But the fight here is complicated by longstanding efforts by city officials to alleviate traffic and reduce drunken driving.

Last week, on his first full day in office as the new mayor of Los Angeles, Eric Garcetti proudly proclaimed that Los Angeles was beginning to leave behind its culture of car ownership. And Juan Matute, director of the Local Climate Change Initiative at the University of California, Los Angeles, said that if more people used ride-sharing services — even just 3 percent of the population, he said — substantial reductions in driving in Los Angeles could result.

“It would be easier to share rides that are incidental to daily life,” he said.

Unlike the new companies, the taxi industry here, which pays about $4 million a year to the city in franchise fees and vehicle permits, is tightly regulated for safety.

In addition to franchise and inspection fees, regulated taxis have to serve far-flung and low-income parts of the city where Lyft drivers need not venture. The city also requires cab companies to offer disabled-accessible vehicles, which William Rouse, the general manager of Los Angeles Yellow Cab, said his company did at a loss. “If they’re going to skirt our entire cost structure,” he said, “then they’re going to be able to get away with charging less and still make a profit.”

Rather than a metered fare, Lyft has a suggested donation, typically about 20 percent cheaper than a cab. Riders can pay as much or as little as they like, although those who frequently shortchange drivers have a tough time getting rides.

Jano Youssefi, an Iranian immigrant who has driven cabs in Los Angeles for 15 years, said his income had dropped at least 20 percent in the six months since Lyft rolled into town. “We are not making money anymore,” Mr. Youssefi said.

Yet, even as the threat of arrest hangs over drivers — some Lyft drivers removed the pink mustaches from their grilles after the cease-and-desist order — interest in the new programs seems to be on the increase. Drivers for Lyft and Sidecar use their own cars, and typically keep about 80 percent of the payments passengers offer, while the company keeps the rest. Uber also offers a service that connects passengers with licensed, professional luxury car drivers, as well as people driving their own cars.

Justin Riley, who began driving for Lyft a month ago, said he enjoyed the flexible hours, which gave him time to work on getting his tech start-up company off the ground. “It’s allowed me to discover L.A.,” he said. “I’ve met so many diverse people, heard so many great stories.”

But Mr. Drischler, the city taxicab administrator, said the collegial environment was part of the problem. Before safety partitions were installed in cabs two decades ago, he said, one cabdriver was killed every 18 months in Los Angeles. “I’m honestly worried for the drivers of those companies,” he said. “I have two daughters in their 20s, and I would never let them drive for those companies.”

The California Public Utilities Commission is writing rules for smartphone ride-hailing services.