Tuesday, September 09, 2014

The Great Wall of incumbent telecommunications infrastructure

Every divided territory has a boundary, border or wall mark its limits. Throughout much of the United States, it's that place where despite having plenty of existing infrastructure, incumbent telephone and cable companies draw an arbitrary boundary where the "footprint" of their landline telecommunications infrastructure capable of providing modern Internet service ends and the digital divide begins.


 
On the other side of the border, digital subscriber line (DSL) signals peter out and can't reliably provide service over twisted pair copper designed for a time when the Internet hadn't yet been conceived. There are also pockets of homes and small businesses in sufficiently close proximity of each other to qualify for cable Internet service, but do not because there are aren't enough along short spans of roadway between them and the wall's edge. Like border signs, the boundaries of these areas are often demarcated with utility pole advertisements offering those in the digitally deprived zone "New Super Fast Internet" that's actually substandard satellite service that should only be offered in the most remote and isolated areas of the U.S.

The U.S. Federal Communications Commission could tear down the wall by enforcing the universal service provision of Title II, Section 254(b) of the Communications Act of 1934 (as amended in 1996) that provides that access to advanced telecommunications and information services be available in all regions of the nation. Section 202 of the law also contains an anti-redlining provision barring providers from discriminating against localities in providing service.



Friday, September 05, 2014

FCC chair signals end of “broadband” era and rise of FTTP

Sooner or later – more likely sooner – the Federal Communications Commission (FCC) will recognize the irrelevance and futility of defining and subsidizing landline premise telecommunications infrastructure based on specified “broadband” download and upload speeds as Internet bandwidth demand growth tracks Moore’s Law for microprocessor processing power, doubling every 18-24 months.
Consequently, it will likely repurpose the mission of the FCC’s Connect America Fund (CAF) program created to subsidize infrastructure construction in high cost areas to instead help defray the cost of deploying fiber to the premise (FTTP) infrastructure in these areas. At the same time, the FCC could also realize that significantly greater funding will be needed to do the job than the $9 billion the CAF has budgeted for its second phase covering the period 2014-2019.

The FCC this year recognized that its current eligibility criterion for CAF subsidies is potentially outdated. It’s targeted to high cost areas where premises are not served by landline connections providing at least 4 Mbs down and 1 Mbs up. The FCC issued a notice of inquiry in August to take testimony as to whether that standard should be increased and modified to include latency as well as speed.

In prepared remarks delivered this week, FCC Chairman Tom Wheeler suggested 25 Mbs should be considered the new minimum. He went on to observe that might also be too low and only a quarter of the throughput that Americans presently expect given their growing appetites for high definition streaming video and multiple connected devices in their homes and small businesses.

“Today, a majority of American homes have access to 100 Mbs,” Wheeler continued. “It is that kind of bandwidth that we should be pointing to as we move further into the 21st century. And while it’s good that a majority of American homes have access to 100 Mbs, it is not acceptable that more than 40 percent do not.”

Relative to high cost areas, Wheeler noted the FCC “will continue to establish requirements for our universal service programs, but beyond that, consumers are establishing their own expectations.” That recognition of end user needs represents a significant departure from existing policy where telecommunications providers and governments tell consumers in these areas what they should expect instead of the reverse. It’s also an implicit recognition that there should be a single standard and not a separate and lesser standard for high cost areas of the nation. Which makes sense given that core content providers and other services are tailored for a single standard of quality at the network edge.

Noting FTTP deployments in several metro areas of the U.S., Wheeler impliedly recognized FTTP infrastructure is replacing the speed-based “broadband” metal wire paradigm of the legacy telephone and cable companies. That model utilizes “bandwidth by the bucket,” speed-based pricing tiers based on the assumption that metal wire infrastructure has limited carrying capacity and that service must accordingly be rationed and priced based on demand.

Wheeler recognized with FTTP, that pricing model that irks many consumers faces obsolescence. “Once fiber is in place, its beauty is that throughput increases are largely a matter of upgrading the electronics at both ends, something that costs much less than laying new connections,” Wheeler said.

Wheeler also acknowledged that mobile wireless services cannot substitute for FTTP. “While LTE and LTE-A offer new potential, consumers have yet to see how these technologies will be used to offer fixed wireless service,” he said.

Thursday, September 04, 2014

FCC's Wheeler: US needs more high-speed broadband competition | PCWorld

FCC's Wheeler: US needs more high-speed broadband competition | PCWorld: U.S. residents lack meaningful choices for broadband providers that offer 25Mbps or faster download speeds, and the U.S. Federal Communications Commission will push for more competition, the agency’s chairman said Thursday.

While more than 93 percent of U.S. residents have access to a broadband provider, fewer than 15 percent can buy service from more than two wired providers that offer “yesterday’s broadband” with 4Mbps download speeds, FCC Chairman Tom Wheeler said during a speech at Washington, D.C., startup incubator 1776.

“At the low end of throughput ... the majority of Americans have a choice of only two providers,” Wheeler said. “That is what economists call a duopoly, a marketplace that is typically characterized by less than vibrant competition.

As long as Internet service providers own the infrastructure that connects customer premises, there will never be any meaningful degree of competition, owing to the fact that telecommunications infrastructure due to high costs and barriers to entry functions in a natural monopoly market. As Andrew Cohill wrote in his 2010 white paper, that's about as inefficient and senseless as having FedEx or UPS operate proprietary roads to serve neighborhoods that are closed to competing shipping services.

The policy of the United States has been to preserve this very market structure of which the Federal Communications Commission chair laments. What's needed to achieve any level of real competition is to encourage and fund the construction of publicly owned open access fiber to the premise networks where ISPs compete to sell services to customer premises. Call it the public option for telecommunications in the Internet age.

Wednesday, September 03, 2014

Broadband and the future of learning | Computerworld

Broadband and the future of learning | Computerworld: Since learning may take place anywhere and anytime, connected learners also need broadband access outside of school. Although 70% of U.S. households now have broadband, millions of households still do not. Private-sector initiatives are helping to expand access. For example, Comcast’s Internet Essentials program offers low-income families broadband service for $9.95 a month, along with the option to purchase an Internet-ready computer for under $150 and free digital literacy training. In its first three years of operation, the program has provided affordable broadband service to more than 350,000 households.

It should be noted that Comcast and other incumbent legacy providers redline many neighborhoods, leaving them without access to modern landline Internet connectivity at any price.

There are also promising public-private partnerships to increase access. In Forsyth County, Georgia, the local school district worked with the Chamber of Commerce to create a directory of free Wi-Fi locations in the community and to provide participating businesses with signs indicating where free Wi-Fi is available. And a middle school in Manchester, Tenn., that has equipped all sixth-graders with iPads had convinced local businesses to open their Wi-Fi hotspots to students to maximize the benefits of their technology tools.

Public-private partnerships need to go far beyond Wi-Fi and help construct fiber to the premise infrastructure to make blended learning possible since it heavily relies on students having adequate access in their homes. A good example is in Utah, where an investment firm, MacQuarie Capital, is partnering with the Utah Telecommunications Open Infrastructure Agency (UTOPIA) to finance and complete the construction of open access fiber to the premise infrastructure.

Thursday, August 28, 2014

Telecom Plan Raises Questions About Future Internet Service | Vermont Public Radio

Telecom Plan Raises Questions About Future Internet Service | Vermont Public Radio: “It’s shortsighted to make that investment in technology that can’t go the whole nine yards,” says Irv Thomae, chairman of the governing board of ECFiber, which currently serves 800 customers in six central Vermont towns.

Thomae says the draft plan doesn’t represent a commitment to the Legislature’s goal.

“If the Telecom Plan says we aren’t to take the 100 Mbps seriously, then we aren’t going to take it seriously,” he says.

Thomae says state funded "dark fiber" projects constructed by the Vermont Telecommunications Authority should be the model for reaching the 2014 goal. These projects enable service providers to lease space and compete for customers.

Thomae says the state should raise money through the sale of bonds to finance an extensive dark fiber system.

Thomae raises a key issue on U.S. telecom infrastructure planning and financing policy. The nation is at an inflection point where the service line extensions of the legacy telephone and cable companies have gone about as far as they can within their business models in terms of making landline Internet service accessible to all American homes and businesses. And possessing the capacity to deliver the bandwidth that will be needed going forward as bandwidth demand doubles every couple of years or so, consistent with Moore's Law on microprocessor development.

Vermont's situation is a metaphor for the United States as a whole and points to the need for greatly expanded public sector financing capacity for this infrastructure that's as critical to the 21st century as highways and electricity were to the 20th.

How big telecom smothers city-run broadband | Center for Public Integrity

How big telecom smothers city-run broadband | Center for Public Integrity: “We don’t quarrel with the fact that AT&T has shareholders that it has to answer to,” Bowling said with a drawl while sitting in the spacious wood-paneled den of her log-cabin-style home. “That’s fine, and I believe in capitalism and the free market. But when they won’t come in, then Tennesseans have an obligation to do it themselves.” 

Republican Tennessee State Senator Janice Bowling puts this debate over the role of the public sector in financing or building telecommunications infrastructure into the proper perspective. It's not a contest over capitalism or any other economic philosophy. It's about the hard reality that markets aren't perfect and can and do fail. When that market is for a service like telecommunications that plays such a central role in the health of the economy as a whole, public sector involvement is entirely appropriate and the interests of a single sector of the economy must take a subordinate position.

At a meeting three weeks after Bowling introduced Senate Bill 2562, the state’s three largest telecommunications companies — AT&T, Charter, and Comcast Corp. — tried to convince Republican leaders to relegate the measure to so-called “summer study,” a black hole that effectively kills a bill. Bowling, described as “feisty” by her constituents, initially beat back the effort and thought she’d get a vote.

That’s when Joelle Phillips, president of AT&T’s Tennessee operations, leaned toward her across the table in a conference room next to the House caucus leader’s office and said tersely, “Well, I’d hate for this to end up in litigation,” Bowling recalls.

Actually, no. Legacy incumbent telephone and cable companies love litigation because it fits perfectly with their strategy of buying time and years of delay since they are unable to invest sufficient funds to upgrade their monopolistic and dupolistic telecommunications markets due the limitations of their business models.

Tuesday, August 26, 2014

Unpacking claims of “unfair competition” when the public sector finances or builds fiber to the premise infrastructure

Incumbent telephone and cable companies often cry “unfair competition” when the public sector invests in or builds fiber to the premise (FTTP) infrastructure. Let’s unpack that assertion. From the point of view of these companies, anyone who builds infrastructure they don’t own is a competitor. They really don’t compete to gain customers in a given geographical area. That’s because telecommunications infrastructure isn’t truly a competitive market characterized by many sellers and buyers. Rather than competing for customers, the incumbents’ true interest is in protecting their monopoly or duopoly status.
True competition occurs in a market where buyers and sellers are on a level playing field and buyers have relatively equal access to market players and information on their services, benefits, prices and value offered. That doesn’t happen in telecommunications infrastructure. Incumbents have the upper hand in deciding which neighborhoods they will serve, what services will be offered and at what price. And they don’t disclose where they plan to build FTTP infrastructure.

The public sector typically gets involved in investing in or building FTTP infrastructure not to compete with the incumbents, but to remedy the market failure they create given their power to pick winners and losers among the neighborhoods they opt to serve and those they choose to redline and not offer service.

Finally, since the public sector typically invests in open access infrastructure and provides wholesale access to Internet service providers (including the incumbents), that’s also not direct market competition with incumbent telephone and cable companies. It’s an entirely different playing field and certainly not the same one used by the incumbents who won’t play ball unless they own the field. Hence, there’s no direct competition, fair or unfair.

U.S. FTTP infrastructure projects falling into 2 categories


The construction of fiber to the premise (FTTP) Internet infrastructure in the United States is falling into two main categories:
  1. Projects in large and midsize metro centers such as those started or planned by Google Fiber, AT&T and Century Link as well as some cable companies. An article in the July 2014 issue of Broadband Communities magazine lists these deployments.
  2. Community or regional projects by local governments, utility cooperatives and public-private partnerships serving less densely populated areas not containing large cities such as those tracked by the Institute for Local Self Reliance.

The bifurcation of these infrastructure projects is distinguished by the economic health of their respective markets. Those in the first category undertaken by investor-owned providers that need a rapid return on investment are targeted to markets undergoing rapid economic growth, Broadband Communities editor Masha Zager writes in her article on large metro projects, citing the FTTP deployment strategy of Cox Communications:

Cox explicitly named rapid growth as one of its criteria for selecting cities for gigabit deployments. In contrast to municipalities, which often deploy fiber in an effort to jump-start lagging economies, large players favor localities that are healthier to begin with.

For the second category of projects, FTTP is clearly an economic development strategy to a far greater extent than the first. Unlike those in the first category financed by the impatient capital of telcos and cablecos burdened with high debt loads and large shareholder dividend obligations, community or regional projects will rely on patient capital. Sources include long term public bonds and creative public-private partnerships that blend public and private funding such as the Utah Telecommunications Open Infrastructure Agency (UTOPIA).

The second category is also distinguished from the first by the ownership and business models of the network infrastructure. In the first category of investor-owned projects, the network is a proprietary, closed access property. The telcos and cablecos that own the networks charge a retail monthly subscription fee to connecting premises.

By contrast, the second category is more likely to utilize an open access business model (such as UTOPIA) where fiber infrastructure is like a public works project such as a road or highway. Instead of selling individual subscriptions to customer premises, an open access model operates as a wholesaler selling network access to Internet service providers who provide services to customer premises. This model is a better option for the second category of projects because it removes the business risk of getting sufficient numbers of premises to sign up for service in order for the network deployment to be economically viable.

Sunday, August 24, 2014

AT&T falls short on California landline infrastructure upgrades




AT&T California has not met requirements for the build out of infrastructure to make Internet-based video services available to at least 50 percent of its California telephone service area as of year-end 2012. 

That’s according to the California Public Utilities Commission’s Sixth Annual DIVCA Report for the year ending December 31, 2012 (issued July 31, 2014). DIVCA – the Digital Infrastructure and Video Competition Act of 2006 – specifies a five-year build out period of 2008 through 2012. (The relevant reference is at page 9 of the report.)

AT&T California qualified for relief from the five-year infrastructure build out requirement under a DIVCA exception in cases where a provider has been unable to sell Internet video services to at least 30 percent of households in its telephone service area.

This in turn has resulted in a significant customer quality issue. Many households in AT&T California’s telephone service territory are unable to order landline-delivered Internet services since AT&T video services (branded as U-Verse and which includes bundled Internet access and voice service) are delivered over decades-old copper cable plant. Instead, these customers are offered only substandard, obsolete dialup Internet service that cannot support the delivery of video services.

Friday, August 15, 2014

Advocates of municipal broadband face resistance over high-speed access | GazetteNet.com

Advocates of municipal broadband face resistance over high-speed access | GazetteNet.com: Foes, including private Internet service providers such as Comcast, AT&T and Time Warner Cable, have a different view. They say they are spending hundreds of millions of dollars upgrading infrastructure to give high-speed access to every American, and that government shouldn’t compete against private companies, which must pay taxes and make a profit.
The assertion regarding "upgrading infrastructure to give high-speed access to every American" is a false statement. These providers segment their markets and redline neighborhoods deemed less profitable and have no plans to serve them, all the while making promises they cannot stand behind. The reason they cannot is they are constrained by inpatient shareholder investment capital and short term business models inappropriate for high cost capital infrastructure that can require decades to produce a return on investment.

The claim that government is unfairly competing with private sector telecommunications providers is also false in a strict economic sense. Competitive markets are characterized by many buyers and sellers. In telecommunications infrastructure, there are many buyers and users but few sellers, making the market a natural monopoly or duopoly. When the public sector steps in to build and/or finance telecommunications infrastructure, it does so because this market environment combined with the previously mentioned business model limitations of investor-owned telephone and cable companies produces market failure on the sell side. That failure has left millions of Americans unable to order modern Internet landline-delivered services at their homes and small businesses.

Monday, August 11, 2014

Latta ascribes wrong cause to constrained investment in last mile infrastructure

Rep. Bob Latta Weighs in on STELA, Title II & E-Rate | USTelecom: On the topic of Title II, net neutrality and broadband legislation, Latta said, “First of all, I believe in an open Internet — a free Internet without government intervention. When you look at where the Internet has come and where it’s going in the future, this has all been done on the private sector. It’s not been done because of what the Federal government has done.” According to Latta, by putting broadband under Title II to make it more like telecommunications using a law from 1935, “What we will see happen then is that the innovation out there that’s spurred about a trillion dollars in private investment is all of a sudden going to be tied up like it would be with a telephone company. We don’t want that. Because once you start that up, then all of a sudden innovation is going to slow up — not only innovation — the dollars put in it and the tens of thousands of jobs being created. So we don’t want that to happen. We want to make sure that it remains free, it stays open and it stays away from government control.”

The problem with this position is regulation isn't the cause of what the Federal Communications Commission estimates as nearly 20 million Americans who are not offered landline Internet connections to their homes. In addition, much of the nation remains served by outdated twisted pair copper plant built many decades ago for analog telephone service and not fiber to the premise needed today and in the future as bandwidth demand grows dramatically.

If legacy telephone and cable companies had innovative solutions to build that necessary infrastructure, they would have pursued them over the past two decades. They haven't been able to do so not because of regulatory burdens but rather market failure on the sell side. It's because their business models are oriented to gaining a return on infrastructure capital investment over time frames far shorter than what's needed given the high costs -- mostly labor -- of deploying that infrastructure. It is this economic consideration that stifles investment in last mile Internet infrastructure in the United States, not regulation.

Saturday, August 09, 2014

Roanoke Valley Broadband Authority, Internet providers disagree on state of broadband - Roanoke Times: Roanoke News

Roanoke Valley Broadband Authority, Internet providers disagree on state of broadband - Roanoke Times: Roanoke News

This story illustrates that local government officials are growing less inclined to rely on what incumbent legacy telephone and cable companies claim they have in place or will build when it comes to last mile Internet infrastructure.

Good for them. Other local governments should follow this example and create and fund special districts and authorities to build the infrastructure they need. Collectively, their doing so will help the United States bridge a persistent Internet infrastructure gap that leaves millions of Americans without adequate connectivity -- many of them still forced to use dialup connections that were state of the art 20 years ago.

Wednesday, July 30, 2014

Study finds Missoula city, county should spend $10M on broadband

Study finds Missoula city, county should spend $10M on broadband: As proposed, the city and county together would invest $10 million toward a $17 million system, with the local government funds leveraging other money, Copple said. The local money would be paid through user fees, not taxes, and it would build roughly 60 miles of an “open access” fiber-optic network.

“Everybody has the same chance to use it,” Copple said.

Thursday, July 24, 2014

Open access fiber telecom infrastructure funded in West Virginia over telco's objections

Competition in telecommunications infrastructure isn't really among major telcos and cable companies. They operate in market that due to the high cost barriers to entry functions as a natural monopoly or at best, a duopoly where service is available from just two landline providers: the phone company or the cable company. In much of the United States, the choice is only one of the two or sadly, neither because they have redlined parts of their service territories.

The real competition is between the business models for premise Internet connectivity: open access Internet infrastructure such as employed by the Utah Telecommunication Open Infrastructure Agency that regards it as a public asset like roads and highways or the proprietary, closed access infrastructure model of investor-owned telephone and cable companies.

This week in West Virginia, the debate tipped in favor of open access after the West Virginia Broadband Deployment Council voted 3-2 to provide $690,000 in funding to the West Virginia Network, a state agency that provides Internet service to schools, universities and other public facilities. The deciding factor was the state wanting more control over the infrastructure and not being subject to the whims of a monopoly provider.

“Frontier is the only provider in the region, and there is no open access to that infrastructure,” one of the council members noted. “You can’t really connect any of the dots [communities] together . . . . We can hopefully connect those rings and enable broadband expansion in the area.”

This is a notable development because it signals that public policy is shifting towards viewing Internet infrastructure as essential as public thoroughfares and thus not best controlled and operated as a hodge podge of private toll roads with high tolls and serving only some areas while leaving others disconnected from the Internet.

Click here for the story.


West Virginia Broadband Deployment Council voted 3-2 to award the money to West Virginia Network, or WVNET, a state agency that provides Internet service to schools, universities and other public facilities. WVNET would own the three-segment fiber network that would connect Snowshoe to Cass, Valley Head to Mill Creek, and Durbin to Green Bank. - See more at: http://www.wvgazette.com/article/20140723/GZ01/140729675/1419#sthash.ldOg1t7a.dpuf
West Virginia Broadband Deployment Council voted 3-2 to award the money to West Virginia Network, or WVNET, a state agency that provides Internet service to schools, universities and other public facilities. WVNET would own the three-segment fiber network that would connect Snowshoe to Cass, Valley Head to Mill Creek, and Durbin to Green Bank. - See more at: http://www.wvgazette.com/article/20140723/GZ01/140729675/1419#sthash.ldOg1t7a.dpuf
West Virginia Broadband Deployment Council voted 3-2 to award the money to West Virginia Network, or WVNET, a state agency that provides Internet service to schools, universities and other public facilities. WVNET would own the three-segment fiber network that would connect Snowshoe to Cass, Valley Head to Mill Creek, and Durbin to Green Bank. - See more at: http://www.wvgazette.com/article/20140723/GZ01/140729675/1419#sthash.ldOg1t7a.dpuf
West Virginia Broadband Deployment Council voted 3-2 to award the money to West Virginia Network, or WVNET, a state agency that provides Internet service to schools, universities and other public facilities. WVNET would own the three-segment fiber network that would connect Snowshoe to Cass, Valley Head to Mill Creek, and Durbin to Green Bank. - See more at: http://www.wvgazette.com/article/20140723/GZ01/140729675/1419#sthash.ldOg1t7a.dpuf

Sunday, July 20, 2014

The FCC wants to let cities build their own broadband. House Republicans disagree. - Vox

The FCC wants to let cities build their own broadband. House Republicans disagree. - Vox

House Republicans would be well advised to consult with or hire new, better economic advisors. They demonstrate a fundamental misapprehension of telecommunications infrastructure in viewing it as a competitive market when in fact it's a natural monopoly as Susan Crawford notes (see the embedded video interview).

Competitive markets (like retail grocery and furniture, for example) have many sellers and many buyers. Due to high cost barriers to entry, telecommunications infrastructure is not characterized by a multitude of sellers and will likely never be a competitive market. It's a utility like telephone service as Crawford correctly observes, and should be regulated accordingly.

Tuesday, July 15, 2014

Another public regional telecom infrastructure project may be ripe for PPP investment


In Utah, several cities are moving ahead with due diligence on a public-private partnership (PPP) to construct fiber to the premise (FTTP) telecom infrastructure.

Another public FTTP infrastructure project in the eastern United States might also be an attractive partner for private investment companies like Australia-based Macquarie Capital Group, which is looking at investing in Utah's UTOPIA regional network.

This one's in western Massachusetts and is a utility cooperative of 42 municipalities. According to a June 2014 update by the Wired West cooperative, it is hoping to obtain state funding to move forward with construction as people in western Massachusetts continue to be vexed by the lack of adequate internet service.

Given the scope of the Wired West project, it will likely need significantly greater funding from the private sector as part of a PPP like that under consideration in Utah.

Sunday, July 13, 2014

Gigabit over fiber altering telecom landscape

One word is exerting substantial influence over U.S. telecommunications infrastructure: gigabit. It is literally exponentially redefining what constitutes modern premises Internet connectivity from megabits per second (Mbs) range to gigabits per second (Gbs).
Gigabit is also aiding in a shift of market power to the demand side and away from legacy telephone and cable companies that supply megabit class service – often in the single digit Mbs range in the case of telcos -- over “last mile” metal wire or cable to customer premises. The limitations of this infrastructure versus fiber optic lines easily capable of delivering gigabit class service have led telephone and cable companies to ration bandwidth, selling it in various “bandwidth by the bucket” consumption tiers like electricity or natural gas. Gigabit over fiber to the premise will obsolete that business model at the same time premise bandwidth demand continues to rapidly accelerate. Drivers include more and higher definition video streams, multiple devices in the home and emerging Internet-enabled home services. Also, more knowledge workers working at home at least part of the work week.

Gigabit over premises fiber service will also obsolete the legacy telcos and cablecos themselves, burdened with decades-old twisted pair and coax cable infrastructure as well as high debt and shareholder dividend obligations. That will give advantage to agile and financially creative overbuilders connecting premises with fiber service.

Since telecommunications infrastructure is a natural monopoly, once fiber first movers have established a substantial market presence, the economics of challenging them with parallel infrastructure become very difficult. This same dynamic has historically deterred those who would overbuild the incumbent telcos and cablecos. But the shift toward gigabit class service delivered over fiber could reverse the advantage of incumbency the legacy telephone and cable companies have enjoyed for more than a decade.

The first mover advantage would be particularly high in parts of incumbent telephone and cable company service territories where the incumbents don’t offer landline Internet connections or very slow ones such as first generation DSL.