Tag Archives: BTOP

People matter, not paperwork, for rural broadband development

by Steve Blum • , , , , ,

The success of broadband subsidies targeted to rural areas should be evaluated, at least in part, on the number of rural subscribers projects actually attract. That’s one of the conclusions of an investigation by the federal government accounting office into $3 billion worth of grants and loans given out by the U.S. department of agriculture’s rural utilities service as part of the 2009 stimulus program.

There were two pots of broadband stimulus money back then: the NTIA’s broadband technology opportunities program and RUS’s broadband initiatives program (BIP). I worked on grant applications to both programs, and found RUS’s standards a little more exacting – they had been in the rural broadband subsidy business for several years by that time. NTIA was completely new at the game.

But after looking at the results in 2012, the GAO said the subscribership data that RUS was using to track project success was “innaccurate” and a year later found it hadn’t improved substantially. Part of the problem is that the projects served more than just rural communities…

RUS does not track subscribership by rural area and, as a result, is not able to show the impact of the BIP program on rural broadband availability. The [inspector general] previously found that RUS’s performance information makes it impossible to measure BIP’s impact in rural areas because the information was not collected by rural area. The Recovery Act required that BIP service areas be at least 75 percent rural without sufficient access to high-speed broadband service to facilitate economic development. The rest of the project area may not be rural.

GAO is recommending formal, annual reports of how well, or not, RUS-funded projects meet the goals set out for them. That’s a good idea. The wrangling over the latest update of farm support programs consumed a lot of time and effort, but expanded rural broadband development resources. More rigorous oversight will help ensure that the money goes toward serving the people who need it most.

Eastern California lights up in July

“We have started and we will finish,” said Michael Ort, CEO of Praxis Associates, the company behind the Digital 395 project. “There have been people who have bet against us and that’s a great motivator. It’s going to happen.”

The ambitious, ARRA-funded network will connect Reno to Barstow, in the California desert east of Los Angeles, installing nearly 600 miles of fiber optic cable. Most of the path runs along U.S. 395, down the eastern side of the Sierra Nevada through towns like Carson City, Mammoth Lakes, Bishop and Ridgecrest.

Ort was giving an update to a community broadband forum organized by the Eastern Sierra Connect Regional Broadband Consortium in the Mono County town of June Lake. The project has been enthusiastically backed by local officials and residents, who are eager to bring high capacity, carrier grade middle mile connectivity to the many small towns, communities and tribal lands along the way.

But even with that support, bureaucracy has been the biggest obstacle he’s had to overcome.

Problems have come from the maze of state and federal regulators that have a say. Twenty three different agencies had to sign off on project plans, a process that ate up two years of what is supposed to be a three year project.

“You have to satisfy everyone’s needs,” said Ort, explaining that different agencies have different objectives and sometimes exactly opposite requirements. For example, one might ban digging along a certain route while another might require it. Approvals included individual permits for almost 400 archeological and historic sites.

Even so, Ort is confident they’ll have it operating by the 31 July 2013 deadline.

The project is funded by $81 million from the 2009 federal stimulus bill, under the Broadband Technology Opportunities Program (BTOP), and $19 million from the California Advanced Services Fund. Ort says construction costs – not counting permits and environmental clearances – are running between $25 and $30 a foot. Much of the work involves plowing fiber into the ground, but some it requires using a giant “rock wheel” to chop through granite.

When it’s done, Digital 395 will dramatically reduce the cost of connecting to the Internet from eastern California. Customers will primarily be major institutional users, Internet service providers and other resellers, and major telecommunications companies that want a back-up path between northern and southern California. Plans are to price 100 Mbps of Internet connectivity in the one thousand dollar range, using traditional T-1 lines as both pricing and quality of service benchmarks. Simple Layer 2 – Ethernet – connectivity would be about $6 per Mbps, under current plans. Pricing would be flat and not dependent on distance.

The broadband stimulus pool is nearly dry

by Steve Blum • , , , , , , , ,

BTOP might have $442 million in the kitty, although almost certainly not. Or $257 million or $15 million or zilch. For BIP, I can’t even estimate what’s left, but my best guess is that money is already gone.

First, I want to give credit where credit is due. Fred Dyste, via his Digital West blog, has been the gold standard for tracking BTOP (Broadband Technology Opportunities Program) and BIP (Broadband Initiatives Program) stimulus grant applications and awards. He’s been delivering invaluable tracking and analysis of who’s asking for money and who’s getting it. Most of the numbers I’m using were provided by Fred.

Yesterday, about $482 million in BTOP grants were announced for several states, along with about $518 million in BIP grants and loans. On Friday, Hawaii received $35.9 million in BTOP grants. Adding those numbers to Fred’s tabulations for the first round of BTOP/BIP grants and his running total to 28 August 2010, the total for infrastructure, public computer center (PCC) and sustainable broadband adoption (SBA) BTOP awards is $3,791 million and $3,516 million for BIP awards (allowing for the cancellation of a $19 million BIP award).

The stimulus program originally gave $4,700 million to BTOP and $2,500 million to BIP. On that measure, the Rural Utilities Service (RUS) has given out an extra $1 billion. They can do that because they are giving out grants and loans, which are accounted for differently, and they have some separate funding for the loan program. Do they have any more money to give out? Maybe.

The National Telecommunications and Information Administration (NTIA) is only making grants and they don’t have deep pockets of their own to dip into. In fact, Congress has already pulled back $302 million from the program to spend on other things, leaving only $4,398 million. The math looks like this:

CategoryGrants Made
Infrastructure$3,506 million
Public computer center$127 million
Sustainable broadband adoption$159 million
Mapping allocation$350 million
Total$4,141 millon

So in theory, that leaves $271 million to spend on non-mapping BTOP projects. Fred has tracked $179 million in mapping grants, so maybe there’s another $171 million available, bringing the total to $442 million. But I doubt it. NTIA is saving a big chunk of the mapping work for itself, and there’s no reason to think they’ll let go of that money.

In fact, NTIA might be keeping a bit more for itself. Originally, NTIA said that only $2,600 million was available for infrastructure, PCC and SBA grants in the second round, and they’ve given out $2,585 million. On that basis, there’s only $15 million left, without even counting the $302 million Congress clawed back.

The original stimulus bill set aside $200 million for PCC grants and $250 million for SBA grants. However, in cutting money, Congress didn’t specify how to spread the cuts around. If you figure things, like Fred does, on a pro rata basis, that means subtracting $47 and $29 million from the PCC and SBA categories respectively, so those programs are tapped out. But NTIA has considerable discretion when it comes to running BTOP, so maybe not. If there is any remaining money, it could go anywhere.

Building community broadband: three things that work without stimulus grants

The California Emerging Technology Fund (CETF) has funded several regional broadband consortia in northern and central California. At its third annual Rural Connections workshop in Redding this week, representatives from six groups presented the results of their work over the past couple of years. Two, covering California’s Gold Country and Redwood Coast, stood out as having made genuine progress toward bringing Silicon Valley-grade Internet service to areas that are otherwise off the broadband map.

Gold Country Connect's interactive web tool
 Gold Country Connect provides prospective investors
 with broadband planning tools
Brent Smith, CEO of Sierra Economic Development Corporation, and Connie Stewart from Humboldt State University had success stories to tell. Three key lessons stood out:

1. Seek out motivated investors, including competitive local exchange carriers and independent Internet service providers, and find ways to improve their business cases and nudge them towards your goals. Don’t waste everyone’s time trying to bribe or bully them into accepting your plans or implementing your programs. A patchwork of operating networks beats a pristine concept with no takers, every time.

2. Do your homework and make sure it’s A-grade. Simple, quantitative market research that identifies market gaps and charts statistically valid demand at defined price points is pure gold to private sector investments analysts. A centralized broadband mapping project with service provider buy-in, like that run by Chico State University, puts the cards face up on the table and lets everyone get down to business without posturing and poor mouthing.

3. Subsidies help, but don’t necessarily need to be large. A guaranteed loan, a little local capital, even a tax break can tip the balance for a potential private sector broadband investor. When bigger subsidies are needed, the lion’s share of the risk can still fall on private investors. The California Advanced Services Fund will do a 40% match against private capital in underserved areas, and that’s been enough for hundreds of kilometers of fiber.

Unified community support is important, and creates a level of comfort that the project can be implemented. Leadership is needed to gain rights of way, permits and variances, and overcome bureaucratic inertia. Business analysts are more impressed by political muscle and professional, statistically valid research than they are by crayon drawings from a third grade class.

Real progress in other CETF-sponsored consortia has been hampered by a focus on community feel-good exercises and unworldly research. Evidently, Chico State’s mapping expertise is not matched by its economics department: someone there seems to think you can do a demand aggregation study without asking tiresome questions about price elasticity. The good thing about this kind of conference is that public sector decision makers get to see what works and what doesn’t, and can respond appropriately.

The last item on the conference agenda was the decision to come back for a fourth year. Expect to see a longer list of success stories.

The stimulus was fun while it lasted, now back to work

It’s time to look past the stimulus program, and re-adjust community broadband planning assumptions. The National Telecommunications and Information Administration’s (NTIA) Broadband Technology Opportunities Program (BTOP) and the Rural Utilities Service’s (RUS) Broadband Initiatives Program (BIP) encouraged local groups to roll themselves up into regional alliances and propose magnificent projects that would meet any conceivable need and serve every user imaginable.

It made sense, because that’s where the money was. NTIA and RUS made some dreams real in the first round last year, and are on track to fulfill a few more fantasies in the second round. But even though BTOP is reopening for what amounts to a stunted, public-safety focused third round, the good times are over and we have to return to the old normal.

It’s a world where the free money is mostly gone. Once the BTOP money is spent, NTIA goes back to being a small agency running small programs. In rural areas, RUS and state programs, like the California Advanced Services Fund (CASF), will provide grants and loans to organizations with a qualifying track record and, in some cases, enough cash to fund half or more of proposed projects themselves.

first round BIP funding funnel
 Adelstein and RUS general
 field representative Harry Hutson showed
 CETF conference attendees in Redding
 how the first round BIP money went
 down the spout
RUS won’t fund projects that compete with their existing loan portfolio, however. Speaking to the California Emerging Technologies Fund’s third annual Rural Connections workshop in Redding this week, RUS administrator Jonathan Adelstein made it clear that the agency will give priority to organizations that it already funds, and won’t subsidize competing projects.

CASF expects it will continue to fund new broadband projects in California, but only in areas where AT&T, Verizon and the cable companies fail to upgrade infrastructure. A few arguable urban pockets aside, it’s the remote rural regions that have a shot.

Elsewhere, community broadband advocates will have to go back to the basics. Tried and true economic development strategies, like public-private partnerships, tax breaks and other incentives, and old fashioned salesmanship, will be effective. But only where public agencies and community advocates can present a focused and well documented business case and be flexible enough to accept that private capital comes with its own priorities.

The old normal is a world where subscriber metrics, return on investment and anchor tenants trump grand visions, sad stories and political grease. Painstaking determination and hard work count again, though. That’s a world worth calling home.

Best Practices Highlight Wireless Broadband Feasibility Study for the City of Oakland

Download the Oakland Wireless Feasibility Study

Like nearly every government agency in California, the City of Oakland was faced with increasing demand for public services and a decreasing budget. An evaluation was needed of the potential for wireless technology to make municipal staff more efficient and allow them to stay in the field longer, and to provide Internet service to residents, either directly in their homes and businesses or indirectly through community anchor institutions. This evaluation needed to focus specifically on Oakland’s diverse population, needs and terrain.

The City’s goals were:

  • Enhance economic development by enabling businesses to operate more effectively and by making Oakland a more attractive place to live, work, and visit.
  • Improve public safety by putting more police officers, fire fighters, inspectors and public works staff into the field, keeping them there longer and letting them work more efficiently.
  • Increase the effectiveness of public, private, and nonprofit organizations through improved access to state of the art broadband wireless technology.
  • Help overcome the digital divide.
  • Improve the quality of life for all Oaklanders.

Tellus Venture Associates was brought in to do a comprehensive feasibility study that would include public focus groups, workshops and a town hall meeting, close coordination with City departments and outside agencies, and a technical survey that included radio frequency modeling over the hills, canyons, flatlands and waterways within the city limits.

When we analysed the research data, the trends that emerged tracked closely with the best practices we’ve developed during seven years of municipal and community broadband experience. The result was a more refined list of those principles:

  1. No matter what the manufacturer says, the laws of physics still apply. No matter what the special interests say, sound business principles still apply. Don’t underestimate the public’s appreciation of physics and sound business principles, or overestimate its regard for manufacturers and special interests.
  2. City-owned and operated metropolitan area networks are a cost effective means of extending information technology infrastructure and resources to local government facilities and employees.
  3. Providing broadband connectivity to targeted community anchor institutions can be financially and technically feasible for cities, and is supported by public opinion.
  4. Providing universal, consumer-grade wireless Interet access is not financially or technically feasible for cities, and is not supported by public opinion.
  5. Cities can better promote digital inclusion by enabling and supporting a competitive broadband environment.
  6. Widespread public awareness and support precedes deployment of a successful municipal broadband system.
  7. Fiber optic and wireless technologies can be effective choices for network backbone segments, depending on capital and operating cost, timing, right-of-way, capacity and other considerations.
  8. Fiber optic and other landline technologies provide orders of magnitude more bandwidth and many more years of useful service life, with lower operating costs.
  9. Wireless technologies can be deployed faster and at much lower capital expense, and provide greater flexibility to change network topologies and service models to meet future needs.
  10. Wireless technologies have the unique ability to support municipal staff in the field, particularly public safety personnel, but should only be deployed after an independent evaluation of technology, terrain and available spectrum.

With these principles in mind, we assessed the Oakland public’s needs and priorities, designed a reference architecture that could meet those needs, and developed a business model that quantified the benefits, demonstrated the value proposition and identified the money to pay for it all.

Our findings were:

  • A point-to-point wireless broadband system serving specific community and institutional needs is financially and technically sustainable for the City of Oakland.
  • The cost of building and operating such a system can be met through identifiable cost savings, efficiency gains and budgetary choices based on the economic value of the benefits produced.
  • Public Internet access by way of community anchor institutions is financially and technically feasible, and universally supported by a diverse range of Oakland residents, organizations, agencies and businesses if it is implemented in a fiscally sound manner.
  • Enabling entrepreneurial opportunities for local businesses on a pay-as-you-go, public-private partnership basis is also backed by Oakland stakeholders and supported by the financial and technical analysis conducted for this study.
  • Providing wireless Internet service to residences or individual consumers is not financially sustainable or technically feasible for the City of Oakland, and is opposed by nearly all stakeholders, who cite the widespread technical and financial failure of such systems in other cities.

The next step was to secure the funding. Some of it came from the cost savings created by replacing a large number of low capacity, leased land lines with a comprehensive wireless backbone, comprised of high capacity point-to-point links using licensed spectrum. Some of it came from money budgeted for expensive cellular data service. In other cases, savings in man-hours and increased productivity, including more and better field audits by tax officers, offset operating costs.

Finding the money to pay for the capital expense was a different problem. Bonds were not an option, given the uncertainty of future budgets. Some of the funding could be raised locally, through public-private partnerships, but not all of it.

Fortunately, the conclusion of the study coincided with the establishment of the American Recovery and Reinvestment Act, also known as the federal economic stimulus program, which included $7.2 billion for broadband projects. With its emphasis on public safety, community anchor institutions and economic development, the broadband infrastructure plan created by Tellus Venture Associates for the City of Oakland was ideally suited to meet the program’s requirements.

The stimulus grant application had to wait until the second round of funding, because the first round emphasized rural projects and all but excluded urban areas from eligibility for broadband infrastructure funding. At the same time Google announced its own broadband grant program, which likewise tracked with the best practices we incorporated into the study. Both applications are now pending.

The final step will be to move ahead with construction of the system. Tellus Venture Associates prepared a draft Request for Proposal, which sets out the specifications for a municipal broadband system that would serve the City of Oakland. In some cases, such as providing broadband connectivity to public safety personnel in the field, the technology that would be employed is necessarily wireless. But in other cases, for example the core network backbone, wireless, fiber optic or other technologies are all possibilities. Those determinations, as well as any decision to release an RFP, will be made by City staff, once funding is secured.

Oakland Wireless Feasibility Study

Printable, high resolution version

City of Oakland staff report

Study presentation to Oakland City Council

City of Oakland wireless reference architecture

Oakland townhall meeting presentation

Follow the money, from the first to the second round of broadband stimulus grants

by Steve Blum • , , , , , , , , , ,

More than a thousand first round hopefuls are still staring into the black hole that swallowed their applications. The second round notifications of funding availability (NOFAs) issued by the Rural Utilities Service (RUS) and National Telecommunications and Information Administration (NTIA) for the broadband stimulus program do not explicitly address the status of first round applications.

The stimulus bill gave RUS $2.5 billion and NTIA $4.7 billion for broadband project funding. In the first round, RUS said it would give out up to $2.4 billion. Now its saying it’ll give out a total $2.2 billion in the second round. The target budget is:

CategorySecond Round
Last mile projects$1.7 billion
Middle mile projects$300 million
Satellite projects$100 million
Libraries, tech assist$5 million
Reserve$95 million
Total$2.2 billion

That leaves $300 million, which presumably goes to first round grants and, presumably, overhead. So far, RUS has only announced $54 million in first round grants. It still has first round applications in the due diligence stage of review, so any applicant that’s made it that far has a plausible hope of winning funding. The lion’s share of RUS’s money is shifting to the second round, so if you haven’t heard back about first round review yet, I suggest you start thinking about round two.

Unless you also put in a joint bid to NTIA. Including broadband mapping grants, NTIA allocated nearly $2 billion to first round projects. It’s allocating a total of $2.6 billion for the second round:

CategoryTotal TargetedFirst RoundSecond Round
Infrastructure$3.55 billion$1.2 billion$2.35 billion
Public computer centers$200 million$50 million$150 million
Sustainable adoption$250 million$150 million$100 million
Mapping$350 million$350 million-0-
Reserve$200 million$200 million-0-
Total$4.55 billion$1.95 billion$2.6 billion

The two NTIA rounds match up pretty closely with the targeted totals. There’s $150 million unaccounted for, but that’s a believable overhead number for a federal operation.

The inference is that the two rounds will be processed, considered and funded separately. As it lays out now, if you have a first round NTIA application that’s disappeared into the process, it’s possible that you might yet advance to the due diligence stage. But that possibility diminishes as time goes on, particularly if NTIA sticks to its end-of-February target for closing out the first round and its 30-day due diligence period.

The second round workshops start next week, and more information should be available by then. My advice to first round applicants who haven’t heard from NTIA yet is to spend this week beginning to form the community alliances that it advocates so enthusiastically. It won’t be wasted effort, even if you slide into the first round under the wire.

Broadband stimulus grant update: first round still under review, second round likely to slip a bit

by Steve Blum • , , , , , , , , , , ,

Anna Gomez, deputy assistant secretary for communications and information at NTIA (National Telecommunications and Information Administration), spoke at today’s Tech Policy Summit at the Consumer Electronics Show in Las Vegas.

Anna Gomez, NTIA
 Secretary Gomez speaks to reporters
 at 2010 Consumer Electronics Show
She repeated previous agency comments about wanting to “get it done fast, get it done right and with the greatest effect possible.”

She described the Broadband Technology Opportunities Program (BTOP) as “unprecedented” at the NTIA.

Lessons learned in a difficult first round would be applied in the second round. Among those lessons is a better understanding of what sort of projects should take priority for BTOP funding.

Her comments regarding the program’s time line were:

  • The notice of funding availability (NOFA) for the second round will be released in a “few weeks”. She wouldn’t say if that means the previous target of mid-January would slip, although she left room for thinking it will.
  • The first round grants will be completed “on a rolling basis over the next two months.”
  • All grants will be made by Congress’ mandated deadline of 30 September 2010.
  • In separate comments, Karen Jackson from the Commonwealth of Virginia’s Technology Office, confirmed that there will be at least a 60 day window for second round applications, rather than the original 45 day deadline in the first round.

I spoke with Secretary Gomez afterward about some of the nuances of the application review time line and progress to date. She couldn’t provide much else in the way of details, although the inference was that the first review stage for the first round BTOP (broadband technology opportunities program) applications is still ongoing, and that not all of the projects that will advance to the second, due diligence stage of review have been selected.

She did say “our goal is to make sure people know their status in time to file in the second round.” Asked whether first round applicants could be in the position of having to simultaneously prepare a second round application and follow up on a first round application, she said “hopefully not.”

Connecting the dots, here’s my take:

  • The second round NOFA will be released around the end of January, maybe even as late as the first or second week of February.
  • If a first round application hasn’t advanced to the second stage of review by the end of the month, it won’t.
  • The second round NOFA will be more specific about program goals, be structured to encourage cooperation amongst applicants, and favor projects that include significant, shared middle mile infrastructure, with or without last mile facilities.
  • NTIA has a much better understanding now of how to run the program and what its goals should be. Don’t be surprised if the first round falls significantly short of its $4 billion target, with unspent funds redirected to specific program goals in the second round.

Secretary Gomez also announced a new program, available at match.broadbandusa.gov, called Broadband Match. It’s an online tool that is supposed to “facilitate partnerships among prospective applicants for a grant.” She said the idea is to further NTIA head Larry Strickling’s goal for the next round of favoring public/private partnerships that take a “comprehensive view” of communities.

She said that they want to ensure that key community members – meaning anchor institutions and government agencies – can access middle mile projects directly and that private companies can make use of it to create last mile services that reach consumers and businesses.

The emphasis in the second round will clearly be on middle mile projects. Gomez spotlighted the grant made to such a project in Georgia last month as an excellent example of what they’ll be looking for in the second round. The objective of the Broadband Match program is to ensure that public/private groups “can put together the most comprehensive application possible.”

First dribble of broadband stimulus funding announced

by Steve Blum • , , , , , , ,

The feds today announced they were giving $182.7 million of broadband stimulus money to 18 projects scattered across the U.S. (but nothing so far for California). 18 projects funded out of 2,200 applications, representing less than 3% of the $7.2 billion allocated.

Not much detail but a few worrisome hints.

The infrastructure grants announced today all appear to be for RUS/BIP-type projects. Even the ones that were funded through NTIA/BTOP. That’s consistent with what we heard back in September: a select few RUS projects were fastracked into the second stage of review.

RUS is going down a familiar path – giving money to rural clients. Unlike NTIA, RUS has the staff and experience to do this work, they didn’t have to start from scratch. Even so, it took four months to process a handful of grants.

From the Associated Press:

The administration plans to award a total of $2 billion in grants and loans on a rolling basis over the next 75 days as it starts doling out the first round of stimulus funding for broadband.

Nice, but the first round was supposed to total $4 billion. Are they cutting the first round in half? Dragging it out past the end of February? Or did someone get the number wrong? Let’s hope it’s a typo. $2 billion is about what RUS was supposed to give out. Maybe they’re only referring to NTIA. Or maybe only RUS has its act sufficiently together to get anything done in the next two or three months.

From StimulatingBroadband.com:

NTIA head Lawrence Strickling “yesterday stated that “300 to 400″ project applications for broadband stimulus funding are now being reviewed…in the due diligence phase.”

What’s not clear is whether the other 1,800 or 1,900 applications are still in the queue, or have been rejected. If 1,800 apps are still sitting in someone’s in-box, we’re in for a long wait. If some or all have already been rejected, we need to know.

NTIA and RUS also just posted the comments they received regarding Round 2. It’s a lot of reading.

This process might take a lot longer than anyone ever thought.

CPUC Approves $5 Million for central California coast broadband project

by Steve Blum • , , , , , , , ,

The California Public Utilities Commission (CPUC) unanimously approved a $4,975,009 grant from the California Advanced Services Fund (CASF) to the Central Coast Broadband Consortium (CCBC) on Friday, 20 November 2009. The grant pays for 10% of the approximately $50 million fiber optic trunk line network planned for Santa Cruz, Monterey and San Benito counties on California’s central coast.

CCBC system map

CCBC’s CASF and associated federal stimulus grant applications are managed by Tellus Venture Associates, which also does the financial planning and budgeting for the project. In August, the CCBC submitted a proposal for a $40 million grant to the National Telecommunications and Information Administration’s (NTIA) Broadband Technology Opportunities Program (BTOP) and the Rural Utility Service’s (RUS) Broadband Initiatives Program. The remaining $5 million has already been committed by consortium members.

CPUC’s approval follows endorsements by California Governor Arnold Schwarzenegger and congressman Sam Farr, who represents the three county region. NTIA and RUS are reviewing the grant proposal, with a decision expected next month.

The project would create a 428 mile fiber optic backbone linking unserved and underserved areas to better served communities, and connecting the entire region to Tier 1 Internet facilities in Silicon Valley. Using a loop architecture, any point on the network would have two independent paths to any other point, and to the Internet.

Current plans are for the system to be operated by a cooperative, which will offer access on a wholesale basis to last-mile Internet service providers and major institutional customers.