The Human Cloud

As Mathew recently said, the human cloud is the capability of a company to get the best talent available regardless of location. With the advent of mobile technology, “flashmobs” and bee “swarms” to accomplish a problem will be more commonplace. An example of this would be responding to a PR crisis: Instead of taking the time to convert a conference room to a “physical war room,” a marketing executive can connect with the resources inside and outside the company (executive team, bloggers, call center, media consultants) to create a “swarm” to extinguish a PR spark before it becomes a media firestorm.

The human cloud is also a catalyst for companies to generate maximum value from existing resources. The Deloitte Center for the Edge recently published the 2010 Shift Index, which rigorously documents a long-term performance deterioration of 75 percent in return-on-assets (ROA) across 15 industries in the U.S. since 1970, despite productivity gains. According to the Shift Index, “the solution lies in empowering passionate employees, those who feel truly engaged with their work and constantly push the performance envelope, by accelerating institutional innovation and driving corporate growth.”

Empowered Employees

Collaborative cultures are now embracing the passion of the whole employee, creating work environments for them to use their devices, home office and professional networks to support the context for creating competitive advantage for the company.

Empowering workers in the human cloud, giving them a choice of devices and leveraging social networks for collaboration is necessary, but not sufficient enough to create truly high-impact collaboration. Different levels of “generational trust” between seniors, boomers, Gen Y and the Millennial Generation create a “digital divide” that can pose a significant barrier to collaboration. Research conducted by David Coleman, Managing Director of Collaborative Strategies, finds the biggest barriers to collaboration are not technology problems but are related to people and process. He has created a framework that illustrates different levels of competence, trust, age and web:

Source: Collaborative Strategies

Companies can act on these findings and create initiatives and social metrics designed to accelerate the progression of digital exiles, digital immigrants and digital natives to digital workers.

On-Demand Infrastructure and Applications

Businesses can now leverage on-demand infrastructure and applications to support nearly every aspect of their operations. Service providers can securely host multiple tenants of one application on one server, which boosts their server utilization rates, cuts administrative costs and lowers server cost per tenant. The resulting efficiencies and cost savings can be passed through to customers using a mesh business model.

At the same time, corporate IT departments are freed from the hassles of traditional deployments, software maintenance and upgrades, and concerns about scalability and reliability. IT can help business units source innovative applications when they need them and avoid getting bogged down with day-to-day maintenance and support. This frees IT up to focus on more value-added activities while being more responsive to changing business needs.

In many cases, these on-demand solutions can be integrated with existing enterprise IT investments. For example, social-media oriented, on-demand collaboration tools can be integrated with Microsoft SharePoint (which offers less-than-optimal collaboration functionality) for an improved user experience that makes employees and external partners more productive, effective and inspired to innovate to establish competitive advantage.

One example of a company turning these ideas into applicable business solutions is Intuit, who developed an innovation management tool called Brainstorm. To read a case study of this process, check out the full report.

Image source: flickr user ChrisL_AK

Related Content from GigaOM Pro (subscription required)

• Rogue Devices: The Consumer Influence on Enterprise Mobility, Part 1
• Transient Apps: The Consumer Influence on Enterprise Mobility, Part 2
• How to Manage Consumer-Grade Collaborative Tools in the Workplace

So which equipment maker is the big winner in the LTE sweepstakes? The 2G wireless hardware market was dominated by Motorola, Ericsson and Nokia, collectively called M.E.N. of the telecom world. The rise of 3G opened up opportunities for companies like Nortel, Alcatel-Lucent and Siemens. The next big wireless shift is going to be dominated by a handful of players as well.

This time there is a new sheriff in town; Chinese vendor Huawei is one of the early leaders in the LTE sweepstakes. The company currently accounts for 36 percent of the contracts for LTE equipment, according to Telegeography, a market research firm. In terms of sheer size (and dollars), however, Ericsson is the big player as it has been supplying gear to some of the largest LTE-rollouts in the world.

Most of Ericsson’s deals are in the more mature markets, while Huawei is going for high growth markets such as China. Ericsson has managed to snag 18 contracts, including deals with TeliaSonera, Vodafone Germany and China Mobile. However, Chinese operators aren’t likely to roll out their networks before 2012. Nokia Siemens and Alcatel-Lucent, with 16 percent and 14 percent of the total market (by number of contracts signed.)

Here are some facts about LTE you might be interested in:

• The total LTE infrastructure market will top $5 billion in 2013. (Infonetics Research)
• The LTE infrastructure market is forecast to reach $11.4 billion by 2014. (Infonetics Research)
• LTE subscribers could exceed 72 million by 2013. (Infonetics Research)
• LTE subscribers could exceed 153 million by 2014. (Infonetics Research)
• TeliaSonera says the average LTE user so far is consuming 15 GB of data every month.
• The number of users of next-generation LTE technology in the Asia-Pacific region is forecast to surpass 120 million by 2015. (Wireless Intelligence)
• A total of 132 operators in 56 countries are currently investing in LTE for next-generation networks.

Related GigaOM Pro Content (sub req’d):

• Everybody Hertz: The Looming Spectrum Crisis
• Mobile Broadband Pricing for Profits
• What Happens When Data Friendly Phones Come to Prepaid?

Huawei, the Chinese telecom equipment maker, wants to be the biggest networking equipment maker in the world. Furthermore, the company doesn’t want to be known as a supplier of cut rate products (though it makes a bulk of its revenues from those products).

So it has started hiring a bunch of smart people, including many of them in its North American offices. Today, it announced that  John Roese, formerly CTO of Nortel is now going to work for it as Senior Vice President and General Manager of Huawei North America R&D. In 2009, the company made a splash when it hired British Telecom aka BT CTO Matt Bross as its chief technology officer.

Huawei, in 2009 had $408 million in U.S. sales, but since then has been ramping up its sales and R&D efforts.

Roese, who in the past worked for Broadcom, Enterasys and Cabletron Systems is going to be based in Huawei’s North America R&D headquarters in Santa Clara, California. In my past interactions with Roese, I have found him to be a clear and long-term thinker. Roese’s hiring shows that Huawei has grand ambitions, and is willing to spend liberally on R&D, in sharp contrast with some of the larger equipment makers in the U.S. and Western Europe.

There has been a perceptible decline in venture capital investments in the networking sectors. It was startups who hired the best and brightest. But as the carriers consolidated, so did the number of equipment makers and thus decreasing opportunities for startups, who found it hard to make a breakthrough at large carriers. Venture Capitalists shifted their attentions elsewhere.

Huawei and its smaller counterpart, ZTE Corp., spotted an opportunity to take a lead in developing new networking and infrastructure technologies. Huawei, according to the World Intellectual Property Organization (WIPO), ranked second in terms of patents applications under the WIPO Patent Cooperation Treaty (PCT) in 2009. It had filed for over 42,000 patents at the end of 2009.

Related GigaOM Pro Content (sub req’d):

• Who Will Profit From Broadband Innovation?
• The New Net-Neutrality Debate: What’s the Best Way to Discriminate?
• When It Comes to Pain at the Pipe, Upstream Is the New Downstream

Heroku CEO Byron Sebastian (foreground) speaking at GigaOM's Structure 2010 conference in San Francisco

Heroku, the San Francisco-based Platform-as a-Service provider is all set to hit the magical 100,000 app mark sometime next week. Heroku’s platform is gaining momentum, thanks to increased adoption amongst corporate and mobile developers.

In November 2009, Heroku had about 40,000 apps on its platform. Five months later, in April 2010, the company had 60,000 apps on its platform, and in September 2010, they had about 83,000 apps. Just over a month later, the company is now hitting the 100,000 mark.

Heroku, which has developed a Ruby-on-Rails cloud platform, has seen an upswing in demand for its service from not only the fast-growing mobile app developer community, but also from companies who are creating brand-centric and enterprise applications to tap the social and mobile opportunities, said Byron Sebastian, CEO of the two-year-old company whose investors include Redpoint Venture Partners and Ignition Partners. The growth in apps is also helping boost the company’s overall fortunes.

While there’s no denying that private or hybrid clouds are going to become a dominant part of the IT spend in years to come, for now, I believe cloud adoption is driven primarily by startups and a new class of products that require Internet scale. Heroku’s success in finding new converts to its platform is a sign of that trend.

Related content from GigaOM Pro (sub req’d):

• Defining Internal Cloud Options: From Appistry to VMware
• What’s Being Done About Cloud Lock-in?
• For Open Cloud Computing, Look Inside Your Data Center

Google’s recent push into tablets and mobile, along with offering new search services such as Google Instant, are pushing up the company’s capital expenditures, which are slotted to grow almost 184 percent in 2010 compared to last year. Next year, that amount is going to go even higher. This spending is a good thing, because it allows Google to leverage its inherent advantage: infrastructure.

A few years ago, I noted in a post that infrastructure was Google’s key competitive advantage. It’s what allowed the company to innovate and outpace its rivals. It allowed the company to give us results faster than our broadband connections could offer, making us more subservient to its search in the process. In the end, we all forgot the directories and instead focused on the search-box as the start of our Internet journey. Today, Google is a gigantic, $7.3-billion-in-quarterly-sales business.

One thing Google knows: It needs to keep spending money on this infrastructure in order to stay competitive and current. The company recently introduced Google Instant, a new feature that allows you to get results even as you’re still typing the search term. It’s a service akin to the days when an Intel chip got multimedia extensions.

In many ways, Google Instant demonstrates the evolution of a product in order to keep up with times; today’s faster broadband means that the search results need to come up faster than one could type. More importantly, Google Instant is a search product optimized for a brave new world where the user interface is touch rather than keyboards, and devices aren’t your classic computer, but instead mobile and tablet-like.

One of the reasons Google was able to launch Google Instant is because it can afford to spend a lot of money on its infrastructure. During the third quarter of 2010, the company spent nearly $757 million, the highest amount since the first quarter of 2008, according to investment bank J.P. Morgan. (In comparison, Google spent a total of $810 million on capital expenditures in all of 2009.) In a conference call with Wall Street yesterday, Google VP Jonathan Rosenberg told the analyst community:

From a revenue standpoint, its impact has been very minimal; and from a resource standpoint, it’s actually pretty expensive. So why did we do it? Well, we believe from a user standpoint, Instant is outstanding and the data that we are seeing actually bears this out.

Google’s spending on capital expenditures (mostly on data centers) had been on a decline. That is about to change. According to J.P. Morgan, the company is going to spend $2.3 billion on capital expenses in 2010 versus $810 million last year. For next year, the investment bank  is forecasting $3.2 billion in capital spending.

Some Wall Street analysts are going to view this increased spending and wring their hands. They’re idiots and short-term thinkers. I see the growth in capital expenses as a sign of health, and that things are going well for Google –actually, really well.

Let me explain; until recently, Google had to focus on a small subset of actions to satisfy its end customers – all of us – and thus make money off of advertising. Throw in YouTube videos and Gmail, if you want, but browser-based search and search-based advertising were its bread and butter.

Google is said to be the single biggest source of traffic on many of the world’s networks and that’s with only a handful of offerings. Now imagine how big Google will be as a percentage of the source of Internet traffic once we start taking their new initiatives into account. That also explains why they need to build their own networks and lay their own fiber pipes.

Now the number of consumer interactions has grown multifold. Google’s Android mobile operating system is an Internet-enabled OS peppered with Google services that are used more frequently because we have access to them in our pockets. This overall growth in data center capabilities is only going to go up as the company becomes more successful with its Android push. By spending on data centers and networks, what Google is ensuring is that Google Android will always have a great user experience. Remember, in a world dominated by cloud clients, nothing matters more than instant access to various Internet services.

Related content from GigaOM Pro (subscription req’d) about Google, and its Mobile Efforts:

• Why Google Should Fear the Social Web
• Report: Google’s Voice Possibilities
• How Mobile Cloud Computing Will Change Tech

Questions around these issues were central to the discussion at today’s GigaOM Bunker Series Event (subscription req’d) entitled The Completely Open Source Enterprise. Is there still an appetite to drive innovation in infrastructure through open source? Will the developer community generate significant open-source enterprise applications? Will SaaS — or some other technology — become the innovation driver?

Open source permeates many infrastructure offerings, and is built into many SaaS platforms. It’s also embedded in the infrastructure of next-generation web platforms like Facebook. And many of the largest commercial players — like Apple and Google — don’t just use open source, but also contribute code back to the community.

But enterprises are finding new ways to create business applications where open source isn’t necessarily the innovation driver. Open Health Tools (OHT), for example, is a consortium of companies and health providers developing new tools for patient records. Open source is part of that, but the service delivery end of the stack is its most important and driving element.

The above example indicates that we aren’t headed to a fully open-source enterprise, but, rather, to a number of different scenarios where new partnerships like the OHT or SaaS become significant drivers of innovation alongside open source. John Robb of Zimbra, for example, pointed out at the event that when developers care enough to address critical app problems, they’re developing it as a SaaS offering. That commitment to solving a core innovation problem might make use of open source, but the real commitment is to SaaS, the hot market right now.

An Accenture survey showed low cost of ownership is the least important priority for CIOs. Meanwhile, the economic circumstances of the day allied to the conservative nature of decision-making in operationally focused organizations work against wider adoption of open source, even though the benefits are clear. Joe Topolski of Accenture’s Research Labs pointed to the high level of focus CIOs have on operations. That disposes them against change other than at the margin. “It comes down to risk mitigation. If the apps are running and you’re not getting calls from the business manager saying why won’t this work….” So why risk innovation?

There’s also a question mark over whether IT departments truly understand APIs well enough to go beyond their operational role. James Turnbull of Puppet Labs pitched in with his experience of banks: “Indemnification is a very big issue. Open source fell away because no one can recover the risk.”

Some of the old arguments against open source won’t go away. But the discovery of the day is that the innovation framework is broadening. Open source is still close to its heart, but innovators are also embracing new ways to get things done.

Image Source: Flickr user stuckincustoms

According to a new report from research firm In-Stat, mobile carriers are going to spend nearly $117 billion by 2014 (up 41 percent from 2009 spend of $83 billion) on “last mile backhaul (including line leasing, new equipment-spending, and spectrum acquisitions.”

Photo via Flickr by Flickr user kickfoto, Creative Commons

According to well-regarded wireless analyst (and our friend and Mobilize speaker), Chetan Sharma, during the first half of 2010, average U.S. consumer was consuming 230 MB/month, up nearly 50 percent from the average data consumed in 2009. With nearly 31 percent of U.S. customers using smartphones, the demand for data is going to keep rising, so much so that in March 2010, data surpassed voice traffic on the mobile networks.  In August, Ericsson noted that mobile data grew 10 times faster than voice, and while mobile broadband accounts for only 10 percent of total mobile subscriptions, it’s growing at a rapid clip: about 225,000 terabytes per month.

Wireless last mile backhaul capacity in Western Europe will more than triple between 2010 and 2014, to nearly 60,000 Gbps. The pressure on the networks is only going increase with the launch of LTE-based wireless broadband networks. Nearly half the backhaul capacity in the U.S. is going to be dedicated to LTE, In-Stat projects. never mind the fast growth in the number of HSPA networks

Much of the spending is going to happen by running fiber to the base-stations, especially in the U.S., but microwave is going to see a big jump as well. In-Stat data projects that the sales of “millimeter microwave radios will grow from $159 million in 2009 to $874 million in 2014.”

The network upgrades are causing carriers to rethink their plans, and many of them are responding by ending unlimited data plans. Companies like AT&T have already moved in this direction.

So, who are the likely beneficiaries of this shift? Of course, it will be companies such as Alcatel-Lucent and Ericsson, but there’s a slew of companies providing bandwidth that are going to do well because of this boom. For instance, the cable broadband providers are seeing strong demand from wireless carriers for their business-class broadband connections. Comcast, Cox and Time Warner Cable, in particular, have built strong business networks. The wireless backhaul boom is going to help companies such as Allied Fiber, whose main business is hawking dark fiber.

Related GigaOM Pro research: Metered Mobile Data Is Coming and Here’s How (sub req’d)

Both companies make broadband access equipment and are primarily targeted at carriers (especially rural and independent carriers) that are building optics-based networks and offering triple-play services. Occam’s expertise is in IP and Ethernet, while Calix is particularly strong in fiber access. Calix went public in March of this year, and this is a good indication of company’s plans: Buy and bulk up fast.

Bottomline: As standalone companies, both Occam and Calix were too small and faced competition from much bigger rivals such as ADC Telecommunications. They also had incomplete portfolios. This is a good combination and a great opportunity for these companies to vie for business from some of the larger broadband service providers who typically opt for the Alcatel-Lucents and Huaweis of the world.

Related content from GigaOM Pro (sub req’d): Did We Really Learn Anything From the Dotcom Crash?

The latest investments go to Adaptive Computing, Ciranova, Joyent and Nexant. Nexant, in particular, develops a host of software for the energy industry and other corporate users, including grid operators, utilities and financial institutions. The software enables the companies to manage customer billing and incentive programs, wholesale market pricing and other enterprise infrastructure.

San Francisco-based Nexant, founded in 2000, also works with its customers to buy and sell energy certificates and credits and lists Nth Power, IBM, Morgan Stanley Global Energy Group and Energy Software and Consulting among its other investors.

Intel Capital announced the $200 million Invest in America Fund in February this year, as well as an initiative it spearheaded to bring together 24 VC firms, including Kleiner Perkins Caufield & Byers, Menlo Ventures, Mohr Davidow Ventures, New Enterprise Associates, Draper Fisher Jurvetson, Khosla Ventures and North Bridge Venture Partners, to invest a total of $3.5 billion into startups over the next two years. Back in July, Intel Capital announced five investments into cleantech startups in energy monitoring, smart grid and other areas.

Here is a rundown of the three other Intel investments:

Adaptive Computing: Based in Provo, Utah, and founded in 2001, Adaptive Computing develops automation software to manage supercomputers and data centers, and counts Yahoo, IBM, Los Alamos National Laboratory, Boeing, ExxonMobil and Merck as customers. Intel Capital led the $14 million Series A round, with participation by Tudor Ventures and EPIC Ventures. Adaptive Computing plans to use funding from Intel to boost its cloud computing offerings.

Ciranova: Intel returns to its semiconductor roots by picking Ciranova, which develops software to enable engineers to design communication capabilities — such as Wi-Fi and Bluetooth — into chips for a variety of consumer electronics and communication equipment. Designing analog and mixed signals onto a chip is a tricky task, especially given the increasingly sophisticated consumer devices that can hop on several types of communication networks. Ciranova, based in Santa Clara, Calif., has worked closely with Taiwan Semiconductor Manufacturing Co.

Joyent: Intel is bullish enough about Joyent to invest in it a second time. Joyent runs cloud computing services, an increasing popular approach that offers distributed computing on demand. The San Francisco-based company’s customers include LinkedIn, Gilt Groupe (online purveyor of luxury goods), Kabam! (social game developer) and Context Optional (marketing software). Joyent said Tuesday that Intel’s funding is part of a $15 million Series C round, which also includes money from Greycroft Partners and Liberty Global. Joyent runs data centers in California, Texas, Massachusetts and China.

Image courtesy of stephen.moore.

Hewlett-Packard has agreed to buy security software maker ArcSight for $1.5 billion in cash, as the computer giant tries to expand the range of services it offers corporate clients. Cupertino, Calif.-based ArcSight’s products are used by corporations and government agencies to detect suspicious activity on their networks. The purchase price represents a 24-percent premium to the software company’s trading price before the offer was made.

Analysts said the acquisition of ArcSight is part of HP’s move to offer more value-added services to corporations who are trusting an increasing amount of data to the cloud, whether it’s a cloud the company itself operates or one run by a number of SaaS providers. As VMWare CEO Paul Maritz said recently, this move places an increasing strain on security systems as data flows out of the corporate network and onto a variety of third-party platforms and mobile devices. In HP’s news release about the deal, HP VP Bill Veghte said “the perimeter of today’s enterprise is porous.”

Aaron Rakers, an analyst at Stifel Nicolaus, told Bloomberg that the acquisition of ArcSight will help HP serve the data-center market, which he described as necessary to the company’s growth strategy as it tries to turn its leadership in PCs into a bigger role in corporate computing. Other hardware makers are also investing in security: Intel Corp. recently agreed to pay $7.68 billion for security software maker McAfee.

HP, which is still struggling with the departure of its former CEO Mark Hurd — who has since joined competitor Oracle as a senior executive, a move that’s now the subject of a lawsuit between the two companies — has been moving into the storage and data-center market with a number of recent acquisitions, and seems to be willing to pay a hefty premium in order to do so: The company recently won a bidding war with Dell for the right to buy storage-system maker 3Par for $2.35 billion, a price more than three times the company’s market value before the bidding began.

Related GigaOM Pro Research (sub req’d): Rogue Devices: The Consumer Influence on Enterprise Mobility

Post and thumbnail photos courtesy of Flickr user wlodi

Cassandra was a tragic figure in Greek myth. She could hear the future and thus was able to foretell what was coming next (usually death and destruction). It’s no surprise that no one wanted her hanging around. It’s ironic that an open-source NoSQL software of the same name has often found itself amidst controversy. Today, Cassandra was blamed for scaling (and availability) problems at Digg, which led to the unconfirmed departure of Digg VP of Engineering John Quinn, who was a big champion of Cassandra at Digg.

This isn’t the first time Cassandra — which was created inside Facebook and later open-sourced — has taken a beating. Back in July, Twitter reversed its plans to move from MySQL to Cassandra for storing its tweets. Comments from Digg founder Kevin Rose as he tries to explain some problems on Digg’s new site aren’t helping Cassandra either. However, a call to Matt Pfeil, CEO of Riptano — an Austin, Texas-based startup — put thing in perspective. Riptano is building its business providing service and eventually an easy-to-implement version of Cassandra for companies (see my video interview with Pfeil here.) Pfeil said that Riptano is working with Digg and noted that he would be “shocked” if Digg abandoned Cassandra.

When asked if the problems Digg has had with its upgrade stemmed from Cassandra, Pfeil said, “We’ve reached out to Digg to ID what those problems are. I don’t know the full extent of them, and am learning more from them about their situation. We know Cassandra can scale to levels that are equal to or greater than a Digg is putting on it and I have full faith in Cassandra, but there are these little knobs that need to be tuned and you have to know where they are.”

For Pfeil, this could be an opportunity simply because helping find and turn “those little knobs” are what Riptano was formed to do. He said Riptano has been involved with Digg since around April, which was soon after Digg announced its plans to use Cassandra. While Digg may be able to blame Cassandra for some glitches, the database technology still seems to be on the upswing. Today, Quest — an enterprise software-database support company — decided to support Cassandra through a partnership with Riptano, and companies such as Cisco, Ooyala and Rackspace are also using it.

As Pfeil points out, Cassandra is still new, having been open-sourced in 2008. “Cassandra has come a long way, especially in the last year or so … there is a lot to be done before it is close to where it will compare in production environments to something like MySQL, but we’re getting close.” So maybe unlike the Greek prophetess, the database technology will be able to rehabilitate its reputation.

Related GigaOM Pro research (sub req’d): Report: NoSQL Databases – Providing Extreme Scale and Flexibility

It’s been almost six months since I wrote about Facebook’s attempts to turbocharge PHP ( a web-oriented programming language) using a set of technologies called HipHop for PHP. At that time, Facebook had open-sourced the technology. Just to recap, HipHop for PHP is a sort of converter which takes PHP code and converts it into C++ code and thus creates a binary that can be run on a server natively.

Since then, HipHop has been adopted by a growing number of developers, and Facebook has also roped in the likes of Drupal, MediaWiki, phpBB and WordPress to get their software running under HipHop. According to Facebook estimates, WordPress has become 2.7 times faster when running under HipHop.

Working with the open source community has come in handy for Facebook, which has seen big contributions from others help HipHop become better and more efficient. HipHop has apparently become 1.8 times faster, which has resulted in big improvements on Facebook itself. In fact, Facebook has entirely replaced Apache and classic PHP with HipHop (which also includes a web server.) 

When I spoke yesterday to Facebook’s open-source advocate Scott MacVicar, he said that thanks to  HipHop, Facebook has seen a 50 percent decline in the CPU usage, a trend that has continued even as Facebook has become bigger and more complex in the past six months. Lower CPU usage means that Facebook needs fewer machines that it normally would have needed. “We want to basically speed up Facebook,”  MacVicar said. And in the process, the web. When I asked Haiping Zhao, senior software engineer at Facebook, how much performance gain can be eked out of HipHop, he quipped, “Even I want to know that.” If there’s anyone who knows HipHop, it is Zhao, who’s been involved with the effort for almost three years now.

In response to my original post, a handful of our readers said that HipHop was essentially a good way for large-scale PHP apps to optimize. Jeff Dickey said it was “going to address the CPU-intensive side of the shared-server pain point. ” Jeffrey McManus said, ”Unfortunately it probably won’t resolve the most common types of Web performance problems — this is the kind of tactic that makes sense for those dozen or so Web companies that are operating at Facebook scale.”

Facebook wants to change that, and it seems the project is getting more traction. The HipHop mailing list has gotten bigger, and now has about 1000 folks signed up for it. Another 67 developers have forked it and are using it to build their own tweaks.  Perhaps this news outlined on Facebook’s developer blog will help attract more independent developers to HipHop:

Hui Chen, a Summer of Code student, has worked to add support for FreeBSD and 32-bit environments. We’ve also added preliminary support for OS X. While we’re not planning to use these platforms at Facebook, they’ll go a long way to getting even more community involvement around the project. For example, there are a number of developers who want to test and develop with HipHop on their 32-bit laptops.

Related content from GigaOM Pro (sub req’d):

Can Facebook or Twitter Spin Off the Next Hadoop?

Some of this year’s findings include:

• 71 percent of 2009 respondents indicated that open-source software was easier to deploy, up from 48 percent in 2008, 38 percent in 2007, and 26 percent in 2006.
• 76 percent of system administrators in large enterprises indicate they prefer to use open-source software whenever possible.
• The No. 1 reason for not choosing open-source solutions was lack of support followed by poor documentation.
• Quality of support was the No. 1 reason organizations indicated they chose proprietary management tools.
• 50 percent of respondents indicated that they are already using some form of cloud technology, including, but not limited to, hosted applications, Amazon Web Services, and/or hosted storage
• The top IT management priorities for 2010 were monitoring, configuration management, patching and provisioning, and security.

It’s refreshing to see IT pros focus on the real issues here — support, documentation and he like — rather than responding to the often baseless fears around open source that many proprietary vendors seem to perpetuate.

Related Post from GigaOM Pro (sub. req’d.): Report: Web Worker Survey 2010

The two competing networks each boast peak theoretical download speeds of 100 Mbps, and this pair of providers holds a combined 78 percent share of the wireless market, says Wireless Intelligence. Although consumers are the real winners here, due to competition and choice, a behind-the-scenes company is reaping benefits too. The infrastructure of UCell’s new LTE network is powered by ZTE, a Chinese-based provider of telecommunications equipment and network solutions.

ZTE is quietly becoming a force in LTE, having implemented seven LTE commercial networks and initiated and over 50 LTE trials. Interestingly, one of ZTE’s main competitors, also from China, is Huawai, which is the infrastructure provider for MTS’s competing LTE network in Uzbekistan. Huawei, and ZTE for that matter, are poised to “win the 4G sweepstakes” according to Om, providing the equipment for tomorrow’s wireless networks around the world.

Before you pack up and move to Uzbekistan for a choice in LTE providers, make sure you don’t plan to enjoy fast mobile broadband on the beach. The country is one of only two double-landlocked in the world — Lichtenstein is the other — meaning that Uzbekistan has no ocean access, nor do the countries that surround it.

Related GigaOM Pro Research (sub req’d):

For Operators who bet on WiMAX, there’s an LTE Plan B

The surprising aspect of open source is not its existence, but its success. People do things for free all the time. Among other things, there is no shortage of people willing to share their videos on the web. However, despite the availability of free videos, viewers are often willing to pay money to watch films made by professionals. Professional producers of films in turn usually make full use of copyright laws. In contrast, in many software domains, open source solutions are preferred. The important question about open source is therefore not “Why do people contribute to a project like Apache?” but rather, “Why can’t companies create proprietary products that can beat Apache on the market?”

The reason is based on what economists call “the hold up problem.” When a business relies on assets owned by another party, it may become dependent on that party’s cooperation in the future. In this situation, the party with ownership of a key resource may gain the ability to “hold up” its partner, demanding an unreasonably high price. Hold up becomes a problem especially when a business needs to make large capital investments that assume future cooperation from the owner of a complementary asset.

The hold up problem is particularly severe in the IT sector. Building an Internet company on a foundation consisting of proprietary software owned by others is akin to building a house without owning the land under it. When software is sold in binary form, the buyer is subject to hold up by the vendor; if the software needs to be changed in the future, such changes can only be done with the cooperation of the original vendor at the price that the original vendor demands. By relying on open source, a company can invest in developing its product without fear of being held up down the road. Therefore, open source is an economically powerful solution to the hold up problem.

This holds true even when an open source solution is originally not as good as a proprietary alternative. A technology company using an open source solution benefits from being able to make large complementary investments (and reap of the corresponding efficiencies) without fear of hold up. For example, Facebook has relied heavily on PHP in building its products. When PHP didn’t meet their needs, they fixed it.

Programming languages and software development platforms are particularly at risk of hold up. Software written in a particular language is only useful in combination with a programming language interpreter. Such software presents an investment that could easily be lost if its use depends on future cooperation of the owner of the interpreter. For this reason, software development platforms are a niche where open source has been most successful.

Flash is a proprietary software development platform and products built on Flash are at risk of hold ups. The recent attacks on Flash, most notably from Apple, are therefore not surprising. Apple is heavily investing in its iPad ecosystem. If software written for the iPad makes use of Flash, Apple’s ability to profit from its investments would require future cooperation of Adobe. Banishing Flash from the iPad now avoids future hold up. With the recent release of the WebM project, Google has joined the campaign to replace Flash with open source alternatives. (Unlike Apple’s alternatives to Flash, WebM promises a fully open source solution for web video).

It may seem ironic that the Flash protest has been instigated by Apple – the owner of one of the most closed development platforms today – but it’s hardly surprising. Actors who control valuable resources often stand to benefit most from the openness of complementary products. It remains to be seen, however, whether Apple’s own closed platform will continue to be successful in the face of open source alternatives.

Michael Schwarz is the principal scientist at Yahoo Labs and Yuri Takhteyev is an assistant professor at the University of Toronto School of Information.

In 2005, the main goal of large incumbent telcos and cable companies was to try for an outright ban on municipal networks. As the public vigorously fought back, incumbents switched to creative assaults on communities’ ability to find or use money to pay for networks. Eighteen states have restrictive muni network legislation (see map) that makes building a community-owned network impossible or difficult, especially when it comes to funding them.

In some of these states, municipalities can’t cross-subsidize their networks to take money from one city department (public works, for example) to support the network, something cities often do on the sly. This type of legislation ignores the fact that incumbents can cross-subsidize to their hearts’ content. In Utah, communities can use no more than 50 percent of their sales tax revenue to cover their general obligation (GO) bonds for networks. Utah and other states enable communities to build the network, but the laws do not allow them to sell services directly to the public, cutting off a source of revenue that could help operate the networks.

There often appears to be a great deal of inconsistency among the various states’ stipulations. In Maine, the legislature recently fought to prevent communities from using bond measures to fund broadband projects (read my previous article to learn how municipal bonds work). In North Carolina last month, the state legislature killed a bill that would have made general obligation bonds mandatory for funding these projects. Another tactic is implementing legislation designed to trigger referendums. Triggering referendums work in incumbents’ favor because they can bury an election with truckloads of money and false claims.

There often appears to be inconsistency among the various states’ stipulations. In Maine, the legislature recently fought to prevent communities from using bond measures to fund broadband projects (read my previous article to learn how municipal bonds work). In North Carolina last month, the state legislature killed a bill that would have made general obligation bonds mandatory for funding these projects.

Peel away the covers a little and you discover these are another incumbent tactic of making rules that trigger local referendums, which work in incumbents’ favor because they can bury an election with truckloads of money and false claims. Alabama, as another example, forces cities to hold a full referendum just one week after introducing a network proposal to the city council, which Opelika, Ala. will do on Aug. 10.

However, increasing public pressure is forcing governors and state legislators to look at removing barriers and preventing new ones. Here are some financing approaches that can help communities skirt legislative restrictions and increase their political capital.

Offense is the best defense.

Execute a preemptive strike in the state capitol. Google’s campaign to bring gigabit fiber to communities motivated and harnessed incredible grassroots-level broadband planning in hundreds of communities. This type of motivation can push legislation that scales back or removes legislative barriers, and give communities the right to determine their broadband future without incumbent interference. For example, even as North Carolina was trying to erect hurdles for municipal fiber networks, the bill that laid down those hurdles had an explicit exemption for Google Fiber.

Turn the rules upside down. If state law requires a referendum, conduct a massive needs assessment project that builds community-wide support for a broadband network before putting the measure on the ballot. Seeds well-planted during the assessment produce results on election day.

Be creative with the rules. Utah Telecommunication Open Infrastructure Agency (UTOPIA), a regional effort involving many cities, was stymied by Utah’s various restrictions, including prohibiting sale of services directly to consumers. Building and operating the network was problematic until UTOPIA brought out a trick borrowed from Comcast and other incumbents.

UTOPIA’s CEO Todd Marriott says it used a contractual utilities enhancement, which uses a specific type of financial instrument that enables the city to hook up subscribers without having to put a lien on their homes. Subscribers receive a bill each month for $22 that pays for the buildout, and another $24 that goes to the networks’ operations costs. Marriott says Comcast puts a similar charge on its bills. Private sector companies offer services over the network. This workaround allows the cities to finance the network and indirectly support services going to consumers without violating the law.

Let’s work together.

Communities can expect pushback on even their participation in public-private partnerships (PPP), despite the presence of private companies (including ISPs). The trick is to wield a big stick of public and political support for the project to keep incumbents at bay while dangling revenue as a carrot. Once it becomes clear that a PPP is going to be a reality, expect incumbents to play nice (relatively) in an open-access network environment.

OpenCape Corp. is a PPP that received a $32 million broadband stimulus grant. It plans to own the infrastructure and license much of it to for-profit entities. CEO Dan Gallagher expects to provide fiber to cellular service providers, develop regional area networks with local cable companies for municipalities, school districts, and libraries, and aggregate bandwidth using a regional colocation center.

“We have built-in discounts for the public organizations, but they will still have to pay for the services they use. There are many ways to generate revenue using this infrastructure beyond Internet access, and we hope to exploit them all,” Gallagher said.

Self-financing: If you’ve got it, flaunt it.

Many communities may want to consider the self-financing approach taken by Santa Monica, Calif., a moderately sized city of 88,000 people. Santa Monica had no money in the city budget for broadband, which is exactly where a lot of communities sit today. The city’s CIO Jory Wolf looked at what city departments and schools were paying for antiquated technology and determined that a city-owned fiber network would save money. His team built the fault-tolerant fiber ring in the central area of the city close to where planners expected to find future businesses, re-development projects and the biggest commercial centers. Wolf said:

We built out fiber for $530,000 for City and school use. This effort helped us look five years into the future. We could calculate future benefits, see how we would replace other technology, and project future costs for adding additional infrastructure. The telecom master fund [that] the savings created allows us to implement new technology such as video surveillance, parking control alarms and parking advisory signage to derive those benefits.

Wolf’s team saw businesses as a separate revenue stream via infrastructure leasing, which went into the master fund to be used to fund expansion of the network and additional applications. Five years later, it’s still working that way. It’s a true self-funding model. The city has never had to borrow money and it’s still running in the black. The team started by generating an initial $200,000 worth of savings, then progressed to $350,000 savings in each of the first three years, and $450,000 in the past two years.

Editor’s note: This column is the second in a two-part series on financing municipal fiber networks. The first column can be found here.

Craig Settles is an industry analyst, broadband strategy consultant and Co-Director of Communities United for Broadband who often holds webinars & talks on the business of municipal broadband. Follow him on Twitter (@cjsettles).