An example of bad image processing on the Web

We discuss the importance of processing, resizing and compressing an image before uploading it to a Web site.

This image, found on the WiFi Alliance Web site, provides an excellent example of what not to do:



As we see in the property sheet, the image is 980 by 664 pixels and the file size is 1.28 megabytes:



Resizing and compressing the image before putting it on the Web site would have reduced it to about 5 kilobytes -- 1/250th the size of the original -- with no noticeable loss of quality. (Quality is not even important in this case since the image is just a button linked to a video).

Furthermore, the image is resized to 177 by 99 pixels, which distorts it by changing its aspect ratio. (It appears that the image was also distorted when it was captured).

We can learn from our and other's mistakes -- can you find other poorly processed images on the Web?

A standard battery charger will benefit individuals, organizations and society

We discuss the importance of standards whether they are developed by a dominant company, an industry trade association, a professional society, a government or international organization, or they begin as an Internet Request for Comments.

The International Telecommunication Union has released a recommended standard design for chargers for cell phones and other portable devices, and a number of large manufacturers have committed to supporting it in the near future.

This standard takes on added importance with recent innovation in and rapid growth of the market for smart phones and other mobile Internet access devices.

Of course, we are stuck with a variety of sockets to plug our chargers into:



It's too bad someone did not standardize sockets before nations deployed electrical
networks (click the image to enlarge):



How will this new charger standard effect manufacturers of phones and other portable devices? How will it effect the consumer? How will it effect the environment? Will any individuals or organizations be harmed if this new standard catches on?

Software refinement -- from research to products -- an image processing example

We have discussed the birth of image processing with Ivan Sutherland's Sketchpad program in the 1960s. As we see in this video, Sketchpad could draw, rotate, resize, copy and combine images.

Since Sutherland's research, there has been tremendous improvement in hardware technology and innumerable engineering refinements to image processing software.

Software refinements require research and the development of prototypes before they become product features. Consider, for example, the PatchMatch algorithm developed in collaboration between Princeton University and Adobe Systems. Here we see a paper describing the algorithm and some demonstration videos. The paper was presented and the videos shown at an academic engineering conference.

Successful research like the PatchMatch algorithm is subsequently incorporated into commercial products. Here you see a demonstration of tools that use PatchMatch for image retargeting, completion and reshuffling integrated into Adobe Photoshop. The tools are being demonstrated at a trade show, and it is probably safe to say they will be in the next version of Photoshop.

This is typical of engineering refinements to software. A prototype is developed in a research lab and it later becomes a product.

Researchers typically work on things that are too demanding for the technology of the day. Some of the Princeton videos are shown at 5 times actual speed because it takes time to compute the new images. By the time these features are released in Photoshop, hardware will have improved and they will execute rapidly.

Note that Sutherland's work was a pure university effort, but PatchMatch was a joint effort between a university and a company. Many significant advances have come from industrial research labs. Can you find some interesting prototypes and projects at Microsoft Research?

Microsoft's answer to Google Apps for education

We have seen that universities like Northwestern, Arizona State and Abilene Christian have adopted Google Apps and Mail.

Microsoft's answer to Google is their Live@Edu program, and they have just entered into a long term agreement with the University System of Ohio for "cloud" services.

Wireless data coverage is uneven

We discuss evolving cellular data generations. Today's third generation equipment provides for speeds of up to 2 mbps, but, as shown in this figure, download speed varies considerably from one city or neighborhood to another. The download speed to your iPhone 3G might be as low as 400 kbps or as high as 1,600 kbps.



The plot was generated by ARCchart, a wireless market research firm, and described by Brough Turner in a blog post. ARCchart monitored over two million performance tests using iPhone, Blackberry and Android phones, then filtered them to focus on major cities. This graph is based on 648,374 downloads from major cities in 103 nations between August 2008 and June 2009.

(Gizmodo performed a more limited test of Sprint, AT&T and Verizon 3G networks in eight US cities and also found considerable variance in download speed).

Of course, in some places there is no GSM coverage. Consider the coverage by AT&T, the GSM provider supporting the Apple iPhone in the US -- there is no coverage in the light-colored regions:



This map was taken from AT&T's coverage viewer in early September 2009, and coverage has continued expanding since then.

However, AT&T states that the maps are only an approximation, not a guarantee, of their coverage, which may be effected by terrain, weather, foliage, buildings and other construction, signal strength, customer equipment and other factors. There are many anecdotal reports of inability to use an iPhone in parts of San Francisco and the bay area.

Network neutrality is a global issue

We talk about the Internet being designed as an end-to-end network in which network operators only make fast connections between computers -- the network does not provide services or differentiate between one person's packets and another's. This approach is generally favored by companies like Google and Skype, which offer Internet application services, and opposed by Internet service providers like Verizon and AT&T, which must deliver those services, and, in some cases, compete with them.

The US FCC recently endorsed the vision of an open, neutral Internet, and The Washington Post has published a story about Skype's lobbyists in Washington, who were quite happy with the FCC action. Skype is lobbying for neutral networks in every nation, not only the US. They feel the example set by the US FCC will have an impact on regulators in many other nations.

The US is among the world leaders in this policy, a nice place to be for a change.

FCC Chairman calls for network neutrality

One of President Obama's campaign promises was to support network neutrality. Federal Communications Commission Chairman Julius Genachowski has now called for new rules guaranteeing network neutrality.

Mr. Genachowski outlined his proposal in a talk at the Brookings Institute. After discussing the importance of the Internet to our economy and society, he reiterated four current FCC Internet governance rules:

• Consumers should not be limited in the content they choose to view online, as long as it's legal.
• Users should be able to run any application they want as long as they don't exceed service plan limitations or harm the provider's network.
• Consumers should be permitted to connect products they buy to their Internet connection, as long as the devices operate within the service plan and do not harm the network or enable theft of service.
• Customers should be able to easily review their options when buying Internet service plans and learn how those plans protect against spyware and other invasions of privacy

and he added two new network neutrality rules he would like to enact:

• Internet service providers (ISPs) would be prohibited from selectively blocking or slowing Web content or applications.
• ISPs would be required to make their network management practices clear and available to consumers.

The network neutrality provisions are opposed by many ISPs like the telephone and cable TV companies and supported by content providers like Google. You can read an argument against network neutrality here.

You can watch a video or read a transcript of Mr. Genachowski's talk and a panel discussion following it here. If you are in a hurry, you can check out this short coverage from National Public Radio. Last, but not least, you can see announcements and give feedback to the FCC here.

Do you support the call for network neutrality? Do you think it should apply to mobile ISPs?

Is technological progress slowing?

We have seen exponential progress in electronic, communication and storage technology during the last 60 years. As a result, computers and networks are now used by billions of people for a great number of applications. This "information revolution" has had an impact on our individual lives, organizations and society.

No doubt, computers and the Internet are a big deal, but Alfred Normann argues that they have not changed our lives as much as earlier inventions -- that he has not seen as much technological innovation as his parents and grandparents did. As we see in his illustration, shown below, a host of important inventions occurred during the 50 years between the Civil War and World War I and many others occurred during his grandmother's life. He feels that technological progress measured in terms of impact on our lives has slowed, not accelerated.

Ray kurzweil would disagree with Normann. He argues that accelerating improvement in information technology facilitates an accelerating accumulation of knowledge, which is leading rapidly toward an understanding of our own biology and intelligence. Kurzweil foresees radical medical breakthroughs, leading to the possibility of immortality, and machines that surpass human intelligence and continue to improve without us.

How does the impact of the Internet on individuals, organizations and society compare to the impact of inventions like the automobile, radio, television, antibiotics, electric lights, telephones, and nuclear weapons?

Will today's exploding knowledge of genetics and biology lead to a longer life for you and your children? Can you think of examples of things computers do today that would have been considered "intelligent" 50 years ago?

The general public is unaware of Internet policy issues and their impact

Geoff Daily just wrote a blog post explaining most people do not understand broadband or bandwidth. As an example, he told about a friend who did not know how his apartment was connected to the Internet and was

pretty much totally oblivious to this language of bandwidth, bits, and bytes. And he certainly doesn't know anything about bandwidth caps or traffic shaping

His friend had invited a dozen people with laptops to his apartment for an online fantasy football draft -- would there be sufficient bandwidth? Would the extra usage exceed a cap and cause an unexpected jump in his Internet bill?

More important, most of the general public is unaware that US broadband connectivity is slow, asymmetric and falling behind that of other developed nations. As shown below, by last year the US had fallen to 11 th among OECD nations in broadband connectivity per capita, and we had very little fiber installed.



What are the implications of lagging connectivity in the US for our quality of life and economy? If all of the freeways in the US were two lane streets, would that effect the quality of your life? Would it effect the economy?

How fast is your Internet connection at home? To your cell phone? Do you have unlimited usage or does your bill increase when you exceed a cap? Does your connection bog down when more than one person in your home is online? Could two people watch a low-resolution TV show without pauses and glitches? Could you watch an HD movie?

For more on Daily's views on bandwidth and bandwidth requirements, click here.

Excellent lectures from The Stanford Technology Ventures Program

The Stanford Technology Ventures Program Entrepreneurship Corner is a free archive of entrepreneurship resources for teaching and learning, including videos and podcasts of talks by successful entrepreneurs.

Many of the talks are by IT people. For example, Microsoft CEO Steve Balmer and Siebel Systems founder Tom Sieble spoke recently. Balmer spoke about the early days at Microsoft, the patience and hard work required of an entrepreneur, and the great opportunities open to Internet entrepreneurs today. That was in contrast to Siebel who spoke about opportunities in health care and energy, and asserted that the time for opportunities in IT is past. In a third lecture, neuroscientist and IT entrepreneur Jeff Hawkins reviewed his career, including the creation of the Palm Pilot and other portable computing devices and the founding of several companies and a research institute.

They have over a thousand two or three minute video snippets from the talks. You can search through them by key word or speaker. For example, there are 24 by Google's founding CEO Larry Page. You can watch one or two, or stream all 24 together.

Balmer is optimistic about the future of the IT business, but Siebel is not -- what do you think? What personal lessons can you learn from these entrepreneurs?

Senators call for investigation of cellular company practices

Several US Senators, including John Kerry and Herb Kohl, are calling for investigation of anti-competitive practices by cell phone companies. The Senators are bothered by restrictions like limiting the Apple iPhone to one carrier, AT&T, and the lockstep price increases we noted earlier.

It is encouraging to see these issues raised, but anti-competitive behavior by the telephone and cable companies goes far beyond cell phone restrictions.

We need to question fundamental assumptions about the role of the Internet as basic economic and social infrastructure, the appropriate role for government ownership of infrastructure, and the service-oriented business model of the incumbent ISPs .

This call for investigation is limited in scope and far from action and results -- let's hope it is just the first step.

Open networks benefit both ISPs and the public

We have spoken of the economic inefficiency of the service-oriented business model of the US telephone and cable companies. They see themselves as selling video, telephone, and Internet service -- the so-called "triple play."

We see the social cost of this business model when we compare the US with other nations, where access network operators open their networks to competing service providers. We pay a lot for inferior broadband connectivity.

It is not surprising that the telephone and cable companies are trying to increase their profit at public expense, but there is ironic evidence that this service-oriented business model may also be hurting them. In a nutshell, the argument is that the owner of an access network can make more money by opening it up to all service providers. True, they would forsake the monopoly profits of being the sole supplier of telephony, movies or television on their network, but they would make it up in wholesale and retail sales volume as many companies compete to sell these and other services.

If this is the case, open networks are a win-win proposition. Both the public and the cable and telephone companies win.

Yankee Group analyst Benoit Felten will present a one-hour webinar on this topic on June 30. The presentation will be archived, so you can listen later if you miss it.

For further discussion of this "dumb pipe paradox" see Brough Turner's blog.

Proprietary research firms are an excellent source of information

Proprietary market research firms like The Yankee Group, Infonetics, and Forrester Research study various industries and technologies. They typically make money by holding specialized conferences and selling reports and consulting services to companies that are in or depend upon those industries and technologies. For example, a report on the future of WiFi would be of interest to both WiFi chip manufacturers and to computer companies that will be using those chips in their products.

Market research firms hire highly qualified analysts and sell the results of their research to many interested companies, enabling them to share the cost of the research. That is good news -- it makes business more efficient. The bad news is that the research reports are too expensive to be used in education.

But, since the market research firms want publicity, they provide some very useful information for free. They publish press releases and executive summaries of their reports as well as blogs and webinars.

Take The Yankee Group for example. They bill themselves as "The Global Connectivity Experts," making their research relevant to our class. Twenty six Yankee Group analysts write blogs with posts in 30 categories. You can read press releases on and summaries of their research here. Yankee Group analysts also present webinars on their research. One can register to watch these online presentations as they occur or download them afterward.

Which of the Yankee Group bloggers or topic categories is most relevant to our class? Which of the previous webinars seems most relevant?

Television: single service; Internet: multiple services

In a recent post, we spoke of the ISP's service-oriented business model, contrasting it with the unfettered delivery of information (bits) on the Internet.

As an example, we cited the NBA playoff games being streamed over the Internet as well as broadcast on television. We noted that the service-oriented business model maximized carrier profit at the expense of consumers and the national economy.

In addition to being economically efficient, the Internet allows us to combine the game coverage with services like the LA Times chat room shown here:

This and other services like providing expert commentary, statistics, alternative play-by-play reporting, supplementary video, etc. would add to the viewer experience and create an online community of fans.

Would you enjoy chatting with other fans while watching a sporting event? Would you rather watch a sporting event on broadcast television or your computer screen? Which would you prefer if the Internet speed increased to the point where it could match the size and resolution of your television set?

How will WolframAlpha change math instruction?

When portable calculators were invented, we debated whether students should be allowed to use them in school and on tests. WolframAlpha takes this question beyond grade school.

Stephen Wolfram, co-founder of Wolfram Research, invented and commercialized Mathematica, an incredibly powerful symbolic math program. A command line version of Mathematica is embedded in Wolfram's latest product, the WolframAlpha Web service. Here are a few examples:

• Enter y = 10 x, and you are told that it is the equation for a straight line with x and y intercepts at 0 and a slope of 10, and the equation is plotted.
• Enter x^2, and you are told it is a parabola, which it plots and gives the global minimum, derivative and indefinite integral.
• Enter x^2 + y^2 + z^2= 10, and you get a plot of a sphere along with its center, radius, diameter, volume and surface area and other information including derivatives.
• Enter sin(x)^2 + cos(x)^2, and you get a plot of a constant 1 along with series representations and integrals.
• Enter derivative x^3, and you get the derivative and its plot plus an indefinite integral and global maximum.
• Enter integral x^2 + 3x dx, and you get the indefinite integral along with its plot. If you click on "show steps," you see the steps leading to the solution.
• Enter definite integral x^2 (0,1), and you get the definite integral from 0 to 1 along with a plot.

You get the idea -- try a few for yourself, and you will be amazed at the power of this symbolic math program.

Should students of algebra, geometry, trigonometry, finance, statistics, calculus, etc. be allowed to use WolframAlpha in class, for homework, during exams?

P. S. For links to further discussion, see this post.

The information-service business model is costing us dearly

The incumbent Internet service providers have a service-based business model, differentiating between Internet, telephone, text-message, basic television, premium television, voicemail and other services.

This enables them to vary prices to maximize their profit and discourage competition. For example, they charge exorbitant fees for text messages and discourage Internet video by charging extreme prices when download caps are exceeded.

This differentiation between types of data or service is arbitrary. It is all bits.

For example, the NBA playoff game on Sunday was televised and streamed over the Internet. Both came over the same cable:



As we see, the television coverage, including the ads, is being delivered over the Internet with a four second delay. The Internet image quality is below that of the television signal, but that will improve when US Internet speeds catch up with the rest of the world.

The incumbent telephone and cable companies profit from their service-oriented business model, so they will resist becoming information utilities, delivering undifferentiated bits.

Can they sustain that position in the long run? Would we tolerate a water company that differentiated between drinking and washing water or a gas company that differentiated between heating and cooking gas?

The cost of high-speed international connections -- Africa loses

Telegeography, a market research firm, compiles a database of wholesale Internet prices -- the prices a global organization or an ISP would pay. This chart shows a few price examples:

The price differences are dramatic -- a 2 Mbps link between London and Johannesburg costs roughly the same as a 10 Gbps link between London and New York.

A recent New York Times article points out that the cost of serving Web content over slow, unreliable links in developing nations is high and the income from advertising is low. This has led some companies to block developing nations from their sites or to offer degraded content by, for example, compressing video or serving "light" versions of pages.

What are some of the causes of high international link prices in developing nations? What other factors contribute to the Internet being relatively slow and expensive in developing nations?

4G cellular is expected to roll out faster than 3G did

We have discussed three generations of cellular technology, and the fourth is beginning to take shape.



It took nearly six years for 3G cellular to reach 100 million subscribers world wide, but market research firm Pyramid Research estimates that 4G Long Term Evolution (LTE) technology will grow faster. They expect a compound annual growth rate of 404% from 2010 to 2014, with 136 million subscribers by year-end 2014.

Note that Pyramid Research expects LTE to rapidly surpass mobile WiMAX.

At 100 Mbps, LTE will not be limited to cell phone service, but will be used for fixed, portable and mobile Internet access.

If you live in a large city in a developed nation, you might look forward to LTE or Mobile WiMAX by 2011. Others will have to wait. (Some for a very long time).

Is 3G cellular currently available throughout the city you live in? Throughout developing nations? What applications and mobile devices do you foresee for 100 Mbps mobile connectivity? What changes may occur in the cellular companies business and billing models?

Substituing communication for transportation -- can we reduce business travel?

My son-in-law Mark flew from Los Angeles to Seattle last Thursday for a job interview. The round trip air fare plus four cab rides cost $619. He had a long day, getting up at 4 AM to catch an early flight, and was not home until early evening. If he had stayed over in Seattle, the a hotel room and meals would have brought the cost close to $1,000.

Three people interviewed Mark, and he feels that a teleconference would have been as effective as flying up for face-to-face interviews.

Ten years ago, I conducted a study of business travel at Hyundai, USA. Nineteen employees monitored their travel for three weeks, identifying trips which could have been avoided if various communication alternatives had been available. Based upon that data, we estimated savings of 3.6 hours and $298.07 per employee per week.

In the ensuing ten years, the cost of travel has risen somewhat and the cost of communication has dropped dramatically. Why did they fly Mark to Seattle rather than conduct a teleconference?

The standard answer is that information is lost in a teleconference -- you don't get the same feelings and subtle clues as when you meet someone face-to-face.

If you have any doubt that a teleconference can convey emotion, watch this excerpt from A Hole is Space, showing clips from a two-way video link between people in New York and Los Angeles in 1980. A Hole in Space was not a business installation, but performance art anticipating today's emerging high-speed connectivity.

A high bandwidth teleconference can convey emotion, but that is not widely accepted in our culture -- yet. But, my grandson has grown up making Skype video calls to his relatives in Chile. By the time he starts high school, he will be video chatting on his mobile Internet-access device. Will a teleconference seem emotionless or unusual to him when he joins the workforce?

Have you done video chats or conferences? If so, was it as effective as talking face-to-face? Would the experience have improved if you had a very powerful computer and very fast connection to the Internet? What are the economic and environmental implications of substituting communication for transportation?

International VoIP is growing fast -- what are the implications?

As a result of globalization, we are making more international phone calls every year. In 2008, we made 384 billion minutes of cross-border calls -- 12% more than 2007.
This graph shows the annual growth rate and the total number of call minutes for both traditional time-division multiplexing (TDM) and voice over IP (VoIP) calls. Both are growing, but VoIP is growing faster.

For example, traffic on Skype, the leading international VoIP provider, increased by 41% in 2008. Skype carried 33 billion minutes in 2008, accounting for 8% of international voice traffic.

This sounds like good news for the industry, but, as we see below, prices are falling as traffic increases, keeping revenue roughly flat.
(This data was from Telegeography, a leading telephony and Internet market research firm and our class notes on VoIP are here).

Do you make international calls? How does the cost per minute using Skype or other VoIP services compare to TDM calls? Do the traditional phone companies make any money on VoIP calls?

What will happen to telephone companies if VoIP continues to outgrow TDM? How will this trend effect developing nations?

Zappos and Flickr -- excellent and terrible customer service on the Web

Here is a short excerpt from a talk by Zappos.com CEO Tony Hsieh, in which he discusses their customer service.

Customer service excellence is a core value at Zappos. They publish their 24/7 toll-free phone number on every Web page. Customer service representatives are expected to give friendly, helpful “above and beyond” service, and Zappos does not time calls or set sales-based performance goals. They view a call as a branding opportunity rather than a necessary expense. Reps even direct customers to other Web sites when appropriate.

Zappos also gives free shipping and return shipping, accepts returns for a year, stocks every item in their own 24/7 warehouse, and often surprises customers with an upgrade to overnight shipping.

(Hsieh's full talk, in which he stresses culture building more than customer service, is here).

Contrast that with my experience with the absurd customer service on the Flickr photo sharing site. On March 13, I submitted a question asking why people could not see photos I had placed in a "public" Flickr group. The next day, I received this response:

Just a quick email from Team Flickr to let you know that we've successfully received your recent Help by Email query and we hope to respond shortly.

We'd also like to take an opportunity to remind you that one query is sufficient and multiple queries regarding the same issue make the Magic Donkey cry.

Lastly, you may not be aware that our FAQs and forums are full of help goodness.

The Flickreenos

They tried to blow me off to their FAQs, warned me not to resend my query, and irritated me with their cutesy language, but, that would be OK if they helped me with the problem. The next day, I received an email saying they had escalated the question to a "senior representative."

But, evidently the senior representative had a five week backlog. On April 24 he got back to me with a non sequitur, boilerplate-laden answer.

Customer service clearly costs Zappos more than it does Flickr, and they are targeting a high-end retail customer, but, no matter what the purpose of your Web site is, you should study Zappos' example and do a lot better than Flickr.

Have you seen other examples of very good or very poor customer service on the Web?

Would you like to own and install fiber to your home?

There are many examples of municipal ownership of access networks, but ownership and control could also be pushed out to home and building owners.

Wu and Slater discuss this alternative in a recent paper and a test is underway in downtown Ottawa, Canada, where fiber has been deployed to serve a 400–home neighborhood, but a service provider has not yet signed on.

The Norwegian telecommunication company Lyse Tele reports that 80% of their 130,000 customers have agreed to dig their own trenches and bury their own fiber in exchange for a discount on installation. This has been good for business -- only .2% of customers who have installed their own fiber switch to another service provider.

While this sounds good, there are questions. Who owns the fiber, Lyse Tele or the home owner? Who controls the fiber -- can their customers reach competing services or are they locked in to Lyse Tele?

I have seen estimates that it costs Verizon about $1,000 to connect a home. I would gladly pay that if it meant I owned and controlled the fiber and could get connectivity from competing ISPs over it. I would consider it an investment -- increasing the value of my house -- not an expense.

I own my own sewer, water and gas lines and call a plumber when there is a problem. I would be happy to own my fiber.

Four excellent videos on municipal fiber networking

We have discussed infrastructure ownership options, and Benoit Felten has posted four excellent videos on municipal fiber networking. Three are interviews of people who gave talks at the recent Freedom to Connect Conference: Terry Huval, Tim Nulty and Bill St Arnaud. The fourth is a presentation given by Felten in New Zealand last month.

1. Terry Huval, Director of the Lafayette, Louisiana Utility Service fiber to the home project.

Huval discusses the motivation and business model that gave rise to the project, their legal battles (3 years of fighting until the State Supreme Court ruled unanimously in their favor), the services they offer, and the applications they will be offering in the future.

2. Tim Nulty, Project Director, East Central Vermont Community Fiber Network (ECFiber).

ECFiber plans to build rural fiber connectivity to 21,866 households in and around 22 Vermont towns. (So far 21% of those households have pre-registered). Nulty presents data on costs and revenue, and concludes that the network is a viable business. They will offer both retail and wholesale Internet service over the network, and he explains why a pure wholesale network like that in Stockholm makes sense in Europe, but would be defeated by the incumbents in the US.

3. Bill St Arnaud, Chief Research Officer at CANARIE, Canada's research network with a mandate to develop next generation networks, applications and services.

St Arnaud describes the "G-commerce" model, which combines connectivity with energy savings and pollution reduction. Installation of fiber to the home will be financed by a 1-2 cent per kilowatt hour increase in electric bills and energy-cap savings from reduced power consumption as high speed communication substitutes for transportation.

4. Benoit Felten, Senior Analyst, Yankee Group.

Felten gives examples from France and the Netherlands showing that network sharing is profitable even for incumbent ISPs. Take-up rate is more important than average revenue per customer, and the fastest way to convert 100% of the population to fiber is sharing it among service providers. He outlines and presents examples of several business models ranging from passive sharing of access and rights of way to offering retail service.

Why is connectivty in Stockholm so much faster and cheaper than in US cities?

We've been discussing US broadband policy and the stimulus package, and this table shows the cost of fiber-based, residential Internet service in several cities (Brough Turner provided the European and Asian data):

Where?	Monthly cost	Uplink (Mbps)	Downlink (Mbps)
Stockholm	$11	100	100
Seoul	$24	100	100
Hong Kong	$35	100	100
Tokyo	$61	100	100
Amsterdam	$127	100	100
Lafayette, LA, Municipal	$58	50	50
Lafayette, LA, Cox Cable	$140	5	50
US, where available, Verizon	$145	20	50

Can we explain the large speed and cost differences?

The Cox Cable offering in Lafayette, Louisiana seems to be the worst deal. It is the slowest and only five dollars a month less than the Verizon network. The municipal network in the same city is faster and cheaper. The Cox network reaches more neighborhoods than the municipal network, and they are forced to compete with temporary sale prices.

Stockholm is at the other extreme. They have a municipal network that reaches every block in the city. Unlike Lafayette, they do not offer consumer service over their fiber, but lease network access to anyone who would like to offer service. The Internet service providers, including incumbent telephone and cable companies, compete on an equal footing.

As a result, there are many competing service providers in Stockholm, and, as Turner points out, the city owns the expensive, long-life assets like fiber, rights of way, conduit, and tunnels, and the service providers own the electronic equipment that is relatively cheap and is upgraded frequently as technology improves.

Many factors determine the cost of Internet connectivity, but the ownership model is significant, and it seems the Stockholm model is superior to those in the US.

Note that analysts at the OECD also endorse the Stockholm ownership model, writing that:

Municipal networks can play an important role in enhancing competition in fibre networks. If these develop, governments should encourage them to be open networks, that is providing dark fibre to service providers rather than becoming themselves service providers. Nor should the existence of a municipal network providing dark fibre mean that investment in other fibre networks in that municipality should be prevented.

Should some of our broadband stimulus funds be used for Stockholm-style municipal networks?

Click here for a paper with more on this topic.

Click here for a PowerPoint presentation on this topic.

Some neat bookmarklets

Would you like to convert a Web page to a PDF file? One way to do that is to install a browser add-on from Pdfdownload.org. The add-on lets you choose to have the converted page emailed, opened, or saved on your hard drive.

Or, instead of using the add-on, you could use the Pdfdownload.org bookmarklet. The bookmarklet does not require installation, works with most browsers, and will not slow your browser down or consume memory. It can even be stored on the Internet for access from any computer. However, it lacks features of the add-on -- it does not offer the option of emailing or saving the PDF file; it just opens it.

Bookmarklets are similar to add-ons, but they may have fewer features. They are small JavaScripts that execute when you click on their icons in the bookmark toolbar. Here is the code of a simple bookmarklet that re-sizes the browser window to 800 by 1,000 pixels (which would be useful if you were preparing a series of same-size screen shots):

javascript:window.resizeTo(800,1000);window.moveTo(0,0);

The Pdfdownload.org script is a bit longer because it calls the PDF conversion service using their API:

javascript:void(window.open('http://www.pdfdownload.org/web2pdf/Default.aspx?left=0&right=0&top=0&bottom=0&page=0&cURL='+document.location.href));

There are tons of useful bookmarklets, and you can see a list of some of the best at this excellent Digital Inspiration post. If you start using bookmarklets, you will also like this post.

What is your favorite browser add-on? What is your favorite bookmarklet?

Three extremely simple Web services

The first day of class, we ask "what is a network-based application?"

Examples provide a partial answer -- a starting point. We do not want to be bogged down in details, so we want very simple examples. Here are three:

Eggtimer is a simple application where the program is stored on the network and downloaded each time it is used. Go to http://e.ggtimer.com/5 for a five-second timer or go to the home page for other options.

http://drop.io/ demonstrates an application in which the program and data are both on the network. With a couple of clicks, users can upload and share text, recordings, images, etc. Here is an example.

(Drop.io is surprisingly useful for such a simple application).

Aviary screen capture is a third example. You can capture and then edit any page on the Web by simply typing aviary.com/ in front of the http:// in the page URL.

For example, to capture the YouTube home page, you would enter:

aviary.com/http://www.youtube.com

Aviary automatically downloads the captured page image and an image editing program written in Javascript. Since Web clients contain Javascript interpreters, you can edit the page image when it gets to your computer.

Do you know of other network-based applications that are as simple to use and demonstrate as these?

AT&T mobile Internet access -- $480 per gigabyte when over the cap

We have seen that the price of sending a cell phone text message far exceeds the cost. What about mobile data plans?

An Oklahoma City woman is suing AT&T because her first month phone bill was over $5,000.00. She had subscribed to a $60 per month AT&T data plan that was capped at 5 gigabytes. She exceeded the cap the first month, and felt she had been deceived when the bill arrived.

The AT&T plan states that if one exceeds the 5 gigabyte cap, they are charged "$.00048 per kilobyte."

Would she have exceeded the cap if the price had been quoted as "$480 per gigabyte" or, better yet, "$350 to download a CD" or "$2,500 to download a DVD?"

Why does AT&T charge $12 for each of the first five gigabytes and $480 for the sixth? Do you think she should have to pay the bill?

The FCC national broadband plan -- can the Internet help?

The American Recovery and Reinvestment Act of 2009 directs the FCC to develop a national broadband plan within one year. This is reminiscent of the energy policy task force President Bush established during his second week in office. Bush's task force operated in secret, but President Obama has promised us an open and transparent administration, using the Internet as one tool.

It would be a shame if large, incumbent Internet service providers -- telephone and television companies -- dominated the FCC broadband planning process the way large energy companies dominated the formulation of the Bush energy policy.

Before deploying next generation access networks, we should consider new technologies and ownership and business models, but that will not happen if the incumbents write the plan.

Can the Internet be used to open the FCC planning process, to take it beyond the "beltway" and its lobbyists? The administration has established the Recovery.gov Web site to inform us on stimulus spending. How might the FCC use the Internet to open its broadband planning process?

Stimuluswatch.org -- fast, cheap, ad-hoc development on the Intenet platform

We have seen that application development and delivery platforms have evolved over the years, and have discussed the ease of building complex applications on the Internet platform.

Stimuluswatch.org provides an excellent example. It was conceived of in a blog post by Jerry Brito, and volunteer programmers, who did not know each other, responded. They built the site using data from another site and existing Web services.

Brito posted his idea on December 11 2008, and the site launched February 2 2009. You can read more on this ad-hoc, decentralized development project here.

Recovery.gov is Web 1.0; Stimuluswatch.org is Web 2.0

The President has signed The American Recovery and Reinvestment Act of 2009, which includes $7.2 billion for broadband access. You can see a brief summary of the bill here.

As we have seen, the Obama administration hopes to use the Internet for transparent, two-way communication with the public. To this end, they have launched the Recovery.gov Web site, which will be continuously updated, telling us "how, when and where" the recovery funds are spent.

As of today, Recovery.gov is definitely a Web 1.0 site -- it summarizes the Recovery Act, requests comments using an email form, and asks us to check back frequently for data on spending. They don't even have RSS feeds.

Contrast that with Stimuluswatch.org, a Web 2.0 site. Stimuluswatch began by importing a database of "shovel ready" projects that was posted by the US Conference of Mayors. Users can search the database by city, keyword and project type, and view the project descriptions and estimated cost and number of jobs created.

But, the main purpose of the project database is to organize user input. Users who are familiar with a particular project can drill down to the project page and:

• Make neutral, factual changes by editing its wiki page
• State opinions and debate the value of the project by posting comments
• Share comments with friends on Facebook
• Vote on whether or not the project is critical

(Check this video for a more complete description of navigation and user input).

I checked my city, Los Angeles, and the results were sobering. The database includes 321 projects with a total estimated cost of $7.3 billion, creating and estimated 82,341 jobs. Of these, only 11 projects had more yes the project is critical than no votes, and they accounted for only 3% of the cost of all proposed projects. Obviously citizen votes are only one factor to consider in selecting projects, but this indicates that we need to be selective if we are to avoid "porkwatch.org."

It sounds as though the administration plans to use Recovery.gov to let us know what they have done, but Stimuluswatch.org could be used in deciding which projects should be funded. That could be done by extending the database to include requests for Federal grant applications that are actually submitted now that the stimulus package has been approved.

Stimuluswatch.org was conceived of and built in under two months -- the administration should be working with them.