Friday, October 28, 2011

Cool historic Usenet posts

Usenet News, is a threaded discussion application that started in 1980, long before today's Web-based threaded discussion forums. Usenet news groups circulated globally and covered many technical and non-technical topics.

Google has archived over 800 million Usenet posts in Google Groups. You can search through them here.

They have also pulled together a timeline listing important Internet-related posts from 1981 to 2001.

For example, this is the post where Tim Berners-Lee announced the Web and invited people to download the "very alpha" software and try it out.

The Usenet map shown here was made by Brian Reid in May, 1993. He estimated that around 87,000 sites with some 2.6 million users were exchanging Usenet news at that time.

Thursday, October 27, 2011

The Energy and Emergy of the Internet

(To find out what "emergy" is, you have to read the rest of this post).

Google and other operators of huge data centers spend a lot of money on power, which leads them to locate server farms near sources of cheap, renewable energy, frequently near rivers in rural areas.

Two recent news items, one on Facebook's planned data center near the arctic circle and a study of the energy and emergy of the Internet remind us that the Internet consumes a lot of energy.

Facebook announced plans to build a server farm in LuleƄ Sweden, about 100 kilometers from the Arctic Circle. They will benefit from low cost electricity generated by dams on the LuleƄ river and from savings in cooling. The average temperature is around 2 degrees centigrate, which will enable them air cool the datacenter, which will consume enough power to run about 50,000 houses.

Powering data centers and the computers and devices that use them is only about half the story according to a recent study by Barath Raghavan and Justin Ma. We must also consider embodied energy (emergy) — the energy used in making the devices and infrastructure that make up the Internet. They remind us that we can save energy by extending the life of our computers and smart phones as well as by reducing operating power.

Taking both energy and emergy into account, Raghavan and Ma estimate that the Interent consumes between 1.1 and 1.9% of the 16 TW used by humanity. That is a lot less than we devote to transportation, so we will still be ahead if we can subsitute communication for transportation.

Wednesday, October 26, 2011

Two studies of concentration of power -- government and industry

A study of the relationships between 43,000 transnational corporations has identified a relatively small group, including many banks, with disproportionate power over the global economy.

(Study details are available in this paper).

The image shown here (click to enlarge) represents the 1,318 transnational corporations that form the core of the economy. Superconnected companies are red and very connected companies yellow. The size of the dot represents revenue. Each of the 1,318 had ties to two or more other companies, and on average they were connected to 20. The anlysis revealed that 147 even more tightly knit companies controlled 40 percent of the 43,000 corporations analyzed.

Characterizing the concentration of power in this way is time timely in light of the recent "occupy Wall Street" demonstrations, and provides background for James Allworth's suggestion that we may be facing a choice between capitalism and democracy.

This sort of concentration is worrisome from the standpoint of stability as well as equity. What is the effect of the failure of one of a limited number of large entities -- like banks that are "too big to be allowed to fail?" Note that the database used in the study was compiled in 2007, so it does not refelct changes that have occured during the current economic crisis.

This study reminds me of another case of concentration of control. Around ten years ago, I worked on a study of the state of the Internet in Singapore for the International Telecommunication Union (ITU). I noted at the time that the government played a central role there. They picked critical areas of the economy -- oil, shipping, banking, information technology, biotechnology -- and acted as a heavy-handed venture capitalist. With the help of my nephew, who was working for Goldman Sachs in Singapore, I put together this diagram showing government ownership in the telephone and ISP industries.
The government was not a passive investor. They hired the best and the brightest coming out of the universities. They created some of the earliest strategic IT forecasts and plans, which were put into effect through significant investment. (The best and the brightest avoid government in the US, but that has not always been the case).

This approach has served Singapore well. Today, they are ranked 19th in the world on the ITU ICT Development Index. The top twenty nations all have powerful governments -- most would be considered socialist failures in tea party circles.

We've seen two studies of the concentration of power -- one in the hands of business, the other government. No study or theory will ever be able to fully comprehend anything as complex as an economy or an industry, but these cases indicate that government has a place in ensuring stability and encouraging the development of infrastructure.

Monday, October 24, 2011

Draft of IT literacy paper

I just posted the draft of a paper called IT literacy – evolution, curriculum and a modular e-text.

In the paper, I review the evolution of the IT literacy course from the 1960s to today, then describe my Internet-era curriculum and the modular e-text (currently 104 modules) I am developing to teach it.

Wednesday, October 19, 2011

TeleGeography’s interactive submarine cable map

I love things like TeleGeography’s interactive submarine cable map.

You can zoom and scroll and click on the map for descriptions of the cables or search their cable database.

It is an anatomy diagram for the global nervous system with us as neurons.

Compare that to this map of the NSFNet in 1986:

(Those are 64kbs links).

Tuesday, October 18, 2011

Worries about OpenClass -- a new learning management system from Google and Pearson

Google and Pearson, a large textbook and etext publisher, are joining forces to offer a free learning management system (LMS) called OpenClass. OpenClass will be competition for Blackboard since it will be free, and, given the skills and resources of Google and Pearson, well done. It will also compete with Moodle, which is open source, but requires a significant staff committment to run.

That is the good news for OpenClass. The bad news for OpenClass is that the LMS market is not competitve -- it is dominated by Blackboard, Moodle and a few less popular LMSs.

Even if OpenClass is better than Blackboard and Moodle -- "open and not clunky" as they claim on their Web site -- it will take time to become a major player because we are locked into current LMSs by the material we have created, the training our students and staff have accumulated, and the systems we have built around them.

Do you use an LMS on your campus? If so, how many students and faculty know how to use it? Do students expect it to be used in a class? Have faculty invested in material that is now loaded into the LMS? Do they use it to automate testing and grading? Is it integrated with student records applications and the textbooks faculty adopt?

Don't get me wrong -- this is not a defense of the current LMSs. My campus uses Blackboard, but it is way too "clunky and closed" for me to use. I am just saying that it will be hard to displace even if OpenClass is superior.

Don't get me wrong on that either. I am not enthusiastic about OpenClass. For a start, I worry that it will favor Google Apps and Pearson teaching material. More important, I worry about the pedagogical impact of any LMS. If OpenClass were to become dominant, would it push us to teach within the confines of the OpenClass LMS -- a Procrustean fit?

On my campus today, on-line education is synonymous with Blackboard. We are in the very early stages of networked education and educational technology. It is too early for such standardization.

Don't get me wrong one last time. I believe the folks working on OpenClass at Google and Pearson are smart and mean well. I just wished they worked for a small startup that did not have such deep pockets.

For more discussion see:
Google Plus discussion
Chronicle of Higher Education discussion
Pearson announcement

Sunday, October 16, 2011

New IT literacy teaching modules

I've recently posted or revised 12 IT literacy teaching modules:
This brings the total to 102 modules. Each module has an annotated PowerPoint presentation and assignment, and many have videos, transcripts and other components. (I am adding them as quickly as possible).

Perhaps you can use one or more in a class or for self study. You can see how the modules are structured here and browse through them and here. You can also subscribe to announcements of future modules.

Friday, October 07, 2011

More on massive, open, online classes (MOOCs)

I recently wrote a post on the mother of all MOOCs, the computer science classes getting under way at Stanford this week.

If you found that interesting, check out this podcast interview of George Siemens, who leads Athabasca University’s Technology Enhanced Knowledge Research Institute.

Siemens is a MOOC practitioner and researcher. His classes have enrollments in the thousands, and he handles the large number by decentralizing -- encouraging students to help each other and form study and discussion groups using whichever social media tools they prefer, He is distributing the teaching responsibility to the network as a whole.

Siemens does not suggest that he has found the optimal model -- he is experimenting. He says "we need to tweak or in some cases completely remodel the university system," and he is trying to learn what works and what doesn't.

One thing is for sure -- there is no single answer. What works for a math course may not work for a literature course and what works for an upper division course may not work for a lower division course.

This interview is the latest installment of a monthly educational technology podcast from the Chronicle of Higher Education. I'd recommend checking that out too.

Monday, October 03, 2011

Open online classes starting soon at Stanford – 130,000 students in one class

Stanford's experiment with free, online classes for thousands of students is getting under way. They are offering online sections of three undergraduate computer science courses:  Introduction to Artificial Intelligence, Machine Learning and Databases.

The classes are organized around blogs, as shown here. The lectures, assignments, exams, forums, course materials, quick guides to software and optional exercises are the same whether you are online or on campus.

I took a similarly open course a few years ago from the Harvard Law School. There were three groups of students -- regular law students on campus, an extension class, which met in Second Life, and an open section for those listening to podcasts. I was in the third group, and enjoyed it very much.

The Stanford class is more highly structured and the experience of the online students will be closer to that of the on-campus students than was the case at Harvard. Also, Stanford's online students will receive a certificate of completion, showing their relative rank in the class if they complete the full course.  There was neither social media support nor formal feedback at Harvard.

Others who have offered massive, open online courses (MOOCs) are generally positive, but they report some problems with privacy and spamming and rude behavior.  Since Stanford will allow open students to take exams and do assignments, there is also the possibility of cheating. (When you enroll, you agree to abide by an honor code).

The Stanford courses are unique in several ways. They are large.  The AI class has over 130.000 students from 190 countries.  Stanford will grade and rank open students who choose to be graded. Most earlier MOOCs have been on educational technology, but these are standard academic courses offered by well-known experts in their fields. They will also be using newly developed tools. The AI course is offered in partnership with a start-up called Know Labs, but, for now, there is no information about their tools on their Web site.

This is a bold experiment -- what are the implications for future undergraduate education if these and other experiments with MOOCs succeed?

For links to and discussion of other MOOCs: