My blood pressure just shot up by a multiplier of 500% while reading this report (nice cover picture!) by Alcatel-Lucent about the benefits of high-speed broadband in New Zealand.

Never mind all the assumptions that are used to come up with 32.8 billion dollars of consumer surplus and productivity benefits (surely an underestimate because it does not include HD streaming porn).

The real fundamental problem with the report is that the costs of the fibre build are counted as an economic benefit. These are real resource costs, that could have been used for something else if we do not build a fibre network. They should be subtracted from the benefits, not added.

And shame on the NBR for reporting that “Crown fibre will pay for itself” and saying that $1.65 billion of government investment will generate $5 billion of GDP.

Repeat after me … SHOW ME THE COUNTERFACTUAL …

Update: Ok, the Alcatel-Lucent report presents the GDP “impact” of the fibre build, not the “benefits” of the build. But the problem is the same – GDP is not going to increase just because you choose to use resources for building fibre. What matters is the economic activity the fibre generates vs what would have been generated if the resources were used for something else.

Buying home contents insurance, the tricky part is estimating how much you need for the total amount of the cover. The insurance sales rep suggested I imagine that I lost everything – how much would I need? But losing everything is not a very likely event (touch wood!). And in the more likely case of a minor loss then less than full coverage is all that you need – if you only insure everything for $10,000 and your $2,000 TV is stolen then you will receive a $2,000 payment.

Which makes me wonder why they don’t scale the claim limits for minor claims in proportion to the total amount insured. Perhaps it’s more profitable to just take advantage of people’s propensity to overestimate the probability of catastrophes and encourage customers to overinsure.

Being a bit of an Apple geek, I watched this 1997 video of Steve Jobs talking about the integration of NeXT and Apple, and his future plans for Apple software. One thing that struck me was his praise for Adobe as a small software developer who had made the revolutionary Pagemaker (desktop publishing) software on the original Macintosh. Of course now Adobe is gigantic and is generally regarded as producing bloated, poorly designed software, while the true innovators are small companies writing iOS and Mac apps. I couldn’t help but wonder what this portends for Apple itself.

Stephen King reports on innovation in tertiary education:

I heard about teaching-focused schools that highlight teaching excellence rather than research. I also heard about exceptional research institutions. I heard about business schools that require all faculty to have spent time (in years) practicing their trade in industry. I heard about institutions with explicit dual track approaches recognising exceptional researchers and exceptional teachers for their equally important contributions.

Stephen laments these innovations being lacking in Australia. I believe the same is true in New Zealand. Lack of reward for teaching was one of the reasons why I left academia. According to my student evaluations, I was an excellent teacher, but based on my publications I was an average researcher. Research was regarded much more highly than teaching for determining job security, promotions and esteem of colleagues. This was despite the fact that staff were supposed to spend equal amounts of time on research and teaching.

Perhaps I shouldn’t whinge and just seek out one of those institutions that Stephen mentioned that reward teaching. But I agree with him that it’s disappointing that many universities have such a narrow focus. Great teaching takes just as much talent and energy as great research.

iTunes U has been around for a while in the desktop iTunes app, but the new iPad app is awesome. There are some interesting courses taught by great lecturers at top universities. The quality of the video lectures that I checked out is very high.

The courses are free, so I assume the universities are not making money from iTunes U directly. So why are they doing this? Places like Stanford and Yale are capacity constrained – there are many more students who want to go there than can be accommodated. So their demand is high, and offering free courses in iTunes U probably isn’t going to bring demand down to the point where the capacity constraint no longer binds. This is because the iTunes U courses don’t give you any kind of diploma and you can’t get feedback on your homework.

So there’s not really any revenue loss for the university, and they get some nice publicity and a feel good factor. But for a less famous university that isn’t really capacity constrained, the effect on their demand could be more significant. So I expect to see mainly famous universities offering these online courses, as long as they are free. The happy side-effect is that this should also maintain the high quality of courses in iTunes U.

In the longer term I can easily see a charging model arise. Some of the courses I’ve sampled are so good that I would gladly pay around the same amount that I’d pay for a textbook. I’d pay even more if I could interact with the teaching staff.

I set my default search engine to DuckDuckGo, to give it a proper test. Having used it for a week, I feel a bit like the first time I installed Linux on my computer – It felt good to give Microsoft the finger, but frustrating and rough around the edges.

I like that DDG doesn’t track you and doesn’t try to filter search results according to what it thinks you are interested in. But many times when I couldn’t find what I was looking for on DDG’s first page of results, I tried the same search on Google with more success.

Part of the problem is that DDG doesn’t appear to have anything beyond a basic web search yet. It doesn’t search news, doesn’t seem to search blogs very well, and doesn’t have image search. It will need all of these things to compete properly with Google.

The other thing about DDG is that it’s just a little slow. Searches that are more or less instant on Google take several noticeable seconds on DDG. While it doesn’t seem like a lot, somehow it increases my perceived cost of searching significantly, and makes searching feel like a chore.

Having helped to design a list-making app myself, I’m something of a connoisseur of apps in the category. So the new Clear app by Realmac Software certainly caught my attention. As you can see in the video, it takes minimalism to a new extreme, with apparently no on-screen controls at all, using gestures for every action.

It looks like a very striking design, brilliantly executed, and I’m sure it’s going to sell extremely well. I have a couple of comments though:

• Gestures are the keyboard shortcuts of touch apps. Clear looks pretty intuitive, but I’m not sure if I’d be able to remember them all. I bought Calvetica because it looked cool, but ended up using the built-in calendar app more often because I couldn’t remember the Calvetica gestures.
• Two-finger gestures require the use of two hands, which is not efficient, and sometimes not possible.

The New York Times has a detailed investigation into working conditions at Apple factories in China. I think there’s no doubt that Apple’s net contribution to economic welfare in China is massively positive. Literally hundreds of thousands of people have jobs that, while not fantastic, are significantly better than the alternatives.

The Times article fairly points out that conditions in Chinese factories of Apple’s competitors are just as bad, or worse. However, a company that has just made a record $13 billion profit in three months can expect to come under intense scrutiny. Apple tries to be transparent with its supplier responsibility reports, but it seems these are mostly bark and no bite. From The Times article:

“If you see the same pattern of problems, year after year, that means the company’s ignoring the issue rather than solving it,” said one former Apple executive with firsthand knowledge of the supplier responsibility group. “Noncompliance is tolerated, as long as the suppliers promise to try harder next time. If we meant business, core violations would disappear.”

Most interestingly, The Times translated their article into Chinese and posted it on a Chinese website. Some of the comments by Chinese readers were then translated back to English. Most of the translated comments are along these lines:

If people saw what kind of life workers lived before they found a job at Foxconn, they would come to an opposite conclusion of this story: that Apple is such a philanthropist.

Or these lines:

The story of Apple is just an individual case. There will be endless problems from Pear or Banana … even if you revealed Apple’s inside conduct. We have to solve the fundamental problems, which include labor laws, corporate social responsibility, China’s industrial policies and others.

Just when you thought Google had wrapped up the search market, along comes DuckDuckGo with fewer ads (ie a lower price of searching), and, more importantly, a different approach to filtering.

As DuckDuckGo explains, Google’s obsession with collecting data on you and using that to deliver more ‘relevant’ search results can lead to you living in a search filter bubble, where your own history and biases can affect what see. Thus leading to a vicious cycle of confirming your own prejudices. But maybe some people like to live in a cozy self-created bubble.

What do people really want? The market will decide …

Traffic is backed up on a main road and moving slowly. A small number of drivers want to enter the main road from a side street. The rules of the road say that drivers on the side street need to wait for a gap in the traffic, but drivers on the main road often slow or stop to let them enter. The cost is a momentary delay, the benefit is getting a smile or a wave from the driver who was let in, and perhaps a bit of good karma.

But there is an externality on all the other drivers queued up behind on the main road, as they are slightly delayed too. Thus there will be too much niceness and everyone gets delayed more than necessary. So next time you are thinking about being nice and letting someone in, don’t do it, and play your part to reduce negative externalities :)

This also made me wonder whether traffic signals use economic concepts when calculating the timing of their phases. With enough sensors in the road it should be possible to optimise (ie minimise) total delay.