vitaminj

Completely agree - I love the Economist as a magazine, but will never be a subscriber again after how much time it took to cancel.

For what it is worth, Australia has 5 minute generator dispatch, but 30 minute settlement. The market operator is presently implementing a project to bring settlement to 5 minute as well.

A few comments are saying that the author is too harsh on Samsung, but I get the impression that he actually admires Samsung's unscrupulous, unethical though ultimately canny business practices - in a similar way that one would admire a drug dealer's ultra-efficient distribution system.

After all, the Samsung presented in this article enters into new markets via wholesale IP theft. It then uses a suite of legal instruments to stall for time in order to build internal technical capabilities and intellectual capital. Samsung could stop here, but chooses instead to actually innovate and improve on the products using the knowledge and experience base it has accrued copying the product in the first place. It's a pretty shrewd, albeit completely unethical business strategy.

There is a third way, one in which research is not a full-time endeavour, but is still a part of your life. It's certainly not for everyone, and is not "pure" in the sense that you still need to make a living doing something with market value. I generally work 3 days in industry and do research the other 2 days in a university setting (a PhD program).

My work is in the broad area of my research (e.g. power systems), and while there's no direct overlap, they aren't entirely different spheres. I quite like my work and I find the industry contact important, so while I could easily live on 1-2 days of work, I've chosen a work/research split that is more biased towards work. Having said that, as a freelancer, I have a fairly flexible arrangement and some weeks I don't work at all.

The key advantage of this approach is that I am not at all bonded to the university. I don't have to participate in the politics of academia, grant funding, the pressure to publish, progress reporting etc, which I see the other grad students struggling with. I suppose I still have to massage the egos of tenured professors, but I can live with that (as a consultant, I'm always massaging peoples' egos anyway). The bottom line is that having independent funding insulates me from most of the pain of being a PhD student and gives me the freedom to pursue ideas that may lead to dead ends.

A few years ago, I volunteered in SE Asia for 18 months using my electrical engineering skills to help rural villages get access to electricity. At the time, I'd been working professionally for around 7 years and was just becoming competent at working independently. Although I'd gotten my PE status a year earlier, I can't say that I was at a senior engineering / consultant level.

So it was a surprise for me to find that I was one of the most experienced and skilled engineers in my organization (and in many other energy-related organizations for that matter). I concur with the OP and have met quite a few western volunteers that were well-intentioned, but generally had no technical skills.

Of those who had qualifications, they were usually in the social sciences, development studies, media / communications, public relations, etc. Useful skills no doubt, but I felt that the country could have benefited more with direct assistance from the hard sciences and engineering, e.g. hydrology, agriculture, civil engineers, etc - those skills were always in demand. In the end, there's a reason why development is often done so badly - they practically let anyone do it.

This guy built a classic run-of-the-river scheme, i.e. the main stream is not dammed. He built an intake channel off the main stream that flows into a forebay, where the penstock and piping system is installed. He also talks about putting a trash filter in the forebay, again a pretty standard thing to do.

Run-of-the-river schemes probably have the lowest impact on the local environment. I'd imagine very few fish would go into the intake channel.

As a grid engineer, I find the Danish case to be somewhat unique because of their strong AC and DC interconnections to Germany and Sweden. Basically, they can free-ride on the large european grid (UCTE), which stretches from Portgual to Russia, to maintain network stability while increasing domestic wind penetrations above 50%.

I think a more interesting case study is Ireland, which has far weaker interconnectors to the UK and operate mainly as an island network. Like Denmark, they are also trying to integrate large amounts of wind (a goal of 40% by 2020, which is equivalent to over 6GW peak), but unlike Denmark, Ireland also have to deal with the resulting stability issues.

An EirGrid engineer I spoke to recently mentioned that frequency stability is already a big issue for them. The main solution proposed in a 2010 study [1] amounted to maintaining a sufficient operating inertial reserve, which would potentially mean curtailing wind generation at times. In the future, I would look to Ireland rather than Denmark for solutions to integrating more wind into the grid, because they are already at the pointy end of it.

[1] EirGrid Facilitation of Renewables study, http://www.eirgrid.com/renewables/facilitationofrenewables/

If you look at the major blackouts around the world over the last 10 years, a large chunk of them were caused by a variation on a similar theme:

- The network is heavily loaded pre-blackout (most large networks are these days)

- A major interconnect trips out (e.g. from a tree strike or protection maloperation)

- The other line is out of service (e.g. for maintenance) or is taken out by the same event (e.g. by storms)

- Power flow is redirected through other weaker interconnects causing voltage instability

- Cascading voltage collapses take out the network

Examples: India 2012, European blackout of 2006, Indonesia 2005, northeast blackout of 2003, Italy 2003 (a bit different because the operators cocked this one up as well).

The point is that you can target one or two major transmission lines (usually in the middle of nowhere) and bring down the system. A coordinated attack on several major interconnects could really cause some damage.

I'm one of the many guest workers in Germany. I'm not an EU citizen and I didn't speak a lick of german when I got here. I'd say over half the other engineers in my company are non-german, and most of them couldn't string a single german sentence together when they started either (some still can't!).

The company has had an open engineering position for a german national (or at least native speaker) since before I started almost 2 years ago... and still haven't filled it, even after half a dozen foreigners have joined.

So I can sympathise with the idea that Germany is struggling to find skilled technical workers. It's telling that a company has to resort to hiring a bunch of non-german speaking foreigners and sponsor their visas and so on. I'd find it hard to imagine a French or Italian firm doing the same thing.

The article alludes to the fact that the Mittelstand firms are spread out across the countryside and many are headquartered in small villages. This is true and could perhaps be a big factor in why they find it hard to attract people - I mean, how many skilled engineers are willing to live in the middle of nowhere? I chose to work to here because the company is one of the leaders in a niche specialist field, and I certainly don't regret the decision given the amount I've learned so far. But frankly, I'd rather live in a bigger city.

By virtue of their tight bundling, all composite multi-core cables have high capacitances. However, onshore AC transmission is typically run on overhead lines, with large line spacings and thus capacitance isn't as much of a problem.

"I was trying to memorize equations, and engineering’s all about the application, which they really didn’t teach too well" - Matthew Moniz from the article

He's absolutely right, and no first year design project is going to fix this if the rest of the degree is structured the same as before. I think the bigger problem is simply that university is the wrong way to teach engineering.

I graduated 10 years ago with an EE degree (power systems major) and have worked in the field since, so I'm going to pick on EE here. During my degree, I was mainly taught theoretical models of electrical systems, interspersed with contrived lab experiments.

Very few 18 year old kids are going to appreciate the standard equivalent circuit of a transformer or a synchronous machine, less so more abstract things like Fortescue's symmetrical components. Because no kid has ever had much of a chance to look at power transformers or machines in service. Sure, they've seen power lines, but how many have seriously looked closely at them? Or maybe even asked themselves why there are strange, ceramic looking things connected to them?

So the first time I saw a real synchronous machine was in a lab. But looking back, it was really a contrived situation, with everything looking like it came out of a Bob the Builder toy set, alligator clips and all. You would never see a machine hooked up like that in real life... but only work experience has taught me that.

When I graduated, I was clueless. And in the intervening years since, I came across countless graduates just as clueless as I was. It's almost surely a systemic problem. I only met one graduate who really had a clue, and it turned out that he was a qualified electrician before he did his degree.

Which brings me back to my point - IMHO the pure university system is the wrong approach to train engineers. Their premise is that you learn the theory first, then apply it in context after you graduate (with some lame attempts at "practical" teaching in between). This works for some, but I'd wager that for most, it's a waste of time and you'll end up having to learn it all twice.

I would prefer to see a hybrid apprenticeship - university system, similar to the way you train tradespeople, but with more coursework components. The work is aligned with the study (or at least the student-apprentice gets exposure to real environments) and modules don't necessarily have to be done in a set order. It could potentially even be set up on a competency basis (like Western Governors University).

I don't know if this scheme would work, but I sure know that universities conceived for training academic researchers do not do such a good job of training practicing engineers.

Okay I'm not defending the IEEE, but I'd like to inject some data into this discussion. From their latest annual report, in 2010 their revenues and expenses from periodicals was $134.65m and $120.63m respectively.

Lord only knows how they spent $120m to curate a bunch of journals and magazines, but the $14m profit is less than what they earned in investment income ($25m).

There is no legal basis for kicking Greece out of the euro. None of the relevant treaties have any provision for member states kicking another state out of the monetary union or a member state leaving of their own accord. I think there is a provision for a state to leave the EU, but only of its own accord and not by force, and they'd have to leave the EU not just the monetary union. So if its going to be legal, there'd have to be some amendments made to the treaties.

Of course, Greece could just turn its back on the treaties, ditch the euro, re-instate the drachma, convert all debt obligations to drachmas and go on its merry way devaluing the drachma against the euro. But they would no doubt face retaliation from an angry EU and will likely get hit with stiff trade barriers on Greece exports as punishment (in which case devaluing the drachma would have limited effect).

I think the more likely outcome will be some kind of debt restructuring, a situation where the foreign bondholders (who own almost 60% of the debt) will have to finally face up to their losses.

Default has a pretty loose definition, but what will most likely happen is some kind of debt restructuring (which people will interpret as a default), e.g. extending the maturity on bonds, writing off some parts of the debt, paying out a lower coupon rate, etc. Basically, german and french banks (who hold a big chunk of the debt) will lose a pile of cash.

I think it's a credit to the CS community that computer scientists actually ask this question.

I'm an electrical engineer and I left a career in oil/gas to do something "good" with my skills. A few of my former colleagues used to ruminate about whether they could be doing something more meaningful in their professional lives. Of course, there were also those who had no qualms whatsoever about the oil/gas industry, that it was a noble goal to provide the world with energy (and FWIW, this is a perspective that I respect even though I don't share it).

But the overwhelming majority of my colleagues hadn't even thought about it, nor did they really care. The job afforded them a very comfortable living, enough for them to fulfil most of their material aspirations. And frankly, that's enough for most people.

Of course, existential crises probably happen to these people once in a while, but only to be forgotten the next day. My suspicion is that in the main, most people just aren't very introspective. So I applaud the CS community for actually bringing this issue up and discussing it openly.

I'm an electrical engineer (power systems) by background and this is my little project - an automated cable sizing tool to IEC standards. A little background to how it was conceived:

Over the years in the design consulting / engineering industry, I noticed that cable sizing was either done by hand (tedious) or by integrated power systems packages (overkill). Cable sizing is a pretty well defined task with all the requirements set out in international standards (e.g. IEC, NEC, BS, etc), so there's really no reason why it should be done manually. And yet there aren't many standalone tools to do it. So I wanted to make a simple tool that was also completely transparent - all the information is there for you to check the outputs yourself against the standards should you desire.

I'd like some feedback on how useful this is and whether or not its clear enough to non EE's. Thanks!

That's true ONLY IF America started to produce textiles. But the textbook trade models (e.g. Ricardian, Hecksher-Olin, etc) all pretty much recommend that America should just stick to producing fish and fur to maximise its gains from trade, and forget about a homegrown textile industry.

"When I spend $100 on materials and labor, and I can -- at best -- sell the result to you for $80, $20 of wealth is destroyed. Why? Because the original $100 came from the creation of wealth equal to that, and I transformed that $100 into something that is now only worth $80. Literally, wealth was destroyed."

In your example, no wealth is actually destroyed in the global sense, only relative to your bank balance (where you lost $20). Most people would call this a negative return on investment rather than wealth destruction.

The $100 you spent on labour and materials doesn't disappear into the ether - it is transferred to the people who worked for you and manufactured your materials. They could well do much more productive things with the cash than you and raise the wealth of the society accordingly.

Granted, Microsoft may not have used their resources particularly efficiently and there are indeed opportunity costs with their strategy, but who's to say that these resources would've been optimally used?

If you've ever been to business school, then you'll know that most of the students clearly don't give a shit about the course material. I mean what kind of 18 year old kid would care about the pricing of a zero-coupon bond?

Fair enough, although I've never actually seen an AC/AC converter without a DC link in real life (the closest thing probably being a direct torque control VSD).

Switch-mode power supplies are essentially DC rectifiers, which will never ever give you 110V, 60Hz (an AC voltage).

One way to change the frequency of a power source is through a double-conversion type AC/AC converter, i.e. 1st stage is a rectifier which outputs a DC voltage - this is an arbitrary DC voltage, but it needs to be within the voltage tolerances of the 2nd stage - an inverter, which reconstructs an AC voltage from the DC link to whatever frequency you like.

This setup is also the basis of most UPS systems btw, with a battery connected in the DC link.

In statistics, you're supposed to come up with a statistical model first before running regressions on the data. But quite a few papers I've read (especially in finance) seem to go the other way around, i.e.

They run regressions on a data set, adding and subtracting independent variables until the t values and standard errors start looking good.

Then they construct the linear model, assume the Gauss-Markov assumptions and sometimes (though not always) try to explain the causal relationship between the variables.

This is obviously very wrong and nobody has any clue what the distribution of the least squares estimators to these models are. But I've seen plenty of examples of this, which is enough to void the results of the paper (even if the model they come up with is somewhat plausible).

Is there any way to change the settings so that I get sent an email once every 2-3 days instead of daily? I don't fancy writing a daily journal, but a twice-a-week diary would be sweet.

Do kids actually learn about power factor?

That point aside, while an intuitive understanding is nice, power factor is really a mathematical construct arising from the multiplication of two out-of-phase sinusoids (i.e. voltage and current).

I think to really understand active / reactive power and pf, you need to look at the maths. It's really not that difficult. And you certainly don't need to know vector calculus, merely high school trig.

It's also worth mentioning that voting is mandatory in Australia, failure of which will result in a fine or legal action (if the fine isn't paid). The fine is paltry - I was fined $25 about 10 years ago when I was out of the country and didn't vote.

Couple mandatory voting with the vast disinterest in politics by the majority of the electorate, and you have the Australian government. I sometimes wonder why our society isn't in more disarray.

I think there are various levels of "badness" when it comes to theft:

1) When you steal your neighbour's car, that's the worst because you're completely depriving your neighbour of the exclusive use and enjoyment of their property.

2) When you copy a digital product that your neighbour has created, and then sell it for profit without paying royalties, then that's pretty bad, but less bad than 1). Your neighbour can still sell / use their works but you have deprived him/her of exclusivity in selling their works.

3) When you copy your neighbour's digital product for your own enjoyment, then you are depriving your neighbour of a potential sale. Still bad, but less bad than 1) and 2). It's only when everyone does it and your neighbour loses all their potential sales does it become as bad as 1), i.e. tragedy of the commons.

I'm not at all suggesting that reproducible digital products are public goods, but they share some similar characteristics in the context of piracy that I think are interesting, namely non-rivalness and non-excludability. And as mentioned before, a tragedy of the commons scenario occurs when everyone (or lots of people) engage in piracy, just like what happens when all fishermen over-fish a lake.

This obviously suggests that some kind of regulation is needed, which I suppose is the point of copyright laws. But what I'd like to know are the economic effects of piracy, for example:

What is the "piracy elasticity of demand"? In other words, how does demand for a product change as piracy levels change? Obviously as piracy levels approach 100%, the demand would be near zero, but what are the changes between 0 to 100%. Of interest is if there are any "sweet spot" levels of piracy where demand actually goes up.

I suspect there's a possibility that at low levels of piracy, the positive effects from things such as word-of-mouth marketing, early adoption / user familiarity (particularly in software) and try-before-buyers could offset the negative effects of lost sales. (Has anyone done this research?)

Anyway my point is that I don't think all theft is created equal, and that some kinds of theft do not dis-incentivize producers as the OP has asserted.

More discussion here: http://news.ycombinator.com/item?id=382899

I agree it's a great site. Kalid is also a regular contributor on HN.

In defense of MBAs, they can actually be useful IF you already have domain knowledge. The problem is when MBA graduates with no domain expertise are unleashed to manage and lead a business. It's dangerous even in small markets, for example the Stanford MBA trying to sell condom keychains (http://www.storylog.com/how-my-start-up-failed/)

I work in a multi-disciplinary engineering firm. In the past, the project managers were typically promoted from the senior engineering ranks (like a lead mechanical engineer). About 5 years ago, the company decided to create a separate discpline for project management. So graduate engineers would go straight into assisting project managers, and then becoming one themselves without ever working on a project as a discipline engineer.

Now we have to deal with project managers who've never actually worked on a project and don't understand how projects really get done. They only know how to manipulate cost and schedule spreadsheets and set unrealistic milestones.

The best project managers I've worked with have had extensive experience working on projects in their discipline, then supplemented these skills with an MBA or similar such management training.

To me, one of the interesting things about how mathematical results have been discovered historically is that they have often been accidental (or intended for something else).

eg. eigenvalues were originally looked at for use with quadratic forms, but people later found that its interesting properties (orthogonality, symmetry, etc) were useful for lots of other things.

There is often also a large gap in time between when a mathematician playing around with a problem discovers an interesting result and when it is actually applied to something useful.

eg. Euler's law regarding complex exponentials was developed around 1740, but it wasn't until around 1807 when Fourier used it in harmonic analysis and 1897 when Steinmetz started applying it to electrical engineering.

But you wouldn't know that from reading a textbook, which is (understandably) arranged in such a way that omits historical context and why people bothered to study it in the first place. Most linear algebra books are classic examples of how to introduce abstract topics without any context.