Here's what most people got out of Design Patterns: "blah blah blah blah SINGLETON blah blah blah blah". I kid you not. I've seen this so many times that it's become a full-fledged pattern in its own right; patterns need a name, so let's call it the Simpleton Pattern.

...

I'll close by saying that if you still feel the need to use Singleton objects, consider using the Factory Method pattern instead. It gives you all of the flexibility of the Singleton, with nowhere near as many problems. Using the Singleton is usually just a sign of premature optimization - I know you dread instantiating objects; you've heard it's way slow. But just relax, take a deep breath, and treat yourself to a few extra instances. You'll be amazed at how fast computers have gotten these days, and you'll be pleased at the extra flexibility using real objects gives you. Enjoy!

Don't do these because "it's quicker" (it's not). Don't do these because "nobody else will use this code" (your future self will). Don't do these because "it's a one-time script" (almost all production software started as one-time scripts). Just do not do them. At all. Ever. Unless you are mst there is no reason to do any of this. The language has moved on. There is a better way.

We might ask ourselves how we would be reacting if Iraqi commandos landed at George W. Bush’s compound, assassinated him, and dumped his body in the Atlantic. Uncontroversially, his crimes vastly exceed bin Laden’s, and he is not a “suspect” but uncontroversially the “decider” who gave the orders to commit the “supreme international crime differing only from other war crimes in that it contains within itself the accumulated evil of the whole” (quoting the Nuremberg Tribunal) for which Nazi criminals were hanged: the hundreds of thousands of deaths, millions of refugees, destruction of much of the country, the bitter sectarian conflict that has now spread to the rest of the region.

At BackType, we are heavy users of Hadoop. We use it to run computations on our 30TB datastore of social data. We've even open-sourced some significant projects that are built on top of Hadoop.

Unfortunately, Hadoop has problems. It's sloppily implemented and requires all sorts of arcane knowledge to operate it. We would be the first to try out a replacement for Hadoop if a viable alternative existed. In this post, we'll look at some of the darker aspects of Hadoop.

Det är förmodligen så man ska se på Spotify; det är ytterligare en livsuppehållande åtgärd för en industri som egentligen borde få somna in för gott. Det kommer heta att det är piratkopierarnas fel, man kommer skylla på ungefär allt som går att skylla på för att försöka skyla de egentliga orsakerna; en trasig och förlegad upphovsrätt och fullständigt föråldrade affärsmodeller inom upphovsrättsindustrin.

...

Spotifys affärsmodell är väldigt rätt, en Internetbaserad tjänst som ger dig ett musikbibliotek som är många gånger större än det du själv orkar sortera och indexera på egen hand, men det hjälper inte när de samtidigt tvingas hålla en obsolet och döende dinosaurie vid liv. Utan den dödvikten tror jag att Spotify hade varit markant mer framgångsrik, artisterna hade definitivt fått markant mer betalt och mycket av dagens kritik hade aldrig ens uppstått men med dagens beslut kommer kritikerna knappast bli tystare.

Another similarity between Japan's current crisis and the recent financial crisis is that the false risk assessment was largely due to the asymmetric distribution of social welfare and individual cost implied by more effective risk mitigation. Both Lehman Brothers and Tokyo Electric Power Company were able to increase their profits as long as the risk they were willing to accept did not materialize. Their management certainly benefited as long as everything went well. When crisis hit, however, the cost of the meltdown exceeded the companies' equity and thus had to be socialized.

...

As in finance, ensuring that the originator of a risk pays the cost seems to be the most sensible approach. If each nuclear-power plant was obliged to insure against the risk that it imposes on society (within and outside the country of its location), it would face the true economic cost of its activities.

So what's wrong with using the Mac as a development machine for Milo, a Python application backed by PostgreSQL and Redis (or any web project, for that matter)? Well, sacred cow, here come the spears.

More than 10,000 people have died in the Japanese tsunami and the survivors are cold and hungry. But the media concentrate on nuclear radiation from which no-one has died - and is unlikely to.

Worst of all, the reporting has put the focus entirely on the Fukushima nuclear plant. As worrisome as the issue is, it has been completely blown out of proportion, with talk of meltdown and massive destruction. The tragedy is that the victims of the earthquake and tsunami are all but forgotten at times. While the world turns away to ponder its own nuclear policies — which, for better or for worse, are far from urgent — people are starving and dying after having survived the disaster itself.

The nuclear plant story is exciting and dramatic and is easy to exaggerate, but it will quickly wear thin as the plant cools and the international public realizes that the fears were deliberately whipped up. By then, it could be far too late for many survivors.

A crappy old plant with inadequate safety features was hit by a monster earthquake and a vast tsunami. The electricity supply failed, knocking out the cooling system. The reactors began to explode and melt down. The disaster exposed a familiar legacy of poor design and corner-cutting. Yet, as far as we know, no one has yet received a lethal dose of radiation.

Some greens have wildly exaggerated the dangers of radioactive pollution. For a clearer view, look at the graphic published by xkcd.com. It shows that the average total dose from the Three Mile Island disaster for someone living within 10 miles of the plant was one 625th of the maximum yearly amount permitted for US radiation workers. This, in turn, is half of the lowest one-year dose clearly linked to an increased cancer risk, which, in its turn, is one 80th of an invariably fatal exposure. I'm not proposing complacency here. I am proposing perspective.