Paul M. Jones

Don't listen to the crowd, they say "jump."


Health-care reform's dirty little secret

The unspoken truth about Mr Obama's ... effort to reach universal coverage is that you may not be able to keep your existing health plan--at least, not at the same price. That is because paying for expanding coverage must involve capping or eliminating the tax exclusion currently favouring employer-based health cover. That single distortion of the tax code costs some $250 billion a year--the biggest kitty of money lying around in Washington. But tapping some of that inevitably means some Americans will see de facto tax increases.

via Health-care reform's dirty little secret | Democracy in America | Economist.com.


The jobless recovery

The weekly [initial unemployment] claims number can't seem to fall below 600,000. ...

This is bad news for many reasons, not least of which is what is suggests about the structural problems in the economy that are likely to persist for years. But the real danger is the threat joblessness poses to an economy that, at least according to most macroeconomic variables, is stabilising. As unemployed individuals exhaust their available savings they'll find themselves curtailing spending and defaulting on obligations, both of which contribute negatively to economic output. Sustained high unemployment also places a strain on state budgets. Faced with growing demands on unemployment assistance, states are forced to cut spending elsewhere. But this is procyclical behaviour, which may act to increase unemployment further, forcing additional budget cuts, and so on.

via The jobless recovery | Free exchange | Economist.com.


Buzz Aldrin’s Plan for NASA

I’m in no mood to keep my mouth shut any longer when I see NASA heading down the wrong path. And that’s exactly what I see today. The ­agency’s current Vision for Space Exploration will waste decades and hundreds of billions of dollars trying to reach the moon by 2020--a glorified rehash of what we did 40 years ago. Instead of a steppingstone to Mars, NASA’s current lunar plan is a detour. It will derail our Mars effort, siphoning off money and engineering talent for the next two decades. If we aspire to a long-term human presence on Mars--and I believe that should be our overarching goal for the foreseeable future--we must drastically change our focus.

Here’s my plan, which I call the Unified Space Vision. It’s a blueprint that will maintain U.S. leadership in human spaceflight, avoid a counterproductive space race with China to be second back to the moon, and lead to a permanent American-led presence on Mars by 2035 at the latest. That date happens to be 66 years after Neil Armstrong and I first landed on the moon--just as our landing was 66 years after the Wright Brothers’ first flight.

via Buzz Aldrin’s Plan for NASA - Scrap Ares I, Fast-Track Orion and Colonize Mars - Popular Mechanics.


Scalable Internet Architecture

This is not strictly PHP, but it is about scalability, and every PHP programmer *ought* to be thinking about this stuff.

Theo Schlossnagle of OmniTI (where I work as a web architect) has this slide deck posted about Scalable Internet Architecture:

View more presentations from postwait.

via http://www.slideshare.net/postwait/scalable-internet-architecture

(Aside: I joke that at OmniTI, my reporting chain looks like this: from God, to Theo, to my director, to me, to my subordinates.)

Developers (PHP and otherwise) should read the whole thing.

Some highlights, with comment and paraphrasing from me:

  • Slide 7: "Lack of awareness of the other disciplines is bad." Developers need to be aware of the advantages and constraints presented by everyone else: designers, DBAs, sys admins, network engineers, etc. Not being aware means making your own work more difficult (bad) or their work more difficult (also bad, especially in teams).
  • Slide 14: Craftsmanship and discipline: learn it, love it, live it. It's very difficult to be consistently attentive; don't let the fact that you cannot *always* be attentive stop you from *always trying* to be attentive.
  • Slide 29: Know the difference between "premature optimization" and "necessary optimization". This is hard.
  • Slide 48: The networking architecture is critical to scaling. Everybody forgets about the network because it's "just there" (until it's not).
  • Slide 59: "Scaling is hard, performance is easier. Extremely high-performance systems tend to be easier to scale, because they don't have to SCALE as much."
  • Slide 63: Combine this with slide 59, and you have the reason why you need to know your application responsiveness (i.e., benchmarks). As part of this you need to know the overhead imposed by your framework of choice, so that you know exactly how far you can optimize your application. This will help you decide where to spend your limited resources: on development to improve application performance, on hardware to improve system performance, or on adding systems for horizontal scaling.

No More TV For A While

After months of having 100+ channels with nothing to watch most of the time*, I realized my main use of TV was to have something in the background as I settled down for bed. At the same time, it was a big time-sink, because I would turn it on, flip through 100+ channels to find nothing, and then flip through 100+ channels again to see if anything new had appeared (usually not), and then again, etc.

So, to save $60/month and recover about 2 hours/day, I suspended my DirecTV service and bought a digital antenna for emergency viewing. (By "emergency" I mean weather emergencies.) I can still watch series episodes by web or DVD, but that has a well-defined beginning and endpoint (as opposed to turning on the TV and just surfing around looking for something).

I've been off TV for about 2 weeks now; we'll see how much longer it lasts. ;-)

(* Notable exceptions: AMC and TCM movie marathons like James Bond, Dirty Harry, old sci-fi, etc. I'm a notorious C-SPAN devotee but I can get that over the web.)


The Illusion of Government Competence

What do these stories all have in common? They all demonstrate that government organizations do not systematically make better decisions in the same circumstance than do private organizations.

Leftists like to argue that, by some magical mechanism, real-world politicians make better decisions, especially better economic decisions, than do private actors in the free market. They usually make this argument after either the free market corrects itself naturally or the government interferes. They then simply assert, without any possibility of empirical verification, that the magic government unicorns could have prevented the problem if only they had been given enough power to do as they wished.

Such arguments are clearly ex post facto. Leftists cannot predict market correction any better than anyone else, so clearly they don’t have a predictive model of the relationship between any particular regulation and any particular market correction. More importantly, when they do have the power, they often do nothing to address the causes of the correction. Last year’s financial collapse demonstrates this clearly.

via Chicago Boyz » Blog Archive » The Illusion of Government Competence.


Bush & Cheney Questions Staged and Pre-Arranged

Oh, sorry, I meant "Obama", not Bush & Cheney.

After the obligatory first question from the Associated Press, Obama treated the overflowing White House briefing room to a surprise. "I know Nico Pitney is here from the Huffington Post," he announced.

Obama knew this because White House aides had called Pitney the day before to invite him, and they had escorted him into the room. They told him the president was likely to call on him, with the understanding that he would ask a question about Iran that had been submitted online by an Iranian. "I know that there may actually be questions from people in Iran who are communicating through the Internet," Obama went on. "Do you have a question?"

Pitney recognized his prompt. "That's right," he said, standing in the aisle and wearing a temporary White House press pass. "I wanted to use this opportunity to ask you a question directly from an Iranian."

Pitney asked his arranged question. Reporters looked at one another in amazement at the stagecraft they were witnessing. White House Chief of Staff Rahm Emanuel grinned at the surprised TV correspondents in the first row.

The use of planted questioners is a no-no at presidential news conferences, because it sends a message to the world -- Iran included -- that the American press isn't as free as advertised.

via Dana Milbank - Washington Sketch: Welcome to The Obama Show' - washingtonpost.com.


Insurance as a Prisoners' Dilemma, Bryan Caplan | EconLog | Library of Economics and Liberty

When the secretary informed me that, "Insurance doesn't cover this treatment," I was positively delighted.

Why, you ask? It's not envy - I don't mind if other people have better insurance than I do. The reason for my delight: As soon as she said that no insurance companies covered this treatment, I knew it would be reasonably priced!

via Insurance as a Prisoners' Dilemma, Bryan Caplan | EconLog | Library of Economics and Liberty.


No Such Thing As Multitasking (Redux)

In most jobs, one needs to be able to switch quickly between tasks, but this is easier said than done. Thought processes from the first task are hard to shut off - “attention residue,” in the words of the author of a new study - and can interfere with the second task. Not only can attention residue impede performance on the second task, but attempting to make the transition when the first task is unfinished - or when the first task did not have a tight deadline - makes the problem worse. The only transition that didn’t suffer from attention residue was when the person believed that the first task had been finished under deadline pressure. In other words, put yourself and your co-workers under the gun, do one task at a time, and then move on.

via Uncommon knowledge: When friends make you poorer - The Boston Globe.