Sunday, January 20, 2013

The concept of safety

I work in the nuclear industry. This is an industry I chose to work in because I think it is important to humanity's prosperity. So you will not be surprised when I get somewhat irritated by this often-repeated rational that we shouldn't be using nuclear power because you can't guarantee that the reactor won't explode.

That is absolutely correct. You cannot guarantee that nuclear is 100% safe. Of course, you can't guarantee that a car is 100% safe, but that doesn't stop basically everybody in Canadian society from riding in motor vehicles on a regular basis. Getting into a private vehicle is probably the riskiest thing that you can do on a daily basis, and yet not only do we accept it, we barely regulate it. We accept a level of risk in our personal lives that would never be acceptable for an industry.

Nothing can be made perfectly safe, simply because nothing in the world is perfect. As builders and creators, we accept this and instead determine an acceptable level of risk. A nuclear plant has to meet more stringent safety standards than a candy factory because the consequences of a failure are higher. This is why you have only seen a small number of events in 50 years of power reactor operation.

(side note: sugar factories can explode and it's fucking impressive to see).

Obviously we have made mistakes. The Fukushima Daiichi reactor disaster is an example of where we fucked up. The people who designed the plant were unwilling to believe that anything could happen to cause it to fail, and ignored evidence that they were not planning for the worst tsunami event possible. You still have to see things in perspective though: the earthquake and tsunami that caused the plant to meltdown killed over 15 000 people. The meltdown has not killed anyone yet, and while there aren't any great estimates for long term deaths caused by the radioactive release, the highest legitimate estimate I've seen is around 100. Perhaps the area housing should have been built to withstand a tsunami.

I will enter into a bit of rhetoric now, but it still needs to be said.

If we stopped doing things when something went wrong, we would not be surviving as a species right now. This is not meant to ignore the suffering of the victims of the Fukushima Daiichi disaster.  This is meant to say that our technological prowess is what allows us to support a population of 7 billion people. In 1984, a pesticide plant in Bhopal released a cloud of toxic gas over the town, exposing over 500 000 people. Death estimates range from 3000 to over 16 000. Yet we still make pesticides. Why? Because people like to eat, and crops are important.

Every year, thousands of people die prematurely from fossil fuel related disease (tons more sources, but I'm feeling lazy), but we still burn stuff to make power.

Hydroelectric dams threaten populations downstream from them. Depending on the dam, tens of thousands of people could die if it were to fail. This seems like a pretty big risk - after all, I can't guarantee that a dam won't burst. Maybe we should stop using hydro power?

I'm not trying to say that a nuclear plant and a hydro dam pose the same level of threat. The idea that I'm attempting to convey is that it's not as simple as Safe vs. Not Safe. Like everything else in life, technology has a lot of grey areas that we have to manage. It is not an option to stop producing electrical power. Very plainly put, society would collapse. That is not me being melodramatic. Without electricity we cannot support the world's population. Without it, the world's population would be decimated, if not worse. Not only do we have to keep making electricity, but we have to increase our ability to do so.

In fact, this can be generalized, and I make the following statement unequivocally: we have no choice but to continue forward in our technological drive if the population is to keep increasing.

You don't run from your mistakes. You look at them closely. You inspect the fuck out of every scrap of data you have and you make sure you never make the same mistake again, but you don't quit. Nuclear power has only been around for about 60 years, depending on how you reckon it. We've done a lot of learning since then, and we're getting better at it.

As we do with everything else.

Monday, December 10, 2012

On driving

I'm not going to claim that I've never complained about other drivers. I won't even say that I've never been annoyed at other people driving slow - it's happened a lot. I am, however, getting a bit annoyed at people losing their minds when people drive "too slow" in freezing rain, as happened this morning.

First, I'm going to address the hazards of freezing rain. Ready?
  • Water falls from sky.
  • Water hits ground.
  • Water freezes.
  • Frozen water makes layer of ice.
  • Ice is slippery, and makes controlling your vehicle hard.
Everyone with me here? Good.

Living in a city like Ottawa, they salt and plow the roads here. This means that the majority of the surfaces are generally not too bad when the freezing rain is light - again, like today. "Generally" is the key word there. Some of the road surface, for whatever reason, does not get turned to slush. So you get black ice patches. Even when the road is slush, snow, and water, hazard conditions are still present. The increase in vehicular collisions every time there is adverse weather is a fairly clear indication of that fact (secondary source).

It's almost like the roads get harder to drive on when there's ice falling from the sky. Weird.

So you get people out on the road who drive slower than they normally would. How much slower is a judgement call, in the same fashion that the speed you would normally drive at is a judgement call. As you have probably noticed, many people have different judgement. Slow speeds are aggravated by limited road infrastructure to create shitty commutes. None of this is surprising.

So now you have people complaining about other drivers being "bad" - presumably, because they are driving slower than you'd like. Now I ask, how much slower does one have to be going to be a bad driver? From an engineering standpoint, I would pick the speed that gives me a comfortable (5-10%) margin over where I would lose the ability to control my vehicle on the worst part of the road that I have to travel on.  I don't want to be the guy explaining to the cop that "the roads were just fine until that one patch of ice" while my car is sitting in a ditch / back of a pickup / front lobby of the retirement home.

Personally, if I had to pick someone to be a bad driver, it'd be the person in the ditch.

Now, the above is based on my own assessment of how bad the roads are, and my own ability to handle my vehicle. These are two separate items that are completely subjective. If I had a high opinion of other people, I would think they would choose my method to select their speed as it provides the optimum method of getting from point A to point B in a storm. Realistically, this is not the case, since more people keep ending up crashing their cars in storms.

Coming back to the concept of a "bad driver". So far we have established that "bad drivers" move slower than one would like. We have also established that this is due to differences in

A) Awareness
B) Handling ability

My personal opinion is that (A) is much more important than (B), probably because my personal experiences with car crashes have been because someone did something stupid when they weren't paying attention - I discount impaired driving for the purposes of this argument.

So what are the solutions to this? Bitch about it on Facebook? Effective. Alternate solution: More difficult licensing. This will have two effects: first, the drain on the system will be HUGE (you think waiting at the MOT is bad now?), with a commensurate increase in cost. On top of that, you will have many, many people losing their license.

That's not hyperbole. In 2009, there were 123,192 collisions that resulted in an injury, and 2,011 that caused at least one death. That's in ONE year. I'm going to assume that most accidents are caused by driver error, rather than mechanical failure (feel free to argue that one). There were just over 23 million drivers, and I'll assume that half of the drivers in collisions lose their license (it's not a terrible assumption; the odds and ends come out in the wash, so to speak). That ends up with a little less than a quarter of a percentage of the population who cause an accident EVERY YEAR.

If we assume that my magical new licensing system perfectly removes bad drivers, another way to look at those numbers is that about 75,000 people per year will not be licensed.

 But every time you do something dumb, you don't cause an accident. There needs to be some combination of factors, like another car getting in your way. I'm going to use an old industrial safety estimation here - the accident pyramid.





http://emeetingplace.com/safetyblog/wp-content/uploads/2008/07/clip-image002.jpg

This pyramid is sort of based on statistical knowledge, so it gives an okay estimation of the relationship between "Oh, shit, that was close" and "OH SHI-". Looking above, the number of fatalities is closer to 1/20 accidents that cause injury, but it's close enough for what I'm playing with. Let's assume that, based on my Highly Scientific (TM) reasoning, for each 30 of those accidents per year that caused an injury, there were 600 incidents where someone did something dumb and avoided the consequences. So now we're up to 1,500,000 people every year who should not have a license. If you remember the above, that's almost 10% of Canada's licensed drivers in a single year. And I'm willing to bet that it isn't completely made up people who's main problem is driving slow.

Obviously, this system does not count for repeat offenders, so the entire population would not lose their license. Still, you can see how that would quickly escalate. You're looking at a huge swath of the population suddenly unable to drive.

But of course, not you, because you're not a Bad Driver (TM). So by all means, continue to complain about people driving slow in the freezing rain.

Just don't be surprised when they take your license away.

Monday, May 21, 2012

Education, museums, and the power of the loudest voice.

By now I'm sure that most people in Ottawa are aware of the exhibit at the Science and Technology Museum, SEX: A TELL-all Exhibition. The reason you are likely aware of this is because there has been some "controversy" over the exhibit's content.

I'm not going to go into what is in this exhibit; check out the museum's website if you haven't already - and I mean that. If you are going based on what you have heard without seeing the exhibit or at least going to the source of the information, you run the risk of not having accurate information and worse, spreading this inaccurate information. [Anecdotal evidence] Mob mentality seems to take hold when people make Facebook posts based on limited data and all of their friends scramble to add their own pointless comments of "You said it!" and "I agree completely!" [/Anecdotal evidence].

Some people seem to have taken issue with what's in the exhibit. The museum received dozens of complaints about the content being "inappropriate" for children (although it was designed for the 12-and-up age bracket), causing museum officials to require children under 16 to be accompanied by an adult. I want to address this very succinctly: sex education is important. The public school system requires it, has required it for many years, and to argue that it is not appropriate for children is foolish and regressive. There are scholarly articles which support the claim that sex education reduces STI transmission, and that "abstinence only" education does not have a significant impact on teen sexual behaviour.

A spokesman for the Heritage Minister made another rather silly claim: that the Sex exhibit does not fit the museum's mandate to foster scientific and technological literacy. At this point, I'm going to make a statement that may be in conflict with things I have said in jest in the past: biology is a scientific field. Sex is a biological function. The fact that we associate a particular value set to sex and the issues that surround it does not diminish this in any way. Adding a sociological aspect to a topic does not remove that topic from its original basis. Indeed, it only adds to the appropriateness of the location - after all, what is technology if not the human application of science?

Some have suggested that this exhibit would be better located at another museum, and have even provided suggestions - all of which I disagree with. The Museum of Nature, while perhaps somewhat appropriate at first glance, is not the best fit: it concerns itself with "the natural world" and the history therein. It tends to focus on the world sans-humans, and to place the Sex exhibit there would be to remove the human aspect of it. The Museum of Civilization would have been appropriate if this were an exhibit exploring how different societies view sex and presented their values and beliefs - but that's not what this is about. (My absolute favourite was the suggestion I heard that it could be sent to the National Art Gallery - obviously based in rhetoric since it is so very silly.)

Then we hit the fall-back position of "this isn't about appropriateness, this is about tax dollars". That is just an obfuscating tactic; saying that an issue is about taxes is the same as saying one doesn't want to pay for it. Why would you not want to pay for something? Because you don't approve of it. Full circle, nothing new has been said and my time has been wasted.

So now I come to the part that actually really bothers me. There is no cohesive argument against this exhibit, so why is it that this issue has seen so much attention? Who has the power to make the museum change its admittance rules? Turns out that when enough people raise their voices through effective means, they get listened to. In this case, that's the dozens of people who complained directly to the museum, and the obnoxious idiots who manage to stir people up through sensationalism and misinformation. It bothers me because rather than respond to carefully reasoned arguments, the museum was forced to listen to a raving mob of people who have more interest in making their presence felt than in actually thinking about something.

And this just snowballs for me. Most of the people I know don't agree with this, so why isn't there a counter-campaign? Because it seems that most of my friends are more interested in complaining on Facebook about how stupid and unfair it is rather than actually doing something about it. If you want to be heard, you have options.

Stop bitching an exercise them. 

Wednesday, February 2, 2011

Metered Bandwidth

I'm a little confused about the issue with metered internet usage. From what I understand, the issue is that the CRTC is going to allow larger companies to pass on bandwidth caps to the smaller companies that are leasing their infrastructure. This means that everyone with access to a service provider that gives "unlimited internet" (apparently 250gb/month counts as unlimited - I have friends who would probably disagree with that) for a low price is going to see rate hikes if they want to continue to consume a large amount of bandwidth.

So consumers don't want to pay more for what they are already getting at a low price right now. Fair enough. Say so. The issue I'm having with this whole 'debate' is the labeling of it as "A Strike Against Net Neutrality!".

Well, no, it's not.

We aren't talking about IP blocking or bandwidth throttling - and these are very real issues that are being lobbied against by many of the same people who are running the petitions against the CRTC's recent decisions regarding metering. This is the meat of net neutrality, but pricing of your bandwidth usage is not. Unfortunately, you do not have a right to cheap media.

Which brings up the next point: the cheap, unlimited bandwidth provided by smaller ISPs that exists now is an artifact of artificially-introduced competition. At some point in the past (and I'm too lazy to look up the specifics right now, so please correct me if I'm wrong) the CRTC decided that there was not enough competition between ISPs, so they allowed the introduction of smaller companies that use the infrastructure of the existing larger companies. This doesn't strike me as something the big companies happily went along with - after all, they do invest billions in all the tubes that bring the internet to your house.

This is where regulating things gets a little difficult. It is hard to ensure that these smaller companies will be able to exist without giving them some large unfair advantage. That's why the CRTC is mandating this 15% discount that the larger companies have to give to the smaller ISPs - they want the smaller companies to survive. Obviously the CRTC wants to maintain proper competition - claiming that these changes in law will prevent smaller companies from competing effectively demonstrates either a lack of faith in the CRTC's ability to regulate the marketplace, or ignorance of the fact that they exist to do so.

Taking a different tack, moving to a pure metering system (instead of the pay-for-overages system that the big companies use now) doesn't make less sense to me. You just end up paying for exactly what you use. Is it an excuse for a rate hike? I would expect the CRTC to force them to keep it revenue neutral, and I've seen nothing to convince me otherwise. As for price gouging and the supposed 100x markup!!!!!1!: The blogs and articles that I've looked at have not shown me convincing evidence of that. Maybe it costs 100x the electricity cost, but not if you factor in the costs of doing business. Telecom is a capital-intensive industry.

Oh, and for those of you who keep comparing Canada to Europe in terms of cost? Have you LOOKED at the size of our country compared to our population?

If you're going to complain that pricing is unfair, show me the proof. If you're with a smaller ISP: I know that it sucks to have your prices skyrocket, but if you want to complain that you're going to have to pay more please make it clear that you are doing so.

I'm tired of people dressing up their personal issues as noble causes.