What's that picture on the save icon?

The antiquated save icon should probably go away now.  Younger people may not even know what that picture is supposed to be.  In case you don’t know, it’s a floppy disk.  They were plastic squares the size of your palm that held a pitiful amount of data.  You could fit thousands of them on a USB stick.  They failed all the time and I don’t even install them in my PCs anymore.  I’m not sure what tiny, obvious picture should replace it.  Hard drive pictures are generally kinda technical and mysterious looking, so they don’t obviously mean “save”.  Funny how things you assume to be obvious are actually pretty antiquated.  I thought this one was obvious until I saw that someone on the internet was asking the question.

When Test is not the same as Prod

One time, at work, a program I wrote resulted in us having to restore our production database from backup. It turned out there was a process active in the production system that wasn’t in the test system. When my program cleaned up a table, this process started trying to reschedule old jobs as they were deleted. After reading this article about some errors that came up during the Apollo moon landing, it looks like I’m not the only one.

It's Homeopathic Awareness Week!

I’ve been reading Trick or Treatment, which is a book about testing alternative medicine. It starts off with an interesting account of the death of George Washington (via bloodletting) and the history of mainstream medicine. (For a long time, mainstream medicine was just ancient Greek medicine that nobody had ever bothered to question and mainly involved bloodletting. It was horrible and ended up killing more people than it helped.) Then, one day, a dude thought, “Hey, maybe we should actually test this stuff.” Everybody else got mad and sued the dude. The book talks about acupuncture and chiropractic therapy, which are both the manipulation of magical energy (chi and “innate intelligence”) in the body that “can’t be detected through any physical means.” It talks about homeopathic medicine, which some German guy came up with in the 1700s (before we even knew about germs) and claims that the more you dilute homeopathic medicine, the more powerful it becomes. So, super-diluted homeopathic medicine (water) can cure almost anything because water has a “memory”. (Who knew?) The descriptions of chiropractic manipulation were just downright scary…having some guy shoving your spine around using 1800s pseudoscience. Sounds like a plan.

But the book also points out that these things actually help people feel better. How can this be? It turns out that if you really expect something to make you feel better, then it may do the job. The example cases were pretty amazing. Because of a lack of morphine at a military field hospitals, an anaesthetist in World War II, Henry Beecher, would sometimes inject saline into a patient and say he was giving a powerful painkiller. The patients relaxed and showed no signs of pain. The book also points out that there is great variety in the quality of studies out there. There are many ways you can cheat in a study…even unintentionally. Biased researchers who want to make their theory come true, lack of any group who thinks they’re getting the treatment but aren’t really (to check for placebo), lack of reproducibility, stacking the deck with different types getting the treatment and the placebo. It takes careful examination of a study to make sure it’s of good quality and not cheating in some subtle way.

So what’s the harm in these treatments if they make people feel better? For one, people may think it’s working and forego something that might work better, as in the case of malaria or even something as simple as eczema. Another problem is the cost. It can all be very expensive for no real reason. As an example, a single duck’s liver goes into making millions of dollars worth of oscilicoccinum (a homeopathic flu remedy). (Remember, the more diluted, the more powerful.) That’s an expensive duck!

The book points out that clinical studies aren’t just a conspiracy to take down alternative medicine, but that such studies also take down conventional medicine theories that turn out to be invalid and even validate alternative medicine theories that turn out to be useful. The book is willing to admit when good clinical trials show a remedy to be effective, as in the case of some herbs that are actually effective in some cases…not because a bunch of anecdotes say so, but because of good testing.