Rob's Blog

In The Cathedral and the Bazaar, Eric Raymond wrote about why open source software is superior to proprietary software. More people look at the code. Even if you don’t look at the code yourself, it can be nice to know you could. For example, I wouldn’t recommend anyone use proprietary encryption software to send secret messages they don’t want their government to read. How can you be sure there isn’t a bug or even a secret government key that will unlock your message? For the purpose of catching the real bad guys, of course.

With open source software, even if you don’t look at the source code yourself, you can tell that others are looking. You can see the bug reports and see the fixes made. If there was a back door in open source encryption software, someone would be shouting about it.

Oddly enough, the openness of open source software imitates the openness of scientists. They publish their research and other people can see exactly what has been done and check that they can reproduce it. If there is something wrong with it, people start shouting, and the next lot of research refines the knowlege until there are few bugs left.

In climate science there is a lot of software. Software is used to run statistical algorithms on data from thermometers and tree rings and other temperature proxies in an attempt to make some sort of coherent temperature history. This is important for the theory of anthropogenic climate change because it gives us a clue as to whether modern temperatures are extraordinary or not. But frequently research is published in a way that makes it difficult or impossible to reproduce what the researcher did. The details of the research are recorded in the software, but the software is not always released; it is proprietary.

There is no way of knowing what bugs it contains, or even whether there is a back door for the government to get its way. I wouldn’t recommend relying on the output of proprietary software to decide whether those proposing to restructure the world’s economy have got their facts straight.

Eric Raymond has made the same point.

We know, from experience with software, that secrecy is the enemy of quality — that software bugs, like cockroaches, shun light and flourish in darkness. So, too. with mistakes in the interpretation of scientific data; neither deliberate fraud nor inadvertent error can long survive the skeptical scrutiny of millions. The same remedy we have found in the open-source community applies – unsurprisingly, since we learned it from science in the first place. Abolish the secrecy, let in the sunlight.

Not only is it bad software, it is bad science. If other scientists can’t reproduce what you’ve done, then you might as well have not done it.

This entry was posted on Sunday, November 29th, 2009 at 3:18 am and is filed under Enviro-Mentalism. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.