Tuesday, 24 June 2008

Death by a thousand blogs

The fact that anyone who has an internet connection to the outside world can publish anything they like doesn't mean that everything published on the net is good. On the contrary, it's likely that as the fraction of the world's population with internet connectivity grows, more drivel will be published -- unless there's a mechanism to edit or select content. For example, although anyone can edit most material on WikiPedia (material subject to malicious edits, or wars over content does get locked down), the fact that the information is in one place on the whole makes it possible to arrive at some sort of reasonable standard.

When it comes to blogs, though, there is nothing to stop anyone from posting whatever they like (aside from laws on libel, copyright, and anything that applies if you live in a police state).

So if you try to find out something, it's possible that you will find a slew of drivel -- especially if there's a campaign going to push a point of view.

Let's try a few experimental searches.

First, the claim that Einstein said humanity would die out if bees disappeared... Search for the words Einstein bees. What do you see? Of course the internet is a moving feast so the hits you will get will not be the same as mine. What I found was a fair number of articles putting the case that this quote was a myth, as well as a few that treat the quote as fact. In this case, it's not hard to determine that the quote should at best be treated with suspicion.

Next, try this one: bumble bees can't fly. Again, there's a fair mix of articles, and it's not hard to arrive at the conclusion that the original story may (possibly) not have been a myth, but certainly has become a bit garbled. A bumble bee's wing area is insufficient to generate lift (taking into account its weight) but that calculation is based on the theory of fixed-wing aircraft. A bumble bee's wings aren't stationary, so the mathematics doesn't apply.

Here's another: HIV causes AIDS. This time the majority of the hits on the first page (when I did this on 24 June 2008) were articles supporting the conventional theory, with a small number opposing – the group who for whatever reason claim that HIV does not cause AIDS.

So it seems we have a general pattern: while there is a fair amount of garbage or controversial material, you get a good balance and can find the most plausible position fairly easily. Relatively few authoritative-looking sources are making strong claims that are hard to dismiss, against the "correct science" position.

How about this one? Search for passive smoking harm. This time, while the majority of articles agreed with the conventional position, I found some surprisingly vehement articles in mainstream media, not just amateur blogs, pushing the line that the science has to be wrong.

Next, let's go to a more current issue, climate change. A couple of searches will illustrate the point. Try climate models fail to predict. Now this one is admittedly a bit different from the others by addition of the words "fail to". But the result is startling. Almost the entire first page of hits is articles claiming that climate models are not able to predict future climate change. Take out the word "fail" and you do get a very different result. With that in mind, I tried adding "less" onto the end of the smoking search: passive smoking harmless. The result? A slew of articles claiming that environmental tobacco smoke was harmless, research to the contrary was fraudulent, etc.

There are two questions that arise out of this experiment. How is the ordinary person with no training in searching to arrive at a reasonable mix of articles? How is someone without a research background to tease apart the mythology from the worthwhile content?

Taking the climate change one again, I spend a good fraction of my blogging time debunking climate change myths. The claim that models have no predictive power is only one of these (in fact, the IPCC validates the models in their previous reports by comparing them against subsequent measurement). Another is the claim that solar variations (search for sun explains all climate change) are sufficient to explain all climate change. Again there is a mix of articles, including some that clearly overturn the claim. This time around, bizarrely, if you change the search to the negative, sun does not explain all climate change, you get a higher fraction of hits pushing the case that climate change is purely down to the sun.

So what's the take-home point from this?

Blogging is not science. Neither, for that matter, is journalism. Blogging seldom is even as good as amateur journalism; very occasionally a whole lot better. Whatever the case, beware of following the line of least resistance, and only reading the material that comes up in the first page of searches. It's not that hard for a small number of people (possibly with an agenda; now who could care so much, I wonder, about confusing people about how harmful tobacco is?) to generate a lot of material, aided and abetted by the gullible who copy their line.

Information on the net is free, but so too is junk. Making life-and-death decisions based on a web search without digging deeper to understand the underlying science, whether it's how to tackle the HIV pandemic, how to deal with the health threats of tobacco or what to do about climate change is silly. Yet many people seem to do exactly this. South Africa delayed its response to HIV by almost a decade. Progress worldwide against public smoking was delayed even more. And the rate of progress on climate change, it appears, is more in the hands of the blogosphere than of informed decision-makers.



Anonymous said...

Yup. Try the same searches in Scholar and the results are better. Few do.

Try the same searches by asking the reference librarian at the local library, and few have ever even been there, let alone know there is a professional available to help them.

Online? hundreds of websites claiming to be reference librarians and tutors. Who can one trust?

Philip Machanick said...

The sad irony of all this is that climate inactivists call themselves "skeptics" when of course a true skeptic would dig deeper and find there is no solid foundation to their position.

Hank, to answer your question, what you need to trust is your own ability to be a skeptic, and that means tracking down primary sources or educating yourself sufficiently to cut through the opinions and get to a reasonable position based on verifiable fact and good science – what journalists ought to do but sadly many are suckered into believing that astroturfing is a genuine grassroots movement.

Many governments think that putting computers into schools is a replacement for sound teaching. Learning to distinguish BS is actually an increasingly important skill as web publishing in various guises including blogging becomes easier. Putting computers into schools without a sound background in BS filtering is a big mistake. Putting computers into classroom doesn't mean good teaching is less important; it is actually more important than ever before.