Non-biologists look away for a couple of lines.
For some reason I got distracted by left-handed DNA earlier. It's a bit of a science geek joke, with a serious point.
Okay, you can look back now.
I then wanted to see what else Tom had to say and came across his Errata & Corrigenda page. And he makes the point very nicely that scientific research progresses through the identification of mistakes and falsification of hypotheses. Additionally, most research claims are false and are corrected by further experiment. There is nothing wrong with this; the observation-hypothesis-experiment cycle is how we do science, and some of those hypotheses will be wrong.
To bring this up close and personal, I published a paper a couple of years ago (in Nature Structural Biology as it was called then) that claimed a hypothesis consistent with the data we had at the time. A year later I published a much better paper (in a less sexy journal!) after my further experiments showed that the original hypothesis was wrong. Did we retract the first paper? Of course not; the work was good and it was merely our interpretation that was wrong at one point. That's how science proceeds, and how we learn about the world around us.
I can not stress enough that I had no particular regard for that first hypothesis. I thought it was a nice little story and it fitted the data. But no one was paying me to find that conclusion, and I had no especial emotional attachment to it. So when I did the next experiments and realized what was really going on I felt little more than an 'oops' moment and moved on. (It makes a nice story for seminars too; students appreciate hearing about it).
That emotional attachment to a pet theory does afflict some researchers, but it is more a reason why you should be particularly wary when an interested party (tobacco companies, pharma houses) funds research, or when a social or political agenda is driving it. This happens more in social 'science' and in totalitarian regimes of course.
So, yes, Tom's Errata page is a useful insight into the philosophy of science. But then he goes and blows it by picking on Creationism and ID. And my gripe with this is that any sane person can see that CS and ID are not, and never will be science. They are politically-driven movements that do not change hypotheses to fit facts; rather they change the interpretation of evidence to fit their postulates. ID/CS starts from a conclusion and tries to prove that by interpreting observation.
Which is not empirical and is not science.
It is also why, unless one of them comes up with a testable prediction, you will not find me discussing ID on this weblog. I do have another website in utero, where I might discuss why ID is not just non-science but bad theology, bad philosophy and just plain bad as well, but that is another story.
My positive last word on the subject, which might clue you in to my own position, is a quote from the comments on Henry's site:
it's okay to admit that ID is vacuous politicized pseudoscience and a pack of lies. Really it is. The actual Christian faith — all but its most deranged interpretations at least — won't be at all weakened by such an admission. The ID crowd never really represented the theology of most Christians to begin with. That's part of the point of all the debate.
