Answers in Genesis BUSTED! points to New Scientist's reporting of the interesting story of contingent evolution in e. coli.
Fortunately, nowadays we have the resource of actual scientist-bloggers rather than being restricted to what passes for science reporting these days, because the New Scientist story drives me nuts. As much as I love being able to hold a magazine in my hands to read, this reminds me why I decided a few years back that New Scientist was too expensive for what I was getting. (I have boxes and boxes of the things. Is there a good charity that needs a whole lot of old science magazines that I won't go broke getting them to?)
Frankly, it's entirely typical of science reporting, and another example of why science reporting sucks, and why blogging by scientists beats it hands down.
Aside from the fact that the New Scientist report has the emphasis wrong, it's just a generally typical sucky example of science reporting, with distortions of fact and silliness. (To be fair, here and there it does an okay job of explaining some bits, but really, I'm no expert, and I could tell it sucked before I'd read about it anywhere else.)
[Actually, what used to really drive me to screaming fits of rage was whenever they let Ian Stewart start talking about statistics. The guy may be a reasonable mathematician, but his explanations of statistics sucked dogs balls.]
I highly recommend reading the Pharyngula
For goodness sake, even the abstract of the paper he quotes (jargon aside) is in many places better written than the New Scientist article.
Let me quote a paragraph from the New Scientist report:
"In the meantime, the experiment stands as proof that evolution does not always lead to the best possible outcome. Instead, a chance event can sometimes open evolutionary doors for one population that remain forever closed to other populations with different histories."
"doors ... that remain forever closed to other populations" is one of the more ridiculous statements I've seen from New Scientist, and they've had some doozys.
forever closed? forever??
Bullshit. If the "tricky part" - the enabling mutation - evolved once, in only 20000 generations (which then further evolved several times to produce the ability to metabolize citrate), it's obviously possible for it to happen again. The probability of it evolving is not zero.
The fact that Blount, Borland & Lenski couldn't get it to evolve independently a second time just means that probability is likely very low. Maybe extremely low.
But forever is actually a pretty long while. You might think it's a long time 'til lunch, but forever is such a really long time that any mathematically nonzero probability of it evolving in a finite amount of time (as was observed to occur) means that by forever, it will have evolved an infinite number of times. And, if you're not real good with the counting thing, take it from me that infinity times is just a tad bit bigger than zero.
And what's with that bit about "evolution does not always lead to the best possible outcome"? Which cryptotheist retarded version of evolution do they think they are arguing with there? All you have to do is take a look at a human eye (heck, just get a small light and grab a handy animal with a spine) to know that. The blood vessels of the eye lie on top of the rods and cones! They're in the way! Anyone can see them! And the nerves, too - that's why we have a blindspot. And, funnily enough, the light sensitive bits point backwards, away from the light. Our eyes are arranged ass-backwards!
(That it needn't have happened this way is obvious from squid and related creatures, whose eyes are not arranged "backwards" like ours. And there are hundreds of little examples of this sort of thing.)
Evolution is completely cobbled-together (i.e. contingent), and it's already completely obvious that it's so - to anyone with eyes. Well, and maybe a brain.
[based on a comment I made at AIGBusted.]