Carl Zimmer has a nice comment on the warts of science: It’s Science, but Not Necessarily Right

Scientists can certainly point with pride to many self-corrections, but science is not like an iPhone; it does not instantly auto-correct. As a series of controversies over the past few months have demonstrated, science fixes its mistakes more slowly, more fitfully and with more difficulty than Sagan’s words would suggest. Science runs forward better than it does backward.

One thing Carl doesn’t get into is that replication isn’t the only way to test results of an experiment. Since repeating an experiment is unlikely to result in publication, what is more likely is for a researcher to change the experiment or at least the emphasis of the experiment rather than simple replication or refutation. Success will depend on the initial discovery being true, but the results will still be novel and publishable. However, that really only works if the original experiment was correct — if the investigation doesn’t pan out, you are still faced with the problem of trying to publish something that is not deemed “interesting” by the journals (another issue raised in the article).