this proves little (Score 2, Insightful)

The article in question was cited 4300 times. That is a lot. This would suggest this must be a fundamental paper for that particular field. How many times has the paper been discussed in classes, discussions, literature clubs, etc.? If so, the scientists are probably very well acquainted with the work, with out having the paper on their desk while typing in the reference. You can easily grab another paper that cites the original, with out digging through your file cabinet full of papers. Managing references can be a huge task. I have a small collection of papers, but even this is over 250. I know others with over 1,000. Any yes, we have read every one of them. That doesn't guaranty I will pull the paper out of my drawer every time I cite it. Many others in this discussion have also mentioned inaccuracies in the databases. It does happen, if you don't agree talk to your local inter-library loan. Contrary to the media's perspective, you should never believe what you read until you have tested it yourself or others have confirmed the work. Didn't Einstein say "Believe Nothing"?