It appears that not all Y2k bugs have been fixed; one was found in NASA's handling of raw climate data.1 After the corrected figures2 have been released, it turns out that 1934, and not 1998, is the warmest year on record. Why a Y2k bug would have changed the readings for 1998 I am unsure. It is very puzzling, but the new data linked to does in fact seem to state that. Still, I wonder if the author is not misinterpreting something (though I would of course love to be able to state that our warmest years were so far back).
UPDATE: 2007-08-17: As I watch the confused reporting on this, I see very little worth linking to. Still, it seems to be that this was not a Y2k bug, though the fact that there was a bug of some sort, and that it has been corrected, is not challenged. Many are also being very quick to challenge the idea that the ranking of various years matters at all.
Mr. Michael Asher. "Blogger Finds Y2K Bug in NASA Climate Data." Daily Tech. 2007-08-09. http://www.dailytech.com/Blogger+finds+Y2K+bug+in+NASA+Climate+Data/article8383.htm ↩
NASA. "Contiguous 48 U.S. Surface Air Temperature Anomaly (C)" http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt ↩