There is no discernable consensus on the format of letters of recommendation in the field, or general rules about standards among letter writers. Any generalizations that may apply probably apply equally to letter writers in all academic fields. For example, it is probably safe to say that most people tend to be charitable by way of avoiding criticism but conservative about strong praise. However, some are prone to extremes. For example, there are cases where a letter writer will exaggerate the global importance of his or her own line of research, and this carries over into evaluations of the research of others - positively, for people working in the same area and negatively for others. Other observed extremes include people who tend to put down anything that is ``not invented here'', or who see everything in black and white - i.e., as either great work or junk. In the recent history of the FSU CS Department the combined judgement of the candidate and letter solicitor seem to have succeeded in selecting reviewers not prone to such extremes.