Scenario: 300 Van versus 1 Million C'Tisians

[Evil Dave]

Uh-Nu-Buh said:

True, that's why they/we have citations. In a study done a few months ago, a random sampling of articles in both a regular encyclopedia written by experts and wikipedia were shown to have an almost comparable number of errors.

And you can't bother to look up this study? The citation is: Giles, J; Internet encyclopedias go head to head; Nature 438, p900-901, 15 Dec 2005. This study was savaged by critics for methodological problems, notably:

numbers of errors were counted, but not types or magnitudes. Both "The Mon is generally believed to have been created by a Mars-sized object colliding with Earth." and "The Moon is made of green cheese." were counted as one error.

no accounting was made for Encyclopedia Brittanica articles generally being longer and more comprehensive than Wikipedia articles. So the error count is nothing like signal-to-noise.

Here are a small sampling of excerpts. Note that they are all accompanied by citations in the article.

"One school of thought rejects the figures given in ancient texts as exaggerations on the part of the victors"

"A second school contends that ancient sources do give realistic numbers. "

"Dr. Manousos Kampouris argues that Herodotus' 1,700,000 for the infantry plus 80,000 cavalry (including support) is realistic for various reasons"

"On the other hand, Christos Romas believes that the Persian troops accompanying Xerxes were a little over 400,000."

The first one dates from 1929, the second from 1930, and the last two are from what appear to be popular magazines*. These are not useful scholarly citations. Archeology and speculation from 70 years ago is generally regarded as junk, and popular writing doesn't have to meet any criteria for correctness.

*: other than the Wikipedia hits, I can find the title of the first offered only by an on-line wargaming store, and the second doesn't appear at all. I'd expect even tiny scholarly journals to have websites under their own names.