|This article is of interest to the following WikiProjects:|
In English, words ending in S do not take 'S as a suffix, but just '. Shouldn't this therefore be Farkas' lemma? Encyclops 03:28, 16 January 2006 (UTC)
- According to Apostrophe (mark)#Things to note, English is not that simple. -- Jitse Niesen (talk) 08:32, 16 January 2006 (UTC)
- Words that should have an apostrophe with no s are plural, eg the cat's feet. Plenty of words end in an s and still have 's when posessing something, eg the double bass's strings. This definitely should be Farkas's lemma, but I'm not going to change it because it'll just be reverted. Loads of papers use the (wrong) Farkas' lemma, so what's the point in fighting it. --RatnimSnave (talk) 11:35, 2 February 2012 (UTC)
- A comment to "Loads of papers use the (wrong) Farkas' lemma": In German, "Farkas' Lemma" is correct. Lots of authors are native German speakers: they just carry their notation from German to English without being aware of the fact that English rules for apostrophes might be different.
From Strunk and White's _Elements of Style_, a broadly accepted reference: "Rule #1: Form the possessive singular of nouns by adding 's. Follow this rule whatever the final consonant." One of the examples they give, "Burns's poems", is especially apropos here. Strunk and White give as exceptions biblical characters, similar to the ones described as exceptions in the above-linked wikipage on the Apostrophe. That page provides several sources on grammar (and even a Supreme Court majority!) that advise the use of 's in all cases, and only offer out-of-date sources or authors who just don't like it as support for the omission of the 's. Farkas was not a character in the Bible, so it seems there's just no support for "Farkas' lemma" (Or Bayes' theorem, btw). B k (talk) 04:12, 31 December 2016 (UTC)
Somebody in the know should check whether this quotation from the book "Polytopes, Rings, and K-Theory" by Winfried Bruns and Joseph Gubeladze (Springer, isbn 978-0-387-76355-2) is applicable for the Farkas' Lemma this article discusses:
"Separation theorems like 1.32 often appear as Farkas' lemma (see Ziegler). A far reaching generalization is the Hahn-Banach separation theorem."
The quote appears on page 21 of Bruns-Gubeladze, and i think that would be very reasonable to mention the relationship between Farkas and Hahn-Banach, as Hahn-Banach is a very important theorem in real analysis (and virtually everybody getting an advanced degree in mathematics has to make peace with it at some point or other).