This article [1] is also very interesting.
I agree sticky menus are quicker to navigate but my worry about the research is how often do you need those functions? To run a similar test imagine us running the following experiment.
1) Go to the Washington Monument page
2) Find out how tall the washington monument is.
3) You discover there is an issue with the page contact Wikipedia to let them know.
4) Find out when the Washington Monument was erected
5) You are now at the White House, search for the Wikipedia article
6) Find out how many snipers are typically on the roof of the white house
7) Find other articles near your current location
In this sort of workflow the user needs to access the menu throughout the article, so is going to be biased towards an interface that makes this workflow quicker however my question is how often does this happen? We are telling the user how to behave but my question is does the user behave that way? For instance if the user uses their browser address bar to search for Wikipedia articles then do they need quick access to a sticky navigation bar? In such a test would they be more biased to using Wikipedia's search function as they feel like that is there expected behaviour? Also how commonly are they going to contact Wikipedia? How commonly will they use a nearby function?
To take Facebook mobile website as a concrete example (which has no fixed nav at least on my Android), when viewing my news feed I don't actually want the fixed navigation - consuming my friend's latest activity is the most important thing. In a Wikipedia article I think the same goes - but I could imagine that it might be useful to have quicker access to certain features - potentially search for example.
Can someone send me the research that says that users use Wikipedia's search tool? I really don't believe this to be the case for the average user and would be interested to know what percentage of our total audience does use it and whether we've done research to show they use it more if fixed.
I'd be keen at some point to do more goal driven experiments with fixed positioned navigation on a wider audience to see if this encourages more searching. For example we could imagine enabling a fixed navigation menu on all iPhone 5 users for a day and see if the search traffic goes up.