Simplicio: Early beer was more commonly of an "ale" type, fermented in open (or only covered) vats, with the produced carbon dioxide mostly diffusing out through the liquid out into the brewery. So early beer was (mostly) flat. Lager-style beer is brewed in a pressurized environment, with the pressure being provided by the carbon dioxide given off by the yeast. Lager-style beers are pretty modern, as these things go. Also, do not (necessarily) look at what comes out of the taps at the pub, most pubs will be using either nitrogen or carbon dioxide to provide pressure in the kegs. This then squirts the beer to the tap and into a waiting glass.

Joseph Connolly wrote: At any rate, I was asked in a phone interview the difference between a linked list and an array. The company does not do low level programming or code for mobile devices which would make the performance between the objects perceptible. There's no need to do "Low-level programming" or "code for mobile devices", it's just needed that you have a lot of data to process. Use the obvious, wrong, strategy to extend your arrays as you accumulate more data and you'll end up with O(N^2) (n-squared) time complexity, where you'd same would be O(N) time complexity using linked lists. On the other hand, you'd then have trade-offs further down the processing chain, so it MAY be better to use one representation in one stage, then convert to another representation down the line. As for "how many" the lots would be, I can't give you a hard answer, but if we're looking at 1500-2000 elements and the O(N) version takes 2 seconds, the O(N^2) version would be over an hour and that may be significant (it mnight also not matter, it depends on how frequently and under what conditions the code is run; if it's part of the annual balancing of the books, it might not matter, if it's code that runs every 3 hours, it probably matters a lot).