3 ms·
Taken from here: http://translate.google.ru/translate?js=y&prev=_t&hl=ru&ie=UTF-8&layout=1&eotf=1&u=http%3A%2F%2Fhabrahabr.ru%2Fblogs%2Fnigma%2F78531%2F&sl=ru&
by alexk 17y ago
Taken from here:
http://translate.google.ru/translate?js=y&prev=_t&hl=ru&ie=UTF-8&layout=1&eotf=1&u=http%3A%2F%2Fhabrahabr.ru%2Fblogs%2Fnigma%2F78531%2F&sl=ru&tl=en http://translate.google.ru/translate?js=y&prev=_t&hl...
- jpwagner 17y agogood old Tungsten-Alpha :)
- qeorge 17y agoThis cracked me up, so I decided to see if the domain was available. Not only is it registered, the guy is trying to sell it: http://tungstenalpha.com http://tungstenalpha.com
- zitterbewegung 17y agoSo what is the actual solution?
- teilo 17y agoThe solution is exactly what Alpha computed: x > 1. The numerator factors to (x-1.5)(x-1), which reduces the problem to x - 1 > 0. Solve for x: x > 1. Not sure what the fuss here is.
- RiderOfGiraffes 17y agoThe fuss is that the inequality is undefined when x=1.5 and so there is no definition of what happens in that case. Performing "obvious" completions of incompletely defined functions is assuming you know the context, and sometimes you don't. This is a sloppy engineering solution that is not guaranteed to be right in all contexts. Without the context, the correct solution is to say that the question is ill-defined/undefined when x=1.5, and all values of x such that x>1 and x.ne.1.0 satisfy the given inequality.
- alexk 17y agoSorry for the bias, the original page is in Russian, so I've used Google translate link to make it possible for others to read.