How can particles travel in a straight line? A particle can be set off in a certain direction by giving them momentum.  Momentum is a vector, so the particle heads off in a specific direction.  But the wave function of the particle allows it to obtain other momentum values, which would steer the particle on a different path. How then can we "shoot" electrons and other particles in straight lines? How can they maintain their momentum in the face of quantum uncertainty?
 A: The uncertainty principle's restriction on the minimum spreads in position and momentum is really small. An electron can be confined to a region in both position and momentum space that's extremely small compared to anything human-sized, but still have more than enough spread to obey the uncertainty principle.
To give you an idea of how small this is, $\hbar=1.05457173 × 10^{-34} J\cdot s$, so an electron with a standard deviation in it's position of 10 micrometers ($10^{-5}m)$ has a minimum uncertainty in it's velocity of about $6 m/s$. Electrons in any sort of beam are usually travelling at some appreciable fraction of the speed of light ($3 \times 10^{8} m/s)$, so this uncertainty is tiny. 
A: Dan's answer is quite nice but I think the answer to this really depends on the interpretation one chooses to use.
The straightforward answer I think is that you probably are confusing collimation with the creation of electrons. 
In classical mechanics one can independently specify the $\vec{x}$ and $\vec{p}$, very much independently for the reason that Dan describes in his answer. To illustrate this suppose you created an $e^-$ with thermal ionization then it's momentum and the position will have distribution which will be given by the wavefunction of the electron. These thermal electrons are then collimated into a beam of electrons which you can say have reasonably sharp peak in k-space. Although it seems that one can make beam as thin as possible and violating transverse uncertainty principle, but after a while when the opening size becomes of the order of $\lambda$ of electron the beam start diffracting, hence saving the uncertainty principle from violating. 
But you might ask what about the longitudinal condition? Well that one actually is already saved as the position of particle a priori is unknown.
I think you should think about a better and more clearer alleged absurdity claimed by quantum theory (BTW this is not exactly absurd since this has been verified by experiments e.g. Davisson and Germer). It is now pretty much a handy rule not to think of the particles traveling in the straight line. Then how would you explain the tracks seen in the cloud chamber which seemingly violates the uncertainty principle. This problem seems to go away when one realise that it is ensamble to which this uncertainty principle apply not to one single measurement. 

PS: I think the last couple of lines might interpreted differently in many worlds and consistent histories.
For professionals and experts, I am just a student yet so do correct me if you find any part wrong.
A: The uncertainties on $x$ an $p$ do not imply any randomness of the trajectory until they are measured. If you don't measure anything, the particle wave function is completely determined and goes in a straight line to infinity, with a given probability distribution of $x$ and $p$. When you measure something, then the wave function is modified and may not go in the same average direction.
