What makes finding astronomical distances so hard is that each distance indicator is usable only over a very limited range of distances. For example parallax is the most direct way to measure distance, but it only works out to about 50 parsecs or so. On the other hand, Cepheid variable stars are a good standard candle, and they're bright enough to see them out to large distances, but none of them are close enough to measure distances using parallax. So, we have to find a distance indicator whose range of usefulness extends out to the nearest Cepheids. We use parallax to calibrate the intermediate indicator; we use the intermediate indicator to calibrate the Cepheids, and we use the Cepheids to calibrate still more distant indicators. This structure is called the Distance Ladder, and it is fairly rickety because an error in any "rung" can be multiplied as it is carried up the ladder, resulting in a totally bogus value of H_{0}. In fact, it was exactly this sort of error that caused Edwin Hubble to get the ridiculously high value of 500 km/s/Mpc for his first measurement of H_{0}. Since the age of the universe is proportional to 1/H_{0}, Hubble's first measurement implied that the universe was only about 2 billion years old, which is younger than the earth, let alone the ancient globular clusters.