One key advantage that telcos and ISPs who provide DSL (Digital Subscriber Lines) push, other than the higher speeds of course, is that you only need one phone line for both data and voice transmissions. This is fundamental to the way DSL works. I will explain it briefly here.
The basic idea is dividing up the frequency range on the copper wire, and using different frequencies for regular telephone usage and DSL. This works because voice transmissions are already limited to the frequency range of 0-3400 hertz. This was done originally to lower analog interference from phone lines to each other. (Note that fax machines and dialup modems also operate at this range, thus you can receive fax transmissions while online with DSL, and technically, if you wanted to, even use a dialup modem at the same time although you'd have to reset some configurations in your computer usually.)
Phone lines can actually handle much higher frequencies. DSL takes advantage of that, and allocates those higher frequencies for digital data transmissions. The two popular implementations are CAP and DMT. In CAP, the frequencies are statically assigned to each task. Upstream data is carried on the 25-160kilohertz range. Downstream is carried from 240khz up to whatever the line can handle. In DMT, the entire band is divided up into 247 channels of 4kilohertz each. Each of these can be used for downstream and upstream, some even bi-directional. The signals switch dynamically between the channels depending on what the line quality on each channel is like at the time.
Usually customers will need to install low-pass filters to telephony devices other than the DSL modem. What these are, are simple filters that block out any signal over, say, 3400khz or 4000khz or so. Basically it lets vocal transmissions through while keeping DSL signals from causing static interference on the line.