I've useed LOTS of LM78xx & LM79xx parts. But, any "3-terminal" linear regulator is just as easy.
I've never built a switching regulator. A switching regulator requires more components (including an inductor) and it's more complicated to design. But where I work, most of our boards have switching regulators. It's not a big deal to add a component or two to a board, and you only have to go through the design process once. Some of our boards have a ~15V to 5V switching regulator, plus a 3.3V linear regulator.
Regulator chips can be fixed or ajdustable/variable. An adjustable regulator takes a couple of resistors to set the output voltage. Of course you can add a pot (and maybe some other circuitry) so that the user can adjust it.
As far as specs, of course the 1st thing to consider is the output voltage and output (load) current.
Then, you can check the input voltage range. It needs to handle whatever voltage you are putting-in. For example, depending on the exact part number you are using, the LM78xx chips can take up to 25 or ~40V. If you go over the rated input voltage, the part might die.
You also need to consider the minumim input voltage. The LM78xx series has a dropout voltage of 2V. That means if you are using a 7805 (5V output), you need to feed-in at least 7V. If you go below 7V, the device will "drop out" of regulation and you'll get less than 5V out. If you are going from 5.V to 3.3V, you'll need a "low dropout" regulator, since you are only going to "drop" 1.7V across the regulator.
If you are using a linear regulator, you also need to calculate power dissipation. For example, a 7805 rated for 1.5 Amps, will probably burn-up if you run it at the full 1.5 Amps with 30V input voltage.
As you may know power (Watts) is calculated as Voltage x Current. In the above case, I've got 25V dropped across the regulator with the 1.5A current flowing through the regulator and the load. There is 37.5W (1.5 x 25) dissipated by the regulator and 7.5W (1.5 x 5) dissipated by the load. You can see this is very inefficient, with more power (and heat) "wasted" by the regulator than is being used by the load. And, the full 45W has to be supplied by the 30V supply. 5V linear supply running off 10V is (about) 50% efficient. (I say "about" because there is also a tiny amount of power required just to "run" the regulator.)
Usually, none of this is a problem if the current draw is low. Unless you are running from a battery and you need the battery to last as long as possible. If heat, power dissipation, & efficiency are important, you can use a lower input voltage, or go with a switching design.
Switching designs are nearly 100% efficient. It's usually not necessary to calculate power dissipation (and it might not be possible to calculate accurately) as long as you stay within the voltage & current ratings.
Since switching regulators are so efficient, nearly ALL of the power goes to the load. This means that the regulator reduces the voltage, but you generally get more current out than you put in.
The exception is at low currents. As an extreme example, with no load connected there is power consumed by the regulator and you have zero efficiency! In very-low current situations, a linear regulator might be more efficient than a switching design (depending on the design details, etc.) because the linear design might require less idol current, or less power to "run" the regulator.
There are variations of switching regulators, such as DC-to-DC converters. I bought (didn't build) a DC-DC convert to convert +12V to -12V for an automobile project that need both positive & negative supplies. Another DC-DC converter application is a "step-up" regulator. For example, a circuit that converts 5V to +12V, or -12V.