How do I know what the field names are in .osm files.

If you want to see what the most commonly used keys are worldwide, have a look at Taginfo. You can use standard GNU/Unix utilities to look at what's in the ".osm" file, if it's too big for a text editor. If you're on Windows and don't have things like "grep" available, have a look at Unxutils, which is a native build of most of the common GNU utilities for Windows, or Cygwin, which is a slightly different approach to solving the same problem.

For example, looking at a small .osm file that I happen to have lying around:

c:\Temp\osm\new>grep 'k="highway"' local_101017_01.osm | wc
    3182     9546   109056

I can see that there are 3182 "highway" items (roads, paths, bus stops etc.)

To see what sort of highway items there are I can do:

c:\Temp\osm\new>grep 'k="highway"' local_101017_01.osm | \utils\sort -u | more

  tag k="highway" v="bridleway"
  tag k="highway" v="bus_stop"
  tag k="highway" v="cycleway"
  tag k="highway" v="footway"
  tag k="highway" v="ford"

(rest of output truncated and chevrons removed)

Have a look at the similar program osmfilter.

It has features to produce a listing for all used keys in an osm file, see Osmfilter#Getting_Tag_Statistics