Rather less common is the completely freelance railway that follows railway practice. Light railways, or 'Colonel Stephens' lines are another variation. The interesting point here is that some recognised railway practice is always observed, despite the fact that it would be possible to start completely from scratch and devise your own standards.
As a fan of both railways and science fiction, this has always seemed quite a normal association but some railway fans would be quite upset if you pointed out the obvious connection.
What has always puzzled me is the rather limited way in which this association is carried out. Why limit yourself? You don't actually have to invent anything, just imagine that some past event didn't happen. For example, why not model a reality where Nationalisation never occured? Or where the 1922 grouping never occured? I didn't want to throw too much at you at this stage, but what about a broad gauge GWR steam layout set in 2007?
That may seem like a giant leap, but all it requires is that:
a. The GWR didn't drop broad gauge.
b. The grouping never occured.
c. Nationalisation never occured.
d. Diesel/electric never occured.
All negatives. All we've done is assume that something didn't happen. Remove just one of those and the possibilities are endless. Broad gauge GWR overhead electric anyone?
Now, what if IKB designed a revolutionary auto coupler that was fitted to all GWR stock and kept secret from all the other companies. It looked remarkably like a tension lock (Triang perhaps) coupling. He also developed a remarkable automatic train control which allowed his locomotives to run with no obvious crew aboard. This had the useful side effect of making open cabs less impractical.
When people say to me, "I see you model the GWR", my answer is,"No, I don't model the GWR, I model my GWR."