What History taught me - the early years

My first couple of layouts were just toys for me and a hoped-for way to engage with my sons.  They were used only casually - built, operated for a while and then left alone till it was time to move house.  I've tried to pull lessons-learned from each of my layouts so as I move forward, I don't make the same mistakes again. As I read through, afterward though, most of these lessons could have been avoided if I had only read and researched more.  The design and operating mistakes seem very elementary now as I look back.

1. I need to be able to run at least one train continuously.  That first layout in the garage in the 70s was a simple point-to-point because I had no room for anything else.  Homes in that part of England have no basements and since the roof timbers are supporting concrete tiles, the attic wasn't an option. We didn't have a spare bedroom either with two kids in a three bedroom house.  So the garage was it.  I soon learned that I go bored just switching - oh I enjoyed it when things worked but running a train took work and continuous attention.
2. Complicated trackwork should be avoided in N-scale.  I suffered continual derailments with lightweight rolling stock being both pulled and pushed through a succession of switches in the station throat.  While I was trying to base my model on a real prototype, I did no research into specific locations. I could observe locations on the Southern region and figured for a terminus station to work, I had to be able to access either inbound or outbound main line from any platform. Adding another station exit to the passenger car storage area was just too much.  The N-scale switches had to be manipulated by hand and didn't always spring into place - I vowed to not entertain complex trackwork ever again.
3. One single locomotive is a ridiculous proposition for a two-track mainline and passenger facility.  I needed at least two more locos and, with no command control back then, would have had to install block sections which I didn't know anything about.
4. A passenger-only layout lacks interest for me. There's a limit to the number of destinations for a passenger fleet.  With a desire for operation, I needed real switching destinations.
5. N-scale was really too small and fiddly. I dare say that 45 years later, the technology has moved on and couplers may have improved, track switches may work better and trackwork, in general, may be more tolerant of substandard benchwork and baseboard.  
6. Must have switching. The second layout , was a free-standing 7' x 5' HO double-track loop with two sidings.  I tried to combine a point to point with  continuous running.  It was a first experiment with gradients an upper level cement works making a tunnel over the back one-third of the layout and an unfinished lower level destination. It provided limited switching on the upper level with a point-to-point run to the lower level.
   
   
   
7. Steep grades don't work.  This layout was more interesting but the ability to run long trains was limited by the steep grades.  My single four axle hood unit couldn't manage more than a couple of cars up the steep grades. 
   
8. Must have a reverse loop.  I had realized from an article in Model Railroader magazine that I needed special wiring but had no real idea how to wire a layout for a reverse loop.  I really didn't know what a reverse loop was and even if I had, I wouldn't have had room for one.  This second layout had provided reasonably reliable operation with Atlas snap switches and code 100 flextrack an began to pique my interest.  This was a purely freight railroad which  I also enjoyed. 
9. Must be able to isolate a second locomotive. The Atlas switches routed power so I didn't have to worry about wiring for isolated electrical blocks.  I'd somehow managed to wire my DC for a cross-over but I can't remember how. 

With my interest piqued, I went out and bought "Track Planning for Realistic Operation" by John Armstrong, an invaluable resource and also fun to read.  I read this book cover-to-cover several times over the next 10 years during which I never built another layout.  I also began subscribing to Model Railroader magazine, I bought a large pad of grid paper and confined my activities to the armchair.  I moved five times during this period, living in England, Connecticut, New York and Arizona.