Winning the battle and losing the war

In 1916 the Battle of Jutland/Skagerrackschlacht took place in the North Sea. Afterwards, both the countries involved – the United Kingdom and Germany – claimed a victory.

In fact, in terms of damaged ships, ships sunk or sailors killed, Germany had an advantage of roughly 2 to 1. On the statistics, it was a clear victory for them. Germany had tougher ships, better ammunition, better rangefinders, and better officers.

But it was not good enough. The object of the battle was to allow the German High Sea Fleet to take command of the North Sea. After the battle, however, it was back in port and unable to leave. The Royal Navy outnumbered the German Navy so greatly that it still controlled the North Sea – and in fact it could probably have lost another Jutland and still retained control.

This emphasises the importance of understanding individual battles in context. For the games theorist, it shows the importance of modelling enough of the overall game to understand the implication of individual results.

Finally

For some time I’ve been thinking of writing about an interest of mine – the computer modelling of simple game engines for well known games.  I’ve decided to do it in blog format rather than try to write an unpublishable book. Nobody may read your blog either, but it doesn’t have quite the same commitment to proofreading, editing and getting involved with publishers or publishing tools. So yes, I’m lazy.

I’ve spent too many years of my life writing Java, C and SQL, so I’ve completely abandoned that approach. I’m retired, I can use what I like. So what gets put up here is going to start off with  Python. It’s a popular language in education, and it has only one major fault; it exists in 2 rather different flavours (2.7 and 3.3 at the time of writing), and because it is an interpreted language you can’t just do a Java and tell a compiler to compile your Java 7 source code to run on a Java 6 runtime. This means that some potentially useful frameworks are only available in 2.7, but if you are a language purist like me, you want to use 3.3 because 2.7 has some features that suggest that the original designer had no idea how big this would all get. Let’s see how that pans out.