Tier 1 Sell-Side Language War: Slang vs Python vs Scala
Although a few months old, Paul’s article in eFinancialCareers is still relevant: If you’re a technologist with Python or Scala skills, consider yourself in demand. Essentially the article highlights the “new” languages that are being used by a few of the leading banks. A 50,000ft view of the ripples in the .NET/Java space from Scala/Python can be gleaned by trawling the usual suspect recruitment web sites. David Pollak offers some finance space views here.
The Scala/Python up-tick at banks has mean that proprietary languages within certain banks have possible finally begun to be irradiated – although it will inevitably take some time to completely removal (if at all possible). However, one banks that doesn’t appear to be heading down this road is Goldman, who as far as I’m aware continues with Slang – maybe forever :)
What does this mean as a net out effect within finance? Essentially looking at the handful of top banks as per Paul’s article. At least three large banks will essentially have a similar “conceptual” platform for their pricing and risk management system. Does this mean that the other banks will now need to follow the “three”? Have the “three” an unhealthy advantage after a considerable amount of investment? Only time will tell, but its clear Goldman’s was perceived to have an advantage by two of its competitors, who have now subsequently done down a similar road.
In the Python financial world I suspect a lot more developers are now aware of the Global Interpreter Lock (GIL), especially given the calculations and threading nature of front-office solutions. Reading on GIL is available here, here, and here as a starting point. Merbist offers a view from RubyConf 2011 on the problem which is mirrored in the Ruby wold.
Final word: As we have seen with the changing Rich Internet Application (RIA) world, nothing stays constant in technology. The language wars will continue.