LSE dumps Windows? NYSE dumped by SDBK

If Microsoft has lost its Windows/.NET foot hold in the LSE and it was related to issues in old versions of SQL Server/Windows, it may at least say some face by disclosing the issues and validating that they are fixed in news version of its database/OS. Speculation alone could damage Microsoft in the capital markets world.

On the flip side, it looks like today wasn’t a good day for the NYSE. Microsoft doesn’t appear to be blamed for the NYSE issues. NYSE SDBK is a server-based system based on NYSE Arca – the SuperDOT replacement. Anyone know what NYSE SDBK was written in? Java on Linux?

NASDAQ OMX Market Replay Europe goes Adobe AIR

NASDAQ OMX Market Replay Europe provides users with instant access to historical quote and trade data from major European exchanges. Filter by exchange for a more targeted view of the market. Analyze execution quality and monitor MIFID compliance within minutes of trade execution.

Video here, or sign up for a free account and install (AIR and Administrator access required) the RIA – 2nd February 2009 data only available on the free trial.

MSDN Magazine – July 2009 Articles Worth Reading and Stuff

Here’s my pick of the articles to read:

• Securing The .NET Service Bus
• Composite Web Apps With Prism
• Distributed Caching On The Path To Scalability
• Request-Response Testing With F#
• Building Tuple
• Usability Testing

Other postings worth reading in blog land:

• My F# Notes
• LAgent
• MVVM light toolkit for WPF and Silverlight
• Azure worker role development tips
• 20 User Experience Books you should own
• PragPub

CEP within FX

e-Forex has an article on CEP and FX: while trading and trade-related projects like smart order routing (SOR) accounted for 48% of existing use cases, areas like risk management and market data were making “double-digit headway”.

“While the initial usage of CEP technology was in equities trading and algorithms, it has since moved out into other asset classes. The need for CEP – be it in FX or other asset classes – is driven in areas where there is high throughput, low latency and a need for faster decisions. Traction is certainly building in FX, especially compared to recent for tunes of the equities market.”

StreamBase Adds Twitter Algo Trading

A new twist on machine-readable feeds. Has StreamBase found a new opportunity to weld social networks and trading? Something for Microsoft to consider with Orinoco.

Matrix and Bloomberg

Building Low Latency Applications for Financial Markets

White paper available here

SIFMA 2009 – Siverlight and WPF

Marc offers a view of SIFMA 2009 – unfortunately being on the wrong side of the pond doesn’t allow me the luxury of attending. I suspect however that Kaazing have something to shout about given that Kaazing Enterprise Gateway was part of the solution that won the Innovator Award from Windows in Financial Services for Silverlight Integration. Shame Kaazing don’t have the demo running on their web site – missing a trick?

Looks like Lab49 picked up an aware for a risk WPF application as well.

Buying Off The Shelf: The Game

There are many flavours of build/buy in the typical capital markets sell-side trading application. Some banks such as Goldman appear to enjoy building their own database, language and then the trading applications on top of their home grown infrastructure. Other sell-side companies are happy to buy off the shelf infrastructure, building only the trading application itself using standard development languages. In general there appears to be a general trend that US investment banks appear to prefer to build rather than buy, where as the European banks are the reverse – my view .

The banks that prefer to build present a challenge for any product company. Product companies normally charge hefty licence fees per CPU, and expect to make a good margin on all sales. In the case of these “builder” sell-side corporations, the first challenge is getting your product to a Proof Of Concept (POC) stage. Post the POC it’s all really down to cost – especially in the current credit-crunch climate. A product company can have the best “solution” in the world, but if the cost isn’t right you could find the door well and truly closed – possible for a number of years until the next POC appears. Product companies need to take a step back and understand that if you can sell your product to a successful project then this is more than likely the best advertising you’ll get within the bank, which will generate future licensing fees coupled with the likely hood that other projects will leverage the successful architecture and knowledge from the initial project thereby generating new licensing revenue for the product company.

Unfortunately, if the product company doesn’t “get” the above they’ll fail post POC with a degree of negative exposure that will probably be well documented on some wiki within the organisation thereby ensuring that the sell-side doesn’t POC again for 2-3years. You can also guarantee that any future POC for product vNext will have to initially overcome the original purchase blocking issue.

In summary, product companies need to understand the impact of their action.

Book: Applying Domain-Driven Design and Patterns

I haven’t read Eric Evan’s DDD book yet, so I suppose in some way I started back to front in the order in which both these DDD books should have been read. Jimmy’s book was a slightly different read from the books I have read of the last few months – the book kind of reads in a paired programming style.

Fluent interfaces (page 176) is just an example of how Jimmy manages to cover so many important patterns in a 500 page book. Linq offers a good example of a fluent interface.

It’s rare to come across a financial pricing sample in a .NET book, but page 369 delivers thanks to Erik Dornenburg – one of Jimmy’s friends who have contributed to the latter part of the book.