Bloomberg - The future's bright, the future's orange
I remember my first few days at Bloomberg where, like many others, life included sending and receiving many emails. Muscle memory told me to click on Outlook and away we go. Although we had Outlook in those days (and bloomberg.com email addresses), it was severely frowned upon. In fact, so much so, that even though I sent colleagues email communication, there was nothing coming back. My managers (you know who you are Brad/Tricia) constantly reminded me, "just say no to Outlook". I finally got it, and I made that transition to the Terminal, to MSG<GO> and, committed to the portal. Inevitably, Outlook was surreptitiously removed from the wider landscape.
The #Bloomberg #Terminal.. thousands of functions, across all asset classes, more technical analysis than you know what to do with, a vast wealth of static reference data, artificial intelligence built in, messaging, sentiment analysis on news. It doesn’t get much better .. or does it? As I mentioned in my last article, Bloomberg supplemented the Terminal with a point of integration, a means of programmatically getting at the power of the Terminal. Bloomberg created what we call the Desktop API (DAPI), or more formally, the Bloomberg API (BLP API). Today, in the spirit of openness, it's on Github.
Back in 2012, I remember accompanying Shawn Edwards (Bloomberg's CTO) and announcing to the industry that the API, was "without cost or restriction". In hindsight, we probably; peaked too early, the market wasn't ready, the industry hadn't really embraced "open".
As a first foray into this new API, Blomberg then created the first extension to the Terminal, the Bloomberg Office Excel Addin available at all good Bloomberg stores and websites.
Bloomberg introduced the Excel add-in, and opened the door to a family of “B” functions, including =BDP(), =BDH(), =BDS() and many more. Some users today refer to this functionality as the “Excel API”, but strictly speaking, it’s an Excel Addin leveraging the BLP API. It's also an example of eating your own dog food, whereby Bloomberg built a product incorporating another of its own products. For the geeky souls out there, the Bloomberg Excel addin is built on the Excel RealTime Data (RTD) framework which, whilst not strictly realtime (it’s based on polling) serves an important function in getting ticking data into an Excel sheet.
This is an important leap forwards, because now you can arrange launchpad screens with your holdings, news, analytics, etc, but also see real-time information ticking in a spreadsheet, combining the information to help make an informed choice about your portfolio. Every time the price ticks on the Terminal, the sheet updates, all the values ripple through the cells, and voila, result time.
It’s probably worth reminding everyone though about Authentication and Authorisation. Authentication is about whether you as an individual are allowed to access the Bloomberg Terminal. This is done through a Bloomberg login using credentials. You’ll have a @bloomberg.net email, you’ll have a password, but you’ll also have that other piece of technology, the B-UNIT, introduced in 2005. The biometrics part of your credentials. It’s back to the.. something you know (your password) and the something you don’t know (the B-UNIT code generation). The B-UNIT only generates the code when you’ve successfully swiped your fingerprint across the sensor.
More recently, Bloomberg introduced both a Google and Apple app for your phone that performs the equivalent secured biometric function.
Additionally, you may be fortunate enough to have a Bloomberg keyboard with a built-in fingerprint sensor (introduced in 2004). Once you’re authenticated, it’s time to determine what capabilities you have.. your authorisation. This can extend from what Bloomberg commands you have at your disposal to what exchanges, brokers, real-time vs delayed, level 1 vs level 2, etc.
But, back to the spreadsheet. We’re authenticated, we’re authorised.. and away we go. At some point, depending on how complex your needs get (and we’ve seen entire risk systems built in Excel!), Excel is probably going to start creaking at the seams. So, where do you go from here? You need to consider writing an application, and by writing an application that means developing some code that you can run on the same Windows device as the Bloomberg Terminal. It’s worth remembering that the fee paid for the Bloomberg Terminal and the fees for all the exchanges, OTC sources and brokers grants you “for display purposes only" benefits. Ultimately, everything you do with the data is to indirectly assist with your cognitive workflow. Yes, it could provide a “signal” suggesting you trade out of your portfolio, but ultimately it needs to be your decision. The point is, don’t get carried away with your creative streak on writing that next application, or at least, wait a couple of LinkedIn articles.
The most obvious choice for programming, especially if you’re a quant or front office is to use Python. Python provides a notebook approach to writing code, with something called Jupyter. It’s a dynamically typed language so great for prototyping and quickly developing functionality. It's not a good match for performance just in case you're contemplating your next low-latency sub-second millisecond algo, but then, neither is the Terminal. For those wanting a more strictly typed language coupled with a severe rights of passage, you could choose C++, C# (.NET), Java and plenty more. For now, because of the tight coupling in terms of Authentication and Authorisation, and to a lesser/greater extent.. licencing, you’ll be developing on Windows, because that’s where the Terminal runs. Bloomberg offers a web page with all the libraries you’ll ever need to develop an application, and if you’re fortunate to have a Bloomberg Terminal look at WAPI<GO> also.
In summary, we started with the Bloomberg Terminal, then we’ve extended off Terminal to using the Excel Addin, and, the possibility of writing an application to provide even more functionality and help support the day job. But consider the what-next on the following;
You hit the Terminal limits
Your application provides so much value that the rest of your colleagues should be running it
Your application needs to run 24/7 (avoiding the logged out at 6pm syndrome)
In the next article I'll talk about a way forwards that still leverages your Bloomberg Terminal capability but steps it up a notch to address the above three challenges.
If you can’t wait, or need some help discussing application architecture, or how to get the most out of your Bloomberg ecosystem, don’t hesitate to get in touch with us here at Sherpa Consultancy LTD or reach out directly to me.
