The Bloomberg terminal is awesome! By now its been iterated to a point where it's like the Microsoft Office of Finance. It does everything and there is no way for one person to understand everything it can do.
It also has the nice lock in feature of the 20% of features that each person needs is different. This is probably the biggest reason why no one is yet encroaching on their territory.
it you trade debt, it has your yield curves
If you trade currencies it has your currency rate matrix.
Back office, you've got PORT and MARS for portfolio and risk management.
Developer, you've got the Bloomberg API that allows you to pull real time quotes for any exchange in the world.
And if you are stuck they do something that seems inconceivable to startups today, they have actual people you can talk to that help you figure out how to solve your problem!
One other often overlooked benefit of the terminal is that Bloomberg unlocked it via Bloomberg Anywhere. This means you can install the Bloomberg terminal software on any machine you want and then login to use it so that you aren't locked to a particular machine unlike the old Bloomberg Terminals.
The downside of this is that you can only be logged in at one place at a time, but pretty much everyone uses this setup nowadays.
And there is IB the Bloomberg instant messenger, just by typing in a name you can talk to anyone else with a terminal, which is pretty much anyone you'd want to in finance.
IB is also a sneaky way for recruiters to find out who they should be contacting:)
The "help" button is very cool. Back when I worked at Bloomberg, I knew a few engineers who actually ended up on a customer support call because a customer found a bug. That's pretty cool; very few large companies will let people chat directly with engineers for bug diagnosis.
One of the most interesting things about the Bloomberg terminal that never seems to get mentioned in all the coverage and commenting is that its UI is optimized for users who actually take the time to go through training.
In particular, everything is designed so that you never have to touch the mouse. If there are 20 choices onscreen, they'll all be numbered, and you type the number of the choice you want and press enter. This allows for incredibly fast interaction, but with a bit of a learning curve. And customers take the time to learn, because their peers and their boss emphasize how valuable that investment of time will be.
If the terminal were being introduced today, without that existing body of zealots to preach the value of getting over the initial learning curve, there's no way Bloomnberg could get away with it. The sales team would visit a potential customer, the customer's eyes would boggle at the UI, and Sales would come back to the office demanding a redesign that looks like Microsoft Office and is mouse-centric.
> If the terminal were being introduced today, without that existing body of zealots to preach the value of getting over the initial learning curve,...
Honest question: doesn't it also have a lot to do with the technical sophistication of the typical user, not just lock-in?
Well, the Terminal's UI when introduced in the early 1980s was cutting edge, but that meant text mode and keyboards - no GUIs or mice back then.
If being launched today, just like back then, it'd be designed to provide both as shallow as possible a learning curve for new users and power and speed for experts. In the 1980s that meant full-screen UI with numeric menus, Go codes, and custom hardware; today, that'd mean a discoverable point-and-click self-teaching GUI for the newbies, backed up by keyboard shortcuts, scripting, and a command bar for power users.
Looking at a video on YouTube of it, it seems it actually has a very modern search-based UI where you type something and it autocompletes/searches commands, and things are clickable by mouse too.
So for instance you type "Google" and it shows a bunch of Google-related stock ticker, you click one, then type "earnings" and it shows you some tools that allow you to learn about earnings, graph, project them, etc.
It doesn't look like you can do much better UI-wise even for beginners.
It seems substantially better than Google/Yahoo Finance though, so there could be a market for building a (ad-supported?) website that offers the most common tools with the same quality of the Bloomberg terminal.
One can always do better :) But yes, "autocomplete" (in Terminal parlance) has made discoverability much better. You can just start typing what you want and chances are you'll find what you're looking for right away. You can even ask it questions and it will answer the direct question for you if it can. (e.g. "What were Google's earnings?" takes you to the exact answer)
> its UI is optimized for users who actually take the time to go through training.
I read another take on Bloomberg terminal's dominance.
A (potential) competitor might look at the Bloomberg screens and think they could recreate that more cheaply with Angular, React.js, or some fancy GPU accelerated DirectX. Maybe they'll have interesting ideas like beaming price charts to Google Glass, Apple Watch, etc.
However, that's just looking at the surface level.
The thing that's very hard to duplicate is the relationships that Bloomberg has built up over the years with all the data providers to get you ultra low latency data feeds. That's hundreds of golf games and 3 martini lunches the competitor would have to duplicate to establish the relationships Bloomberg has. The programming of the hardware and GUI would be trivial compared to the wheeling & dealing that has to be done.
There are some of us HNers that develop the Terminal software/infrastructure (or used to)... so if there's ever anything you were dying to ask, AMAA. :)
Yes, we run one of the largest (if not largest) private financial network in the world. Something like 180+ countries, close to 1,000 core routers, etc. with proper DR (dual-homing, separate fiber, risers). It's not like we run a NOC in each customer's office. It is all standard stuff -- you concentrate in big cities and run lines to individual customers. Lines are exactly the same type of lines you'd provision for Internet access (anything from a T1 to OC-3 and beyond, metro ethernet, etc.)
You can certainly connect to us over the public Internet as well -- but you have to factor in down-time, latency, and other factors. There are big Internet outages that have happened in past years (e.g. anchors cutting cables) where we kept our customers online while the Internet was down in the entire country.
The API requires a subscription. The API SDK was open-source licensed so that either open-source or proprietary applications could include it in their distribution so that they can consume Bloomberg data "out of the box".
An example is OpenGamma -- http://www.opengamma.com/ -- it is an open-source app that ships with the BLPAPI and if you run it on the same PC as your Bloomberg Terminal install, it will just work and pull data from the Terminal API end-point.
The incentive is for apps to support Terminals / Bloomberg data natively from the get-go and it makes it easier for these startups/smaller players/open-source apps to have a chance of selling into firms without lengthy custom development efforts to adapt other data sources.
It's not really as simple as that. Just because you have data doesn't mean you have (sometimes dizzyingly complex) analytic tools to make sense of the data. An API and Excel plugin exist for people who want to do things themselves, but the Terminal app is definitely not secondary for the vast majority. You're paying for service, which includes everything as well as continued enhancement, new features and data sources, and 24/7 technical and analytic support.
FiveThirtyEight's "What's the Point" podcast recently had an episode mostly devoted to the Bloomberg Terminal. I found it fascinating. I would have never known it had its own sections/apps/services for travel booking, classified ads, chat, etc. It would be so fun to poke around at one in real life.
As someone who recently joined the firm to develop software for the terminal I find it heartwarming that it is so well regarded in the HN community.
I did struggle a lot with the UI initially though I hear that most of users are hardcore about keeping it as it is.
Outside of Wall Street, or rather non-U.S. financial industries, do they also use Bloomberg Terminals for their markets or do they use something home grown or nothing at all?
What's the value advantage that a network like this gives to the U.S. vs. other places?
27 comments
[ 3.1 ms ] story [ 79.3 ms ] threadhttps://news.ycombinator.com/item?id=10341334
It also has the nice lock in feature of the 20% of features that each person needs is different. This is probably the biggest reason why no one is yet encroaching on their territory.
it you trade debt, it has your yield curves
If you trade currencies it has your currency rate matrix.
Back office, you've got PORT and MARS for portfolio and risk management.
Developer, you've got the Bloomberg API that allows you to pull real time quotes for any exchange in the world.
And if you are stuck they do something that seems inconceivable to startups today, they have actual people you can talk to that help you figure out how to solve your problem!
One other often overlooked benefit of the terminal is that Bloomberg unlocked it via Bloomberg Anywhere. This means you can install the Bloomberg terminal software on any machine you want and then login to use it so that you aren't locked to a particular machine unlike the old Bloomberg Terminals.
The downside of this is that you can only be logged in at one place at a time, but pretty much everyone uses this setup nowadays.
And there is IB the Bloomberg instant messenger, just by typing in a name you can talk to anyone else with a terminal, which is pretty much anyone you'd want to in finance.
IB is also a sneaky way for recruiters to find out who they should be contacting:)
In particular, everything is designed so that you never have to touch the mouse. If there are 20 choices onscreen, they'll all be numbered, and you type the number of the choice you want and press enter. This allows for incredibly fast interaction, but with a bit of a learning curve. And customers take the time to learn, because their peers and their boss emphasize how valuable that investment of time will be.
If the terminal were being introduced today, without that existing body of zealots to preach the value of getting over the initial learning curve, there's no way Bloomnberg could get away with it. The sales team would visit a potential customer, the customer's eyes would boggle at the UI, and Sales would come back to the office demanding a redesign that looks like Microsoft Office and is mouse-centric.
Honest question: doesn't it also have a lot to do with the technical sophistication of the typical user, not just lock-in?
If being launched today, just like back then, it'd be designed to provide both as shallow as possible a learning curve for new users and power and speed for experts. In the 1980s that meant full-screen UI with numeric menus, Go codes, and custom hardware; today, that'd mean a discoverable point-and-click self-teaching GUI for the newbies, backed up by keyboard shortcuts, scripting, and a command bar for power users.
Much like Excel.
So for instance you type "Google" and it shows a bunch of Google-related stock ticker, you click one, then type "earnings" and it shows you some tools that allow you to learn about earnings, graph, project them, etc.
It doesn't look like you can do much better UI-wise even for beginners.
It seems substantially better than Google/Yahoo Finance though, so there could be a market for building a (ad-supported?) website that offers the most common tools with the same quality of the Bloomberg terminal.
I read another take on Bloomberg terminal's dominance.
A (potential) competitor might look at the Bloomberg screens and think they could recreate that more cheaply with Angular, React.js, or some fancy GPU accelerated DirectX. Maybe they'll have interesting ideas like beaming price charts to Google Glass, Apple Watch, etc.
However, that's just looking at the surface level.
The thing that's very hard to duplicate is the relationships that Bloomberg has built up over the years with all the data providers to get you ultra low latency data feeds. That's hundreds of golf games and 3 martini lunches the competitor would have to duplicate to establish the relationships Bloomberg has. The programming of the hardware and GUI would be trivial compared to the wheeling & dealing that has to be done.
You can certainly connect to us over the public Internet as well -- but you have to factor in down-time, latency, and other factors. There are big Internet outages that have happened in past years (e.g. anchors cutting cables) where we kept our customers online while the Internet was down in the entire country.
Could you set up your own stock dashboard or application with just that API, or do you need to subscribe to get database access as well?
An example is OpenGamma -- http://www.opengamma.com/ -- it is an open-source app that ships with the BLPAPI and if you run it on the same PC as your Bloomberg Terminal install, it will just work and pull data from the Terminal API end-point.
The incentive is for apps to support Terminals / Bloomberg data natively from the get-go and it makes it easier for these startups/smaller players/open-source apps to have a chance of selling into firms without lengthy custom development efforts to adapt other data sources.
So I imagine Bloomberg must negotiate for access to this data, and that's what I'm really paying for? The actual terminal application is secondary.
http://fivethirtyeight.com/datalab/podcast-wall-streets-just... - Bloomberg Terminal discussion starts around 5:30.
What's the value advantage that a network like this gives to the U.S. vs. other places?
edit: Also very important is native speaker tech/analytic support in all major customer languages.