Hacker News Reader

Comment by zimablue | Hacker News Reader

This is hyperbole, I've worked on dev side in finance a while now and bloomberg data is often unreliable, everywhere I've worked people have built layers on top of it, checking vs other sources and fat fingers. Also I'm actually currently working for some oil guys and they use bloomberg for some things but use specialist providers for oil macro, shipping and refinery events. I'd say their main advantages are familiarity and some information oligopolies

View on HN

I would argue they have established a sort of universal minimum threshold of information within the financial industry, across a ton of financial markets. This provides a certain amount of certainty/consistency within markets(really any sort of pseudo monopoly in a knowledge space does). Particularly in the news space, it sort of lets people assess to what degree they might have a competitive advantage by the degree to which their data is more accurate/earlier than that on the terminal/data side.

It seems they must be getting the data through agreement with the companies they're collecting the data on. I wonder where the line is drawn between what they're doing and commercialised insider trading.

You're kidding right? It is more than 99.99% correct guaranteed by an SLA. You can have errors as with any system but they're very open about it and correct it when its wrong & inform you, which is extremely rare (seen it only 1ce myself)

> This is hyperbole

What I love about HN. I read one thing that sounds great and authoritative in a subject that I know nothing about. Just to have someone else rain on the parade! Which point of view is correct? I don't know (so I upvoted both..)

> sounds great and authoritative

"sounds" being the operative word. Lots of ... lets say 'articulate' people on HN, in general. Current thread excepted.

Hacker News is the ultimate waste of time. Correction: Hacker News _comments are the ultimate waste of time.

You're correct data can sometimes be wrong, but this is mainly in the basic datasets provided, and still exceptionally rare. I once highlight a figure was wrong to their helpdesk, in an obscure money supply aggregate, bt public data that I happened to be pretty intimate with, and their helpdesk were all over a solution in seconds.

For added-cost datasets, they're mainly a reseller for a 3rd party data provider and SLAs need to be checked to see what you're getting.

For checking via other sources, this is a minimum in finance. Reconciliation / validation needs to be done at every step something new is introduced, and just for the heck of it usually anyway.

Reply on news.ycombinator.com