Freigeben über


What kind of data person are you?

andrewI have been working with data in one form or another since 1991, but I wouldn’t describe myself as a DBA, data scientist or a developer so what am I, and am I alone?

I spent ten of those years doing business intelligence and most of that was working with databases, designing systems data marts, and analytical models, as well as getting into search and unstructured data.  What these had in common was my desire to provide the business I was in with a better view on the decisions they were making.  However what I was creating in each case was a space in which business users could explore that data to come up with their own answers rather than me doing that as the questions changed quickly and needed a deeper insight into the business than I had

More recently I have been getting back into working with data again and what’s interesting about that is that there seem to be a lot of other data orientated people like me who are exploring the many new data tools that exist both on the desktop and more interestingly in the cloud.  Some have deeper technical knowledge while others are really close to the business.  In all cases they are curious but cautious in that they will try test and evaluate rather than ignoring a new opportunity or taking someone else’s word for it.

I would classify us (or should I say cluster us) as data scientists however that term seems to be reserved for those who have a deeper mathematical / statistical background so what should collective term should I use to describe us?  In the team I work in we call ourselves data developers and I kind of like this as the word development implies that we do all do some sort of coding (such as R, Python, SQL, MDX and F#) and it also means we are using some sort of work bench like the SQL Server data tools in Visual Studio.  It also suggests that we are developing the data into something in the process, for example in Machine Learning that something would be a prediction. in BI it might be an analytical model and in real time analysis that might be a trigger to fire an event when a condition is met.

There are other groups around this field as well. I have already mentioned the data scientist and it might be that their role might be to oversee the models we build and sign them off for production much as an actuary would do in an insurance company.  Other data related roles are the visualisation experts who like user experience designers on an application team put the gloss and usability on data to ensure that its true meaning is self evident.  Then there are curators/librarian types who understand how disparate data sets fit together.  Perhaps one are that will emerge that might get you thinking is an ethicist - as the laws on data privacy are far behind what is possible these days so the question is not “is this illegal?” but “is it ethical?”. Just because someone didn’t uncheck a box on an online form, and just because someone can now mash that preference data with their social media data exhaust and a  bunch of commercially bought 3rd party data that enables a data scientist to infer some really intimate details about someone’s life, doesn’t mean we should.

Anyway back to data developers.  I think there’s a lot of us out there and that we have had some sort of background in business intelligence, data management etc. and while we may have done some maths in the past it’s not the day job.   My conversations at recent events seem to support this and so our plan is to try and help you transition you into this new world with in person and online content which the MSDN UK team have curated onto a dedicated data developers portal

What we need from you is feedback on what we are doing to ensure you are getting the content and events you need so it’s just like any  other science project.