Over on the "developer side" of the Internet, I regularly encounter articles of the type "Great Programmers <do blank>", where the blanks range from "use source control", to "write reusable/debuggable/self-documenting code", to "work on open source projects in their spare time".
There are several potential problems with such articles. At their worst, they appear arrogant, a self-proclamation of greatness on behalf of the writer ("I am great, and I do this…"). Occasionally, they read like a strange form of blackmail ("If you don't learn this you are a bad programmer"). Perhaps most problematically, such statements are usually an over-simplification. Not all great programmers try to write reusable code. Not all great programmers always use source control (Linus Torvalds wrote Linux without source control, apparently, though maybe he's an exceptional case). In general, these blogs do little to clarify what skills and training are essential to make a good developer.
By contrast, "Great DBAs <do blank>" articles are largely absent. Why is this? Perhaps it is, at least in part, because Brad McGehee and Louis Davidson have cleared the air with their book, How to be an Exceptional DBA, and What Counts for a DBA blog series, respectively. Each spells out the skills and qualities that make a good DBA and it is difficult to argue with their detailed prescription. What I like abou tboth is they give a clear message that anyone can become a good DBA, armed with a good ability to absorb knowledge, sound judgment, tact, and the ability to assess critically a given situation and make sensible decisions on how to act. Of course, some of these skills come only with experience, but they do not require a special talent, or an extra lobe to the brain, just application.
Brad, of course, is a classic Production DBA rather than a database developer or BI Specialist. He spoke for the traditional DBA role. By contrast, we're still straining to find the consensus understanding of what makes a good data scientist. Shouldn't DBAs, as a profession, take an active role in defining the optimum skills and habits of these emerging specialties before the HR people, journalists and recruiters do it for them? Maybe it will just mean we're afflicted with endless "Great BI specialists | database developers | data scientists <do blank>" blogs!
Cheers,
Tony.
