Big Data Ethics and DBAs
For many, many years we’ve spent time here on SSWUG.org talking about ethics. We even went so far as to create a community-based code of ethics for DBAs. It went repeatedly through many revisions and has "rested" since. (Here’s a link to the finalized look at the statement)
Wired just published an interesting article on what amounts to being the state of the state in very brief form, for big data. Where we’ve come from, where things are, etc. You can read more about that here.
This really got me thinking and wondering if the tasks and responsibilities have changed, or whether it’s really just the environment. In other words, we still collectively are responsible for the "goods" but the structure of our world has changed. Same responsibilities, different building.
In many ways, I think it’s true. We’re responsible for protecting data and for making it available to the right people at the right time. Where this may be emphasized at the moment is the stewardship of that data. We have to be looking forward to the implications of the systems we’re supporting. As mentioned in the Wired article, the NSA revelations (and does anyone really believe we fully understand all that is being/can be done?) shine much-needed light on the information managed.
I think this is where we can lend very strong leverage. How is information protected, what types of things are important to manage, what types of systems are needed? This goes beyond compliance and security technologies, though they are critical of course. Included in this is the responsible management of systems.
I actually have seen some recommendations where the suggestion is made that systems are so cheap, so easy to line up, that no real disaster planning or management is needed. Just duplicate everything several times over and voila! All things are protected and will always be available. That’s it. Nothing else to see here.
Still more talk about the lack of a need to pay attention to queries, infrastructure, design and such. Since systems are, again, so inexpensive, any lack of performance or utility can be overcome by throwing new physical resources at the issue.
What do you think? Does big data (and all that that means), mean a different or enhanced code of ethics is needed? Or… is it more, perhaps much more, of the same?