Clearly the world is full of the news and gleeful sniping at Google’s little fopar on Saturday 31st January. What I have not seen quite so much of is comment on the process, or perhaps lack of, that allowed such a catastrophic (for Google) human error to go “live”.
It seems that the need to get it out is greater than the need to get it right (and I am sure that anyone with a blog or website has managed to make an occasional fopar too), forgivable, perhaps, for the legions of us who are posting for leisure and social purposes, not so for professionals and corporates who should know better. Whatever happened to the process of development, test, review, authorise, release?
Let’s not just focus on Google, there are other newsworthy problems out there this week. Social bookmarking site Ma.gnolia was taken down completely on Friday (30 January) citing data corruption and loss “that will take days, not hours” to fix, going on to say they cannot guarantee that users bookmarks can be restored at all. Now imagine if this were the web services you most depend on.
In the case of Google we now know that human error was the stated cause, in the case of Ma.gnolia one can only speculate. Whatever the cause there is little excuse for a lack of process and infrastructure to prevent (Google) or recover (Ma.gnolia) in a managed way.
As we move towards more use and reliance of cloud based computing, as we place more of our “stuff” on social networking sites and data repositories we must assure ourselves that the sites and companies we put our trust in are up to the job.