In some ways, Hadoop is a like a fine wine: It gets better with age as rough edges (or flavor profiles) are smoothed out, and those who wait to consume it will probably have a better experience. The only problem with this is that Hadoop exists in a world that’s more about MD 20/20 than it is about Relentless Napa Valley 2008: Companies often want to drink their big data fast, get drunk on insights, and then have some more — maybe something even stronger.