Time. One of the grandest illusionsof the universe. And yet it leads us all to believe that another week has started. In other words: time for a soundbyte! This week featuring ‘Slice of time’ by David Crosby. It’s from the album Croz, which contains some great and timeless songwriting.
“My teacher said time is elastic
I wonder just what
I’ll find in it”
Ungraspable as time might be, it is one of our most precious resources. Take a moment to ponder before reading on. What did you spend your slices of time on last week? How will you use your time to achieve your goals this week?
“Pluck out a day
A week or an hour
HOLD IT UP, HOLD IT UP TO THE LIGHT”
With the Luminis Technologies crew, we decided to experiment with a new way of sharing how we spend our time. Since we work remotely and on different projects, it’s a challenge we need to tackle. So every week, we’ll post an internally readable snippet of what we’ve worked on and accomplished that week. Everyone contributes and can feel the pulse of the whole, regardless of which project you’re working.
So let me kick off that habit in public. The past weeks had several highlights, including receiving another JavaOne Rockstar award for two of my JavaOne talks. Also, we worked very hard with our team to deliver a first release of a new project. Another highlight was speaking at Jfokus, together with Paul Bakker. For more than just a snippet, read the upcoming Conversing Worlds which features a Jfokus experience report.
“Freeze the frame,
Really look at the faces,
With all of your sight”
One of the talks I gave at Jfokus and JavaOne is about event-sourcing. As developers, we suffer from a fascinating compulsion. That is, the compulsion to model data in our application in an atemporal manner. We put data in databases where we create, update and delete to our heart’s content. The database forms our shared, mutable state. We all know the perils of shared, mutable state. Event-sourcing turns this model around, by recording immutable events that capture all changes in your application, rather than only tracking the current state. In other words: get all the facts straight for each slice of time, and the rest will follow. If you find this stuff interesting, you can watch a recording of the talk on Parleys.com.
Another trend that struck me at Jfokus is the massive increase of scale that modern web-applications experience. What we are seeing is a corollary of Parkinson’s law in action on massive cloud infrastructures: ‘data expands to fill the space available for storage’. Datacenters are the factories of our computing revolution.
“Images, images, images, images
Telling the truth to us all”
Facebook recently built a sixth datacenter. Its one and only goal? To serve as ‘cold-storage’. In other words, data in there is seldomly requested, yet Facebook doesn’t want to lose it. With 350 million (!) images uploaded each day there will be a staggering amount of data for cold-storage. It’s mind-boggling that Facebook is building a dedicated datacenter for storing cat pictures nobody cares about. (And before you know it, your Facebook pictures are deleted by some random stranger anyway.)
“The land going by seems level
But really the tracks are
Reading facts like this, it almost seems like we reached some form of post-scarcity economics for ‘big data’. Well, that may be the case for actually storing enormous amounts of data. But not so much for actually understanding it and making good use of said data. We are massively underestimating the new skills we need for this as an industry. Yes, there’s a lot of hype around data science. But don’t confuse that with a lack of substance. As developers we need to get serious about data, not just code. In that vein, I can highly recommend a course Mining massive datasets running on Coursera. I’m currently enjoying the lectures and assignments. And there’s plenty more where that came from.
All this talk of data inevitably makes me think of the privacy implications. I don’t use Facebook. Not because I don’t think it’s useful in some ways. But to me, the type of information exchanged coupled with Facebook’s privacy policies just doesn’t sit right with me. On the other hand I do use Twitter and LinkedIn, but only in a professional capacity. Throw Google into the mix and the line becomes blurry to the point I’m not even sure who or what I’m trusting anymore. Fortunately, the paper (pdf) ‘On the duality of resilience and privacy’ gives me a shimmer of hope. May be there is a future that is native to the true roots of the distributed web. A future that doesn’t exhibit the privacy quagmires of centralized cloud services. I do encourage you explore for yourself where you draw the line between evidently useful and ‘slippery slope’ creepy for the services you use today. Who owns your slice of time?