Went to a great talk by Anand Babu (known as AB) at the TiE Silicon Valley on Tuesday. AB co-founded Gluster and as well as that seeming like a very neat way to solve some of the strategic big data issues that I can see for AdTotum on the horizon and to make my life much easier with some 'massively big data' AI projects that I've had up my sleeve for a while, AB brought up some great ideas on how to robustly structure new super-high performance technologies. His example of how the compression in Gluster is coded fit in exactly with what I've been trying to achieve with the AdTotum API and indeed all the code that I think has any extended lifetime. Of course this might mean that I should be thinking of moving from Ubuntu to Red Hat.
Very good insights into solving problems by design and about how so called experts - who gain their expertise by experience rather than by deep understanding, often have insurmountable barrier to what they can do due to being locked in to existing wisdom.