Best of this Week Summary 21 September - 27 September 2009

• Always feeling a bit less up-to-date when using RSS/Atom feeds? Well, convince your RSS/Atom provider to implement the PubSubHubBub protocol: it provides a callback-hook, which when registered to as feed subscriber, causes an immediate update (ping) to the subscriber (instead of waiting for the next RSS/Atom feed request).
  Or will rssCloud be the winner of this realtime RSS extension? Here's the basic (simpler) flow of rssCloud:
  
  
  Differences between the two are described here (including some issues) and here. A real old-fashioned flame-war has even started between the creators of the two protocols.
  And a good comparison which explains quite well why PuSH is the better choice of the two. Note that both protocols seem to be server-to-server only! (e.g think when you're behind a firewall or NAT...)
  See this article for an explanation of the PuSH flow of notifications/information in below's diagram:
  
  
  Update: PubSubHubbub explained by co-creator Brett Slatkin (an engineer at Google).
  
  
• Want to get to know Hadoop? In short "Hadoop allows you to write and run your application in a distributed manner and process large amounts of data with it. It consists out of a MapReduce implementation and a distributed file system." Check the rest of this introduction for more details.
  


• Five Java anti-patterns to prevent (out-of) memory problems.
  


• Three Best and three worst practices in BPM and SOA.
  


• "Tom Killalea, Vice President of technology with responsibility for infrastructure and distributed systems engineering at Amazon.com wrote an article on ACM queue on building scalable web services. He outlines guiding principles to building scalable web services with a lot of real-world examples, the core theme of which is “build only what you need”."