Long time ago, in a galaxy far away… there was OGSI, there was WSRF, there was HTTP, there were services… and there were heated discussions :-)
Ian Foster suggested that we write a paper together explaining our different approaches in building applications which deal with “state”. If my memory serves me well, that took place a couple of years ago. We weren’t 100% in-sync back then and the effort slowed down to a standstill (I was to blame mostly for giving priority to other things).
Few months ago, Ian suggested that we revisit the paper and took the initiative of producing a new draft, with which he did a great job. After few edits, the “How Do I Model State? Let Me Count the Ways” article (by Ian Foster, Paul Watson, Mark McKeown, and myself) was sent to the Communications of the ACM and now appears in the September issue.
With that, Ian, Paul, and I agreed to be put to rest the discussions around services and state (we couldn’t contact Mark to confirm with him too but it’s a safe bet that he’s moved on as well). It’s funny how, today, it feels like we were investing too much energy on details but, as Ian told me, those details seemed important at the time :-) It has been a very interesting and exciting road to here!
It was great working with Ian on this and I am really looking forward to working with him on new things. Clouds vs Grids perhaps? :-) (just joking)
BTW… for the record… HTTP emerged as the clear winner when dealing with resource state :-) (at least in my mind :-)