Neil’s comments on my blog entry about whether IONA’s idea of an extensible ESB is a good thing were very informative – thanks Neil.
I was particularly interested in IONA’s value assessment program referenced by Neil; anything that helps users justify investments and assess/measure results is most welcome and worthy of applause. I shall take a closer look at this – expect another blog entry on it in the near future! And before I go on, just in case anyone thinks I am having a go at IONA, I want to make it clear that I actually think Artix has some definite technical competitive strengths.
Having said that, I don’t want to let go of this ‘extensible’ point yet. The points Neil makes about incremental investment and deployment, as opposed to ‘expensive infrastructure build out up front’, are based on one of the raisons d’etre of ESBs. By definition, an ESB has distributed intelligence and allows implementations to be built up gradually, with additional investment based on success at each step. The key to my argument is the packaging chosen for this ‘extensible’ ESB. Because functionality is broken up into value-add options to quite a low level of granularity, even when a new node is brought into the ESB domain there may have to be additional investments as functional usage increases. Other ESBs tend to package up a lot of that functionality at one price. That isn’t to say that others do NOT ask for more money for functional increments, but simply that the increments are more coarse – eg I want to start using orchestration now. My concern remains that, although the IONA approach gives more flexibility, this can be a two-edged sword. Of course, a user does always have the option of ensuring the basic order includes all the features that are expected to be needed at least in the medium term, but then the question is how that affects the price against competitive offerings.