intertwingly

Please elaborate - Are you dissing Raindrop?

Exhibit A’s in both mine and Joe’s posts (i.e., Unicorn and Raindrop) very much appeal to me.  “Values ... regardless of the actual technical architecture”... not so much.

But these two things aren’t even remotely on the same level, while the process-vs.-thread discussion certainly is.

Not the same level is exactly the point.

Stefan: I understand your point of view.  I’ve been there myself.  Enough times that I see the pattern of the types of solutions that approach ends up with.

I very seriously doubt that any reasonable application of a SOA approach to Raindrop by Mozilla personnel would have ended up with an Erlang (process based, BTW) server for Raindrop persistence.  But no-one can deny that a JSON, JavaScript, and REST based API is a perfect fit for an important class of applications, apparently including Raindrop.  Similarly, compare github’s application of Unicorn to the comments on Joe’s blog.

I’ll make what likely will be startling statements: there is nothing wrong with SOAP or WSDL.  HTTP, and by implication REST, can often be very ugly.

But the approach to the problem that latter two tend to encourage inevitably ends up being far superior to the approach to the problem the former two tend to encourage.  It totally is an I wouldn’t start from here thing.

But in any case, peace.  We each decide how we wish to spend our time.  I hope you find your efforts there rewarding.  Where you are spending your time is an area that I personally chose to avoid.

Fair enough.

Stefan, now that you have posted your comments on the SOA Manifesto, I feel even more like the right answer is: RUN AWAY!  I can’t even get past the first point.

Business value over technical strategy

False dichotomy. 

This may seem very obvious, yet we see this value violated all the time. People build up grand theories of how things should be, or come up with huge and complex products that attempt to solve all imaginable problems.

Begging the Question.

Clearly only those despicable technologists come up with complex products.  Things like Unicorn and CouchDB.

But however great that theory is, however powerful a product, the focus needs to be on delivering business value. (To use Anne Thomas Manes’s words, focus on services, not SOA). We put this at the very beginning to emphasize that we consider it to be the most important value. Note that adhering to a technical strategy (the right side) is also a value - but when in doubt, business value wins all the time.

Pop quiz: pick the best alternative from the following choices:

• Simple, feasible, and modest return on investment.
• Complex, impractical, and very lucrative.

As I said, I’ve seen the types of solutions a “business values wins all the time” approach end up with.

For my tastes, I continue to prefer “A”.

I agree < I feel even more like the right answer is: RUN AWAY!...