Making ourselves heard
I love Peter Lacey's 'They can’t hear you' piece. He really nails why enterprise's don't get REST, and why they are so WS-Hairball (love that term) oriented. Quite simply it is because that is what the enterprise vendors are selling to the CTO/CIO's, particularly on the back of their over-hyped SOA marketing themes.
REST would also be very difficult for those vendors to sell, they like to sell middle ware platforms, the complexity of the WS-Hairball is perfect for these kinds of plays. Often these vendor oriented enterprise platforms are layered like onions, encapsulating the various levels of complexity introduced as standards by all of the vendors in this space, Microsoft, IBM, SAP and Oracle et al (MISO). In other words MISO writes the standards, delivers the platform and then charges for the privilege of cooking with the onion. Even worse if you are an outsider trying to peel the onion it tends to end in tears.
Curiously though I do understand many of the issues that the WS-Hairball is trying to solve, having been there with clients and projects over the years. But like many big ideas or projects, the WS-Hairball is trying to solve something too complex and too wide-ranging to fit into any mere mortals mind, particularly whilst juggling all of the other project goals and complexities. Hence the attraction of the middle ware toolsets and the SOA magic pixie dust transit system, has the enterprise CTO/CIO's lapping it up.
What is even more disturbing is that once implemented it often grows unwieldy and results in expensive outsourcing to those same vendors (and there lapdogs) to sort out the new layers and complexities, this pattern has happened for decades - rinse and spin. Thats the trouble with the WS-Hairball just like previous middle ware generations, these fat complex beasts are to I.T. what fly-by-wire planes are to aviation, they are too complex for us humans to operate. That to me is a mistake, technology should make things simpler and this is particularly true of business development software, otherwise we will end up with even more layers of legacy in our enterprises.
What we have seen is the emergence of agile tools and social software in the web world, far more complex than the enterprise, but yet kept simple in order for us humans to try and manage it from the bottom up. Simplicity works, us mortals can get enough of the problem space and tools into our heads at anyone time to edge things forward piece by piece. The enterprise tries to build the big picture from the top down, but never finishes delivering, like painting the Golden Gate bridge. Each enterprise iteration more grand and complex than the one before, with ever more powerful vendor platforms promised, poised like some sort of super power conflict escalation trying to take on the unwieldy information technology beast. But I am sorry to say this goes no where, and solves nothing, except for the vendors insatiable appetite for revenue and relevance. The enterprise needs to adopt some of these technologies omitting the WS-hairball. But most of all the enterprise needs to unravel it's plans and trust the little guys implementing at the bottom in conjunction with the participants (enterprise employees) and start building bottom up, middle out and solving the real issues item by item, requirement by requirement in agile fashion, guess what it freaking works guys c'mon!
But I diverge from the point and should respond to Peter's great post, we need to work together to try and penetrate that CIO/CTO + vendor shield (enterprise reality distortion field) and counter those vendor arguments. Let us not waste time with posts around REST vs SOAP (good call Peter) , but rather converse around delivery of alternatives to the WS-Hairball. I for one am looking at how we can help utilise agile tools that work both in the web space and the enterprise space (watch this spot for more soon). For that we will need both REST, RSS and SOAP, but we can avoid the WS-Hairball trap and concentrate on delivering rapid, incremental benefits that have clear return on investments. But for this to happen the CTO/CIO's will need to be a little brave and step out of their safe enterprise vendor zones, we know the adage 'no one ever got fired for buying MISO..'. Rest assured we at Folknology will be making ourselves heard over the next few months offering views outside of that safety zone, if you're brave or know someone that wants to be invite them along for the adventure.
Technorati Tags: Business2.0, Enterprise2.0
