Much has not been discussed recently on the growing corporate silos of business users who are taking back control of their own IT legacy systems and budgeting within their own budgets for future enhancements or even worse, new applications all together. This is particularly true within the BI space. Having moved to "the dark side" only a month ago, but well versed in the BI space, this was extremely true for my own personal migration. Having grown tired of waiting for project prioritization to bubble up my project request for scorecards, but before going truly rouge and renegade in my approach, I went on an approval campaign about 3 years ago now. Huh?
An Approval Campaign – where a business user goes IT group to IT group, asking whether they would a) be interested in building a scorecard app or any app for that matter, b) could do it within my 1 year time frame, c) could REALLY do it within my 1 year time frame and d) would be willing to support it within an IT regulated production environment. D was the only taker, and really is the only one that really is a win/win for both parties.
What the approval campaign gives you is a CYA mechanism; the ability to point out years later when the label "under the desk shadow app" gets placed on your pride and joy platform, that you DID, in fact, seek out the ‘proper channels’ and was turned away by the not so subtle use of lower priority terminology to basically elicit a response similar to a canine’s hackles going up when one climbs into their territory. While it may seem innocuous, it is anything but that. It is a way for IT to scapegoat, to blame, and cast aside any wrong doing in service disruptions; to enable them to point fingers and deny any culpability to the eventual service issues that will arise ( and believe me they will). This is exponentially compounded by those orgs who have undergone a greater than average mass exodus of employees, aka your corporate attrition, without a solid means of knowledge sharing or succession planning on the parts of HR, the employees or managers enforcing those departures.
Seems like a conundrum beyond means of course correction, or does it?
Let’s start to draw some parallels with the concepts of ‘Open Source’ and ‘SOA for BI’ in terms of a fable that I am going to share with you:
~ Once upon a time, in a land far, far away, there were many software companies and consortiums, who would hire very intelligent engineers to help them build their ‘visions’ of the future. Their constituents, "the business", kept changing priorities and drove this ‘flavor of the month’ sense to development that no one knew what to concentrate on, and thus, all did it ‘a li’l bit different’ — standards, what standards, was our core value, along with the notion that documentation is for the weak. Some people got so frustrated with the hard coded system integration and recoding of work, that they left the kingdom, and so left the knowledge of the architecture supporting the kingdom. Even worse, the evil dragon named nonstandardized, continued to burn any of the attempts to create pocketed standardization guidelines because their wasn’t any support beyond the fiefdom’s attempts at the kingdom level.
But all was not hopeless in the land of spaghetti coded platforms – Oh no… Why? Because as long timers left, so came newbie’s with their unabashed impression that by architecting a retooling of software like our friends next door in the land of best practices were doing, levering open standards, modularized applications as services with standardized interfaces and most importantly, a means to address the evil King, ‘Disparate business process’.
And lo and behold, one day, a prince, named middle-ware, came in, riding on his noble widget, breaking down anyone claiming "but we’ve always done it that way" in his path. His ride took him from software land to software land, breaking down silos, and smoothing business processes in his way, that were once isolated and forced into their siloness due to the technology barriers that existed between the loyal tenants of this kingdom’s applications.
And he dropped his widget onto a vast and empty space that had standardized the look and feel to resemble all other spaces schema. But instead of mirroring the space next door, who placed his hub widget in the middle of the space, leveraging all offered supporting informational spokes around the perimeter, he placed it and nothing else at the top of the space, and called his new land, the land of ‘Scorecard’ — he enabled the same level of information for his people by linking and drilling deeper into this primary scorecard widget, and by doing so, creating a space worthy of the King’s time (short, sweet and to the point), but also gave a path to deep dive into the meaning of his widget for those users more accustomed to seeing more, not less information."
And people started to follow. From one space to the next, the kingdom underwent a radical face lift, and together the silos, without realizing they were conforming, started to conform; but they didn’t live happily ever after….
Because the root cause wasn’t addressed with why the silos initiatives happen in the 1st place – Thus, the open source concept that I am so passionate about enters. Little has truly been discussed about this space….Part 2 will talk about how to leverage an open source platform with an SOA for BI infrastructure and by doing so will not only stop the business from their own silo’ed initiatives but also, give your developers the ammunition to build and customize apps while adhering to your set BI standards and guidelines.
PS – Did I say I love my new job…?