Junk Drawer What?
“Junk-drawer architecture” is a term I use to refer to architectural artifacts that have no formal or centralized repository. Similar to the way we rummage through a junk drawer to find a relevant knick-knack, architectures are often crafted in numerous formats then circulated to an absurd number dissimilar locations.
Need an example? A junk-drawer architecture is that PowerPoint of current CRM business processes a business analyst emailed you last week. It’s the Visio diagram of the network those consultants drew up six months ago. It’s that application portfolio spreadsheet your vendor manager asked you to complete by end of week.
What’s the Big Deal?
Junk drawer architecture is inefficient and wasteful. Assuming someone takes the time to create an artifact, that artifact clearly has a dollar amount attached to it, and that money is wasted if you can’t find the artifact. If the diagram, flowchart, or business process is sent out in email then saved to some obscure network location, one could argue it lives in multiple places. But what about new employees who missed the original email? They could argue that the artifact doesn’t exist at all, and they’d be right!
Yet another reason junk drawer architectures are so wasteful is due to the expensive software we use to create them. Microsoft PowerPoint, Visio, Adobe this-or-that. There are so many! This is the era of the web, so what’s with all this desktop software?
How Life Gets Better
Killing junk drawer architectures is a two-step process. First, build a common repository for your architectural artifacts. The solution should be in the cloud (so it’s accessible to all), secure, and entirely web-based. Everyone should have read/write access without needing to “install” anything and the solution must work for all computing devices. I’m talking Mac, Windows, and Linux. Mobile devices mustn't be forgotten, and should have at least read access as well.
Secondly, you need a standard for information organization that is initially devised and subsequently reviewed by a curator; ideally your friendly enterprise architect. I’m not referring to the ivory tower types who want to impose their overly-rigid “guidelines” on content taxonomy. I’m talking about “get-it-done” types who understand that enterprise collaboration of the masses is far better than tightly-controlled edits by an elite few.
Moving to an online, collaborative standard won’t happen overnight. It takes time for people to change. However, they’ll never change if you don’t provide the path to improved efficiency.