« Securing Your Databases | Main | Cover Letter Inspiration »
Storing the Juicy Details
By Eric Gross | October 26, 2007
I have been working recently on what is, for me, a large-scale project. I realize that this pales in comparison to the Manhattan Project and sending people to the moon, (US Progress of the past, Right Now) but still - for me this has been quite an enhancing experience. Our software product requires that we understand the checks and activities including and surrounding database provisioning so that we can automate all of these tasks. I knew little about what was required at the outset - I am familiar with the concepts around the database as well as automation principals, but this project deals with delving into hundreds of documents, thousands of lines of code, and some information dating back 10 years or more.
The first few weeks involved delving into the depths of existing processes documentation and code. Had we not created a simple repository for the details collected we surely would have failed. Culled from many of our customers all over the globe was a few hundred ‘activity pearls’ that would satisfy the stakeholders. No single party had a grip on the entire puzzle - some had a fuzzy wide-angle perspective of one regions code, some specialists had details about the implementation of some details, some customers had information about this and that - and so on. It was up to us to form the patchwork quilt from a million threads. Each item was represented in a spreadsheet alongside all key identity points, and when necessary a directory for each item stuffed with all the juicy details.
The key lesson for me during the early phases of the project was that every detail must be stored in the proper location in a spreadsheet or file structure in a logical way. Search is vital for finding reference data, but it’s only useful if the details have been saved. Emails are a good last resort for missing details, but using your Inbox for your filing cabinet will cause failure when the project is sufficiently complex.
Implementing, testing, and integrating the lot is also proving to be very interesting - more on that later.
Topics: Data Management, Organization