Engineering Rewrites and Old Code

Closely related to my Balls o' Mud post about old code bases, the latest Silicon Valley Product Group blog posting is about the necessity of engineering "headroom" for code maintenance during product lifecycles. And about inevitable rewrites to scale, perform, and I'd say, accommodate design of new features, although the article doesn't focus on that issue per se.

I knew ebay had gone through growing pains, but the article suggests some absolute heroics in the form of two mid-growth rewrites of their code that barely touched their business performance. I know of some (plenty) UI designers who've left ebay because of their unwillingness to touch their UI (it sadly needs it), but I do applaud their engineers for their effectiveness at communicating the need to do code work to survive. From the SVPG article:

The deal with engineering goes like this. Product management takes 20% of the capacity right off the top and gives this to engineering to spend as they see fit – they might use it to rewrite, rearchitect or refactor problematic parts of the code base, or to swap out data base systems, improve system performance – whatever they believe is necessary to avoid ever having to come to the team and say “we need to stop and rewrite.” If you’re in really bad shape today, you might need to make this 30% or even more of the resources. I get nervous when I find teams that think they can get away with much less than 20%.

If you are currently in this situation, the truth is that your company may not survive this. But if you are to have a chance of pulling through, you’ll need to first do a realistic schedule and timeline for making the necessary changes that engineering identifies.

If your product is slow, hard to modify, and your code base is really old and hard to understand by your own developers, you should be worrying about your future in the business. Can the people you want to modify your product succeed in actually modifying the product without total internal collapse, at version N? Here's another pointer to that ball of mud article on software architecture growth issues.