Why are you even bothering? What are you trying to solve? • Does the state of your current application make it difficult to add more features? • Is it hard for you to recruit because the job market for the tech you use is shrinking? • Make sure you have a good reason for tackling this large tech debt problem before you begin.
a stop the world solution and one that is rarely a good idea. • It takes a lot of time to get to parity with your existing application. • Hard to rediscover all the edge case bugs that were fixed over years. • Stops work on new functionality. Which can be hugely damaging to a young company. • There are times when it's the right call… • …but those are the exceptions, not the rule.
migration • Having an Evergreen Legacy Application means your old code and new code will exist alongside each. Possibly forever. • You have to be ok with living in this hybrid world. • Prepared to take a small cost when switching context between the old and the new. • It's ok to not have an end path to removing legacy code, but you have to be ok with living in a world of flux
to do it • Rule 2: Don't stop and rewrite everything • Rule 3: Prepare to be in a state of constant migration • Rule 4: What is new today is legacy tomorrow
to do it • Rule 2: Don't stop and rewrite everything • Rule 3: Prepare to be in a state of constant migration • Rule 4: What is new today is legacy tomorrow