Monday, 10 September 2012

To Fix or Not To Fix...

1 comment:

Dom Finn said...

<div>Agreed it's not always completely possible but you can often do something.?</div><div><br></div><div>Even if that something is just:?</div><div><br></div><div>Refactoring a specific function, tidy the code up, return early instead of late etc... as long as you are keeping to the return contract of the function you can't go far wrong.?</div> <div><br></div><div>Renaming, rename variables from foolishness like cuIDoW to CustomersIdOnWednesday (at least you can identify the foolery better then).?</div><div><br></div><div>Indeed, there is the risk of making it worse. Sometimes the spaghetti is so bad you can barely tell what the knock on effects of something are but you can still pick some small safe battles to fight and at least begin to identify what is rotten for the future.?</div> <div><br></div><div>You know the future of the project, the future is that it's coming back to get you. Like some foul curry beast from the night out of yesteryear. If it's not you it will be some other poor fool that is tasked with keeping it going 5 years after most of the team has moved onto other systems.?</div> <div><br></div><div>Is it the code you wrote yourself? Probably. Even if it's not, you can't generally blame the people involved, we are all on a journey. Although there's always the occasional Orc blasting out lol-code.</div> <div><br></div><div>As I say, it's just down to you what you do after you do know what you're doing. Do something, doooo it!?</div>

Comments System

Disqus Shortname