Whether we’re an industry veteran or a newbie to a company, we all have to deal with the legacy system that No-one Knows, Just Works and Ain’t Broke so Don’t Fix It. But what happens when it does go wrong, or the server falls over, and suddenly that old single point of failure becomes a business priority? In this talk, Christopher will cover a variety of techniques to help you understand the cruft and ensure you can modernize it efficiently and without (much) fear.
Last active
August 29, 2018 10:40
-
-
Save choult/e0a216472cfee9ed3a295efcdc2a7711 to your computer and use it in GitHub Desktop.
Orchestrating Manoeuvres from the Dark
In the UK 'newbie' is a contraction of 'new boy' in public school slang, so people who object to phrases like 'guys' might object to 'newbie' as well.
@jimholmes Fair points, of course - I suspect this will shift more towards a CI, monitoring and logging talk, and it was certainly more than just a legacy talk.
@chrisseaton Being schooled at a grammar in the UK myself, I'd not heard of this explanation for the derivation (we called our intake year "bugs"), and certainly https://en.wikipedia.org/wiki/Newbie provides multiple explanations for the word, but I'll happily work on dropping it from my vocabulary if people feel it is inappropriate.
I'll just toss in my +1 to the above feedback. @GeeH and I both reviewed a similar abstract earlier this week.
@elazar Yeah, I think shifting it from a legacy talk to a "What's going on in production?" talk is more appropriate anyway, it's just that there's an overall suitability of the ideas for modernizing.
I suspect the final abstract will be significantly different :)
@choult I agree with the other points about distinguishing yourself from other legacy system talks. Abstract seems very broad, perhaps also offer some of the main takeaways? Also, you probably are already aware, but to me, the two bibles when dealing with legacy systems are Martin Fowler's Refactoring and Micheal Feathers' Working Effectively with Legacy Code. Perhaps those are worth a skim and see if there's something of interest.
@luijar thanks for the tips; I was definitely going to cover Strangler, so Martin Fowler is certainly on my radar :D
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
There are a fair number of people talking about dealing with legacy codebases/systems. You need to differentiate your talk from those others. As an organizer, what's unique and valuable about your talk? As an attendee, what am I going to take away from the talk that will help me deal with that awful system when I'm back at work?
As others mentioned, you need some specifics to flesh out the abstract. You might also consider laying out what "legacy" means in your context. For some it's an age of the codebase thing, for others it's any codebase without automated tests. For others... You get my point.