|
@Carnage4Life | |||||
|
Microsoft Research came up with an A.I. model to predict the number of bugs in a code base and the top predictors in order of accuracy were
- company org structure
- code churn
- code complexity
- dependencies
- code coverage
Anyone else not surprised?
augustl.com/blog/2019/best…
|
||||||
|
||||||
|
Sundar Narasimhan
@sundar_narasimh
|
19. pro |
|
Umm.. no mention of developer experience? Or problem/product novelty? Odd!
|
||
|
||
|
Terci
@Tercicatrix
|
19. pro |
|
Virtually every software project is novel.
Software is infinitely reusable, unlike bridges or commodities, so if something calls for a non-novel solution, the existing solution can be reused.
|
||
|
|
||
|
Steven Sinofsky
@stevesi
|
18. pro |
|
Literally the things we measured for 25 years.
|
||
|
|
||
|
Harold Combs
@DevHarryC
|
18. pro |
|
Given #1 is Conway's Law, #2 correlates to Brook's Law, no :)
|
||
|
|
||
|
John Merchant
@_jmercha
|
19. pro |
|
Software is a reflection of the organisation or individual that created it.
|
||
|
|
||
|
John the ArmchairTech
@thearmchairtech
|
20. pro |
|
Due to context switching, I consider an individual an organization. The me that writes the html is not the me that writes the Dockerfile or the me that writes the database model.
|
||
|
|
||
|
David A.
@dmatech2
|
19. pro |
|
Large codebases are almost analogous to large organizations. Both have various layers that need to interact and communicate, and there's "refactoring" being done constantly in organizations. It doesn't always work.
|
||
|
|
||
|
Rubberduck VBIDE add-in
@rubberduckvba
|
19. pro |
|
That's "refuctoring" though 😉
|
||
|
|
||
|
Downtym
@downtym
|
18. pro |
|
|
||