Development methodologies: Difference between revisions
From ym2149.org
Jump to navigationJump to search
No edit summary |
No edit summary |
||
| Line 1: | Line 1: | ||
* Sweep it under the carpet-driven development - never investigate any root cause, just throw code at the screen until the problem appears to go away | * Sweep it under the carpet-driven development - never investigate any root cause, just throw code at the screen until the problem appears to go away | ||
* Plate spinning-driven development - we don’t have time to write automated tests, instead n developers must perform an increasing amount of incomplete manual testing forever | * Plate spinning-driven development - we don’t have time to write automated tests, instead n developers must perform an increasing amount of incomplete manual testing forever | ||
* Fear-driven development - tests would expose bugs so don't write any, deployment to prod would expose bugs so let's not do that, and so on | * Fear-driven development - tests would expose bugs so don't write any, a cool new lib would break things as we don't have enough tests so ban that, a cool new lib would confuse our developers because we treat them as a liability instead of investing in people, deployment to prod would expose bugs so let's not do that, and so on | ||
* Pearl clutching-driven development - any hint of creativity must be extinguished | * Pearl clutching-driven development - any hint of creativity must be extinguished | ||
* Gotcha-driven development - instead of publishing guidance to empower teams to avoid problems, send the results of scans straight to the top of their backlog without context | * Gotcha-driven development - instead of publishing guidance to empower teams to avoid problems, send the results of scans straight to the top of their backlog without context | ||
| Line 8: | Line 8: | ||
* Playing the security card - get a team to drop everything to fix a nebulous security issue, never explain the impact of leaving things as they are | * Playing the security card - get a team to drop everything to fix a nebulous security issue, never explain the impact of leaving things as they are | ||
* Aspect-oriented programming - an admission that the code is unmaintainable, and commitment to keeping it that way | * Aspect-oriented programming - an admission that the code is unmaintainable, and commitment to keeping it that way | ||
* Reddit-driven development - FIXME | |||
[[Category:Programming]] | [[Category:Programming]] | ||
Revision as of 12:28, 22 March 2024
- Sweep it under the carpet-driven development - never investigate any root cause, just throw code at the screen until the problem appears to go away
- Plate spinning-driven development - we don’t have time to write automated tests, instead n developers must perform an increasing amount of incomplete manual testing forever
- Fear-driven development - tests would expose bugs so don't write any, a cool new lib would break things as we don't have enough tests so ban that, a cool new lib would confuse our developers because we treat them as a liability instead of investing in people, deployment to prod would expose bugs so let's not do that, and so on
- Pearl clutching-driven development - any hint of creativity must be extinguished
- Gotcha-driven development - instead of publishing guidance to empower teams to avoid problems, send the results of scans straight to the top of their backlog without context
- Emoji-driven development - replace logging critical to investigating incidents with pretty logging, or no logging at all
- kim jong un-driven development - so much critical info only exists in the engineering manager’s head you have to frantically jot it down when they’re speaking, and don't dare log off until they've stopped
- Playing the security card - get a team to drop everything to fix a nebulous security issue, never explain the impact of leaving things as they are
- Aspect-oriented programming - an admission that the code is unmaintainable, and commitment to keeping it that way
- Reddit-driven development - FIXME
