4 ms·
I once had a coworker who, when working to pointlessly extend the lifespan of an obsolete external tool, wrote a translation layer. Then he stopped and re-imple
by Kalium 7y ago
I once had a coworker who, when working to pointlessly extend the lifespan of an obsolete external tool, wrote a translation layer. Then he stopped and re-implemented it in Rust. This added nothing to the tool except time to the release schedule.
I don't eschew new technologies. But increasingly I need to see an expected value to justify the uptick in cost.
- hvidgaard 7y agoWith wisdom you come to the conclusions you have. The value is somewhere in between. You shouldn't mindlessly reimplement working tools, but you shouldn't keep patching old tools either. Somewhere in between, you project the current and future requirements. Sometimes you have the old tool on life support, sometimes you restructure parts of it, sometimes you reimplement it from scratch. You will often make a wrong decision and that is okay - you just learn that minimal work is often the best way forward unless you positively know otherwise.