Re:Designing very large JavaScript applications
by aje403 on Friday April 20, 2018 @05:56PM
(#16888742)
I think "How confusing is this going to be to the one consuming this" (assuming you're not writing the front end as well) is the more important question to ask
[ Parent
| on HN ]
Re:Designing very large JavaScript applications
by joeyespo on Saturday April 21, 2018 @11:22AM
(#16891737)
This way of thinking was my main motivation behind Grip [0]. Use Readme-driven development and "play" with the API--whether it's in the form of a CLI, a library, or web API-- through usage examples.
I've found this to be a wonderful way to get into the heads of dev users before you approach them directly.
Re:Designing very large JavaScript applications
by abritinthebay on Friday April 20, 2018 @06:15PM
(#16888827)
And the one most rarely asking in my experience too
Re:Designing very large JavaScript applications
by silversmith on Saturday April 21, 2018 @07:44AM
(#16891052)
In my opinion, asking this question is the key benefit of test-driven development. You get to be the consumer, write down your expectations, and then work backwards from yours-as-consumer expectations.
Re:Designing very large JavaScript applications
by sbov on Saturday April 21, 2018 @08:02AM
(#16891107)
I find the opposite tends to happen in Java. In the pursuit of testability, people make libraries much harder to use and understand.
Re:Designing very large JavaScript applications
by purerandomness on Saturday April 21, 2018 @11:06AM
(#16891681)
Writing good tests is a skill that has to be developed, just like anything else.
Re:Designing very large JavaScript applications
by abritinthebay on Sunday April 22, 2018 @01:59AM
(#16894913)
It should be, but too often people write bad tests that test whst they wrote and not what it’s supposed to do.
Writing good tests is a skill and most developers don’t seem to be as skilled in it as they think.
The same concepts power much of Agile too though: XP & Scrum both complement this.