Docs: HTTP Client Tutorial#617
Conversation
|
@arnetheduck I've finished the tutorial. It took me much longer than I had expected but at least I am now sure I do understand how to do HTTP requests with Chronos. |
|
@arnetheduck the CI is failing but it doesn't seem to have anything to do with the docs. |
|
Since it's a non-code PR I'm not going to wait for a formal approval and just merge it on Feb 20. |
Since this ends up being a recommendation for how to use chronos, let's wait until it's been reviewed - ie before we commit it, we should also probably take a look at whether we can simplify the http API itself so that the tutorial itself becomes more simple, ie let's use this PR as a way to talk about those simplifications. |
OK, sure, let's do that.
IMHO simplifying the API and improving the docs could be done independently as parts of a bigger process: update the docs, collect feedback, improve the API, update the docs, etc. I.e., the docs could accurately reflect the current state and that would be valuable per se. They also could inspire some changes in the API but that's a different independent value. I always get nervous when I see a PR depending on another PR, I'm scared that the scope will grow out of control and the task will take more time than it should... But maybe I'm overreacting. Anyway, I'm trusting you with the way to treat this PR, all above is just an opinion. |
It's a bit like introducing a feature and discovering that the codebase needs refactoring - we tend to extract the refactoring and do that first so that the feature can be kept clean, using the feature as a driver and motivator for the refactoring. In any case, I expect both docs and API updates to go out in the next release together, so it's not far off anyway - that, and even without http api changes there are things to address in this tutorial (though a proper review needs a bit of time) - users anxious to get going can simply look at the branch in its current state and get started already. |
No description provided.