We're all familiar with things like HTTP codes and content types, but there's so much more we can do when developing an API to make life easier for consumers. How many times have you used an API only to find out that every endpoint is slightly different – some use `snake_case`, others `camelCase`, sometimes the field is called `id`, sometimes it’s `user_id`. How about pagination? Error responses? What about API documentation? Trying to standardise on all these things can kill an engineering team. There are so many options out there it’s difficult to know where to start. Come along and learn what works for our team! We’ll cover contentious topics (should the version be in the URL or a header?), lesser-known standards that are great (RFC 7807 springs to mind) and a couple of things that aren't an issue right up until they’re a really big issue (like pagination).

Comments

Comments are closed.

Gareth Ellis at 17:04 on 22 Feb 2019

So much practical and pragmatic advice from an intelligent and concise speaker. Thanks Michael!

Alex Youngs at 11:52 on 23 Feb 2019

Good topic and nice examples. There were some good nuggets of advice throughout though I felt once or twice the slides disappeared a bit quickly.

Michael is a confident and knowledgeable speaker and delivered a good overall talk - thanks!

Interesting talk, especially about the lifecycle :-)

Peter Fox at 23:29 on 24 Feb 2019

This was my top talk of the Friday. API design is still something I struggle with and I've yet to see a talk delve into APIs like this one did, focusing on providing real insights.

It's a shame it wasn't a main talk as I think there's something in there that most developers won't know about.

Also just a really good, clear speaker in general.

awsome one