DX, API Design & Documentation

The State of Simple CRUD API Creation

10views

With all the talk of APIs you think it would be easier to publish a simple Create, Read, Update, and Delete (CRUD) API. Sure, there are a number of services and open source solutions for publishing a CRUD API from your database, but for me to just say I want a CRUD resource, give it a name, push a button, and have it—there isn’t much out there. I should be able to just write the word “images”, and hit go, and have a complete images API that I can add properties to the schema, and query parameters to each method. After ten years of doing this I am just amazed that the fundamentals of API deliver are still so complicated and verbose.

We even have the vocabulary to describe all of the details of my API (OpenAPI), and I still can’t just push a button and get my API. I can take my complete OpenAPI definition and publish it to AWS, Azure, or Google and “generate my API”, but it doesn’t create the backend for me. There has been waves of database or spreadsheet to API solutions over the years, but there is not single API solution to plant the seeds when there is no existing data source. Over the holidays I managed to create a Postman collection that will take my OpenAPI from a Postman-defined API and generate a AWS DynamoDB and AWS API Gateway instance of API, but it was the closest I could get to what is in my head across AWS, Azure, and Google. Why can’t I just hit GO on my OpenAPI, and have an API in a single click? Nio matter which cloud provider I am on!

The reasons why I can’t immediately have a CRUD API are many. Some technical. Most are business reasons. I would say it is primarily a reflection of our belief that we are all innovative special snowflakes, when in reality we are all saying pretty much the same things in similar ways—we just can’t 100% agree on the details. So we all just all continuing developing schema and APIs in isolation. I would say there is also a very narrow proprietary view of what APIs should or shouldn’t be, and the business and technical wizards don’t want it to be easy. It needs to be complex, and something that is done with specialized tooling by specialized individuals. When in reality, we’d all be better off if most of things were standardized and interoperable—then we could get down to business doing the actual unique things we are good at, not just managing technical debt.

As technologists we are really good at keeping things complicated so that we are needed and have something to sell. The state of API deployment after all of these years demonstrates that the API lifecyle is governed more by business and political decisions than they are ever by technical ones. Showing that us technologists aren’t as in control over what is going on as we think we are, and that the “right” technical approach isn’t the “right” answer to most API questions being asked.

This article originally appeared here.

Kin Lane
Kin Lane is a writer, storyteller, and recovering technologists. Kin is the Chief Evangelist at Postman, and is helping share the story of how Postman is the next generation of API development environment (ADE), while also continuing to tell API stories on API Evangelist about what is happening across the API sector.

APIdays | Events | News | Intelligence

Attend APIdays conferences

The Worlds leading API Conferences
with 9 Conferences in 2019:

Singapore, Zurich, Helsinki, Amsterdam, San Francisco, Sydney, Barcelona, London, Paris.

Get the API Landscape

The essential 450 companies

Get the API Landscape
Industry Reports

Download our free reports

The State Of Api Documentation: 2017 Edition
  • State of API Documentation
  • The State of Banking APIs
  • GraphQL: all your queries answered
  • APIE Serverless Architecture