When do you know you can trust an API to do what it said on it’s box? Developers are bombarded with countless new releases of “really cool” APIs, promising easy integration and inspiring the creation of ground breaking applications. What they don’t say is that this is just a teaser, and in fact, they’re about to change everything and turn your project on its head.
Obviously it’s up to the creator of a specific API to decide how it will be made available to third parties and at what cost, and that’s fine; but make that clear up-front. It’s just unproductive for anyone making use of these to have them constantly change.
Here are some of the issues that may arise:
The API provider decides to change the service from free to paid -
This leaves the developer held at ransom; pay or your application dies. The issue is that if developers are not generating income by using this API, how can they justify or even afford paying for it?
The API provider changes the method for calling (using) the API -
The problem here is that web applications calling this API will simply stop working, and often implementing the new version can force developers to have to re-code large portions of their projects, sometimes even using different programming languages.
For example, this wolfish behaviour has been evident in Google and Bing APIs. Both have changed their service from free to paid, and both have eliminated certain methods for calling the API.
Again, this is all fair enough, but it would be better for all involved if these intentions were made clear from the start.
(Cartoon by Rob Cottingham, Social Signal)