This is a guest post from Emma Barnes. Emma is co-founder of General Products, and indie publisher Snowbooks. General Products is the company behind FutureBook-award-winning Bibliocloud, the web-based all-in-one publishing management system.
API is one of those acronyms you hear bandied about. “APIs”, people say, “are vital. Vital!” But if, as you fervently nod in agreement, you’re thinking of another acronym — “WTF?” — then this article is for you.
“API” stands for “Application Program Interface”, which means “a way to interact with a program”. Actually, the sorts of APIs that people get most excited about nowadays are RESTful APIs. In English, that means “a way to send and receive data between programs on different computers, over the internet”.
If you can get computers to send data to each other, that can be very useful indeed. (It’s why the notion of the internet was so moreish.)
For example, you can get data from Amazon’s computers onto the computer that runs your website, which can display it. Even better, you can get your own computer to look at the data that Amazon’s computer sends you and check if it’s correct — so you don’t have to visit 100 Amazon pages to check with your human eyes. You can get the data of your Twitter feed from Twitter’s computers and display that on your own website — or, even better, you can ask Twitter’s computers to send you every tweet that contains a mention of the next 90 books you’re launching and display them on your website, without you having to look each of them up.
You can write your own API so other computers can access your data, too. You could turn the chapters and paragraphs of your forthcoming book about Edinburgh Castle into data that other people’s computers could get, and they can display it on their own websites, or in an app, alongside their own design, pictures and content.
You probably knew all that. But you might not know how it actually works. First off: what does an API look like?
Here’s the result of a query, or ‘call’, to the Google Books API. Here’s the call that was made:
Ooh, that’s familiar! It’s a website location! When a computer queries an API, it’s going to a normal old website location, handing over any username and password that’s required, and fetching whatever’s there, or posting new data to the program.
Click on that link so it opens in Chrome or Firefox and you’ll see that the result of an API call is… data. Structured data, though, arranged in a format that’s more convenient for computers than humans to understand (in this case, JSON). It’s still got English words in it, though, so we’re in with a fighting chance of understanding it. You can pick out “authors”, “title”, “ISBN_13” and so on in all that.
Now – this is the cool bit. You’re going to run your own API query.
change the text ‘isbn:9781909679368 to something of your choice. So: say you want to search for books about quilting. Change it to this:
Now click on Run, top left. And see what happens in the results pane! You have made a little scrolling AI all about quilting. Try something else: churches, or snow, or programming. Tinker with anything else on this fiddle and see if you can use other bits of data from the API.
If that’s whetted your programming appetite, you should register your interest for the super-enjoyable Try Programming course that we run in Oxford where you can find out exactly how this code works.
So well done: you’ve had a tinker with the Google Books API. There are loads of APIs out there – from Flickr, LibraryThing, Goodreads, Open Library, and Bibliocloud, of course. Here’s a list of 53 of them. The possibilities are endless and it doesn’t take long to learn enough code to be in control of the world’s data.
Exciting, isn’t it?