How APIs should be: Drop in keys, running in 1 minute(blog.singly.com)
blog.singly.com
How APIs should be: Drop in keys, running in 1 minute
http://blog.singly.com/2012/05/31/how-apis-should-be-drop-in-keys-and-be-running-in-1-minute/
6 comments
At Pwinty- we collect emails (conversion rate 5-10%), and then try and guide users through the process.
Firstly- its a great way to engage with users where you might not offer exactly what they want (do you print canvases, how about xyz). We get most of our feature ideas from our users wanting something that we don't yet offer.
I also try and schedule in a Skype call with all of our users to understand exactly what they want from us, and how we can help them (I think this also convinces them that there's a real person there they can do business with).
Once we've got their emails, we probably convert about 20% into 'making API calls'. However, I'm expecting that to increase over time (lots of people sign up, but are just working out how they're going to build their app, rather than actually doing it).
It's also worth following up non-active users after 1,3 and 6 months. You can often get useful info that you'd otherwise miss 'we decided not to use you because x,y,z; we canned the project but are keen to use you in the future etc)
Firstly- its a great way to engage with users where you might not offer exactly what they want (do you print canvases, how about xyz). We get most of our feature ideas from our users wanting something that we don't yet offer.
I also try and schedule in a Skype call with all of our users to understand exactly what they want from us, and how we can help them (I think this also convinces them that there's a real person there they can do business with).
Once we've got their emails, we probably convert about 20% into 'making API calls'. However, I'm expecting that to increase over time (lots of people sign up, but are just working out how they're going to build their app, rather than actually doing it).
It's also worth following up non-active users after 1,3 and 6 months. You can often get useful info that you'd otherwise miss 'we decided not to use you because x,y,z; we canned the project but are keen to use you in the future etc)
I very much regret that I have only one upvote to give for this comment, because every API company should be doing that. (Most of the very successful ones do already, but since many API companies start with two hackers and a couple of boxes of ramen, you can greatly increase your chance of success by taking this comment and executing on it.)
> The bottom line is that the single most important step in converting developers to permanently use your API in their product is how easy it is for them to get started after they sign up.
I take issue with this. If the single most important step to converting developers to /permanently/ using your API isn't something like scalability, security, performance, or functionality, there is something wrong with this world.
Just because you get more people to succeed in trying your API at the beginning, if it doesn't work for their use case over the long haul they are just going to leave and do something else: the key word is then "permanently".
In comparison, Amazon's APIs are often ludicrously complex. Even if you try to use Amazon's client libraries, they are often ludicrously complex, requiring you to construct object factories to get arguments even for the most common request paths.
But, by and large, Amazon's services don't stop working when you get too much data, have well-thought-out authorization mechanisms, have never become randomly slow as lots of users sign up for the service, and are sufficiently flexible to support many different kinds of tasks.
The fact that the first time I sat down and used Amazon's APIs it took me a while to get them working (and that some APIs from them I still haven't gotten around to learning all the properties of) is then irrelevant: I could tell they would do what I wanted once I figured it out, and I have never been disappointed.
Meanwhile, I've come across tons of random startups, often that have websites that are a "little too Web 2.0" (normally a negative indication for an API provider, btw), whom have APIs (or libraries, or daemons, or services, or programming languages) that are great at first but actually sucked.
(I mostly felt the need to make this comment, btw, due to someone else in this thread seemingly arguing against OAuth because, despite being secure, it made the API sufficiently complex that they could no longer use curl with it: that is the opposite method from how you should be choosing an API.)
I take issue with this. If the single most important step to converting developers to /permanently/ using your API isn't something like scalability, security, performance, or functionality, there is something wrong with this world.
Just because you get more people to succeed in trying your API at the beginning, if it doesn't work for their use case over the long haul they are just going to leave and do something else: the key word is then "permanently".
In comparison, Amazon's APIs are often ludicrously complex. Even if you try to use Amazon's client libraries, they are often ludicrously complex, requiring you to construct object factories to get arguments even for the most common request paths.
But, by and large, Amazon's services don't stop working when you get too much data, have well-thought-out authorization mechanisms, have never become randomly slow as lots of users sign up for the service, and are sufficiently flexible to support many different kinds of tasks.
The fact that the first time I sat down and used Amazon's APIs it took me a while to get them working (and that some APIs from them I still haven't gotten around to learning all the properties of) is then irrelevant: I could tell they would do what I wanted once I figured it out, and I have never been disappointed.
Meanwhile, I've come across tons of random startups, often that have websites that are a "little too Web 2.0" (normally a negative indication for an API provider, btw), whom have APIs (or libraries, or daemons, or services, or programming languages) that are great at first but actually sucked.
(I mostly felt the need to make this comment, btw, due to someone else in this thread seemingly arguing against OAuth because, despite being secure, it made the API sufficiently complex that they could no longer use curl with it: that is the opposite method from how you should be choosing an API.)
Good points are covered in this blog post, however what if we would push it even further?
Why would people need to download a package? Why would anybody need to learn a programming language you package designed for (or assume you distribute packages for many programming language)?
When using APIs for integration or mashups, especially in business world, person who know how solution should looks like may not have time to code. So we at http://elastic.io think API could be via visual designer.
We believe it's also a way to make API usage _simpler_ for enduser, even the one who can not, or have no time to code.
Why would people need to download a package? Why would anybody need to learn a programming language you package designed for (or assume you distribute packages for many programming language)?
When using APIs for integration or mashups, especially in business world, person who know how solution should looks like may not have time to code. So we at http://elastic.io think API could be via visual designer.
We believe it's also a way to make API usage _simpler_ for enduser, even the one who can not, or have no time to code.
We implemented something similar for The OpenPhoto Project where every URL a user visits is a visual representation of an API endpoint.
My site, for example, is at http://jmathai.openphoto.me/photos/list. You can add ".json" to the end to see the JSON representation of that page. http://jmathai.openphoto.me/photos/list.json. The same rule applies across the site.
We wrote about it a while back.
http://blog.theopenphotoproject.org/post/8462620991/clean-ur...
My site, for example, is at http://jmathai.openphoto.me/photos/list. You can add ".json" to the end to see the JSON representation of that page. http://jmathai.openphoto.me/photos/list.json. The same rule applies across the site.
We wrote about it a while back.
http://blog.theopenphotoproject.org/post/8462620991/clean-ur...
Good call. Maybe it was misworded in the post, but the packages available for download are fully functional app skeletons that consume the API and are ready for you to build on. The API itself is just JSON over HTTP.
Making an API accessible without code is an admirable goal. Reminds me of Yahoo Pipes and the application-via-visual-blocks frameworks.
Making an API accessible without code is an admirable goal. Reminds me of Yahoo Pipes and the application-via-visual-blocks frameworks.
Exactly! We are Yahoo pipes on steroids.
I now know how I like Web APIs to work: Give me the URL and let me go at it with cURL and some scripts to parse the your JSON.
[Been using the Salesforce REST API recently, and apart from the process of getting your OAuth token, it's really rather good].
[Been using the Salesforce REST API recently, and apart from the process of getting your OAuth token, it's really rather good].
Yes, I was rather pleasantly surprised with the Pivotaltracker API. I managed to get it integrated in our hour registration within 30 minutes.
Check; https://www.pivotaltracker.com/help/api
All calls have ready cURL examples you can copy/paste and try. Prototyping is fast.
And no OAuth. OAuth could be good if it were standardized (and thus would work with a single library without any extra tinkering); it's horrible all over the place. I understand this fanatical obsession with security and such, but just make me login and get a token back.
Anyway; Hooray for Pivotal! (Again)
Check; https://www.pivotaltracker.com/help/api
All calls have ready cURL examples you can copy/paste and try. Prototyping is fast.
And no OAuth. OAuth could be good if it were standardized (and thus would work with a single library without any extra tinkering); it's horrible all over the place. I understand this fanatical obsession with security and such, but just make me login and get a token back.
Anyway; Hooray for Pivotal! (Again)
FYI this is the same org which is running the $10K app challenge using their API this weekend; https://singly.com/appchallenge
Re-posted to the Factlink blog at http://blog.factlink.com/
(I am unsure of whether I like the make-a-play-in-the-browser-utility suggestion from the perspective of the business. I have never work directly on that question for a client, so I don't have even anecdata for y'all on this, but the developer in me says "That's a great idea! That way I don't have to do any work to kick the tires." and the marketer in me says "If we sold 'tire kicking' that would be wonderful now wouldn't it")