Next Steps for the Touchatag Developer Program
Greetings,
My name is Ted Haeger, and I just joined the touchatag team this month as the new developer relations manager. I've spent the past couple weeks getting to know the touchatag team and assessing how we can better serve software developers. Now I'd like to share with you some of our plans and hear your feedback on them. If you'd like to know more about who I am, check out the About page on my blog.
Developers are an important part of our longterm goals for touchatag. One of our key focus areas for 2009 is to provide more developer interfaces into the touchatag service. I'll be working directly with touchatag management and engineering to represent the needs of 3rd party developers. So, starting with this blog post, touchatag will serve developers through our new program, the Touchatag Developer Network.
Our goal for Touchatag Developer Network is simple. Rather than trying to deliver every possible use of the touchatag service on our own, we want to provide developers with economic opportunities and programming interfaces into our service so that 3rd party innovation can run far and wide. We believe that a well run developer program fosters mutual success and prosperity between its sponsoring company and the 3rd party developers who integrate with the company’s services.
But enough with the lofty words. Let's get down to some of the changes that we hope to realize. Over the next several weeks, the major tasks that we want to accomplish are:
- Create touchatag's first "Developer's Guide"
- Release a powerful new API for the touchatag service
- Build the Touchatag Developer Network infrastructure
That's the quick-hit list. Read on if you'd like a little more detail on any of these items.
Touchatag's First Developer's Guide
When I first started investigating how developers could start integrating the touchatag service, it was hard for me to know what interfaces were available to me, and what usage scenarios were possible. Today there is only the technical documentation of our current SOAP API and block system. The developer's guide will explain what you as a developer can do with touchatag, and how to get it done. And, as my wise colleague from engineering Bruno advises, it's just as important to include information on what cannot be done. So I'll be drafting the first edition of that. (And Bruno has unwittingly signed up to be my technical editor. Thanks, Bruno!)
A Powerful New API
After reading through a lot of feedback from developers--in the current and previous Google Groups forums, from surveys the team has conducted, and conversation the team and I have had--the list of interfaces that developers would like us to provide is extensive. However, the one that stands out above all others for me is enabling the touchatag Application Correlation Server to make outbound API calls. The use cases for this are many.
With an outbound API, you could make a logging application for recording whenever a tag is read at a reader. One quick practical application of would be to make an application for tracking whether mobile personnel--maintenence, medical, or security--are making their rounds. For another, parts or packages could be tracked more easily around a production facility.
The current plan for implementation is to make it a
RESTful API. That is, it would be HTTP-based, providing CRUD-like operations via HTTP verbs such as GET, PUT, POST, and DELETE. The engineers here at touchatag are currently designing this interface, so now is the time to weigh in with your comments, questions and suggestions. (Reply to this post.)
Build the Touchatag Developer Network infrastructure
To make it as easy as possible for developers to succeed with our service, we also need to consolidate the various locations where developers can find information about developing with touchatag. We also need to facilitate better bi-directional communication so that developers can interact with the touchatag team and each other.
To that end, I'm working at bringing all developer content into the main touchatag website. That includes the upcoming Developer's Guide, the API documentation, the developer blog, and any how-to documents we might create. That will address the outbound information from touchatag to you.
I'm also considering options for two-way communication. We currently use the
touchatag Google Group as a forum. Our website is built on Drupal, and we current use it for our end-user forums, so it might seem obvious to spin up a developer forum there. It sticks to my "One Search to rule them all" principle. But I don't want to make that decision just yet. I want to strike up a discussion about it. So watch for my next blog post.
That's it from me at the moment. Now it's your turn. Leave a comment and let me know what you think about any of the conversation above.
Ted Haeger
Manager, Touchatag Developer Network
2
API features
Personally, I'd like to see a couple of different APIs.
(1) A more thorough, generic application for building 'applications'. Something that can connect to, interact with, and optionally return data from, any site or service.
(2) An API for provisioning applications/tag associations. For any mid-to-large scale tag deployment the current manual process is just too cumbersome. Take the Last.fm app -for example. Who wants to do this for 10 or 15 artists? Try 400. (The number of CDs in my collection). No way I'm going to enter 400 tags via the current interface. But if there was an API, so I (or you) could write a desktop app, to associate the tags quickly, I'd do it.
And while you're defining standards,
(3) A standard for clients, so that the community can build touchatag clients for non-supported desktops.
My 2¢ ...
RE: API Features
bear454:
Well, 1 & 2 pretty much sums up my current priority list. :)
As I see it, as soon as we deliver "Server-side REST-out" (and make that something that can also respond back to the client), developers will demand to have a better tag management mechanism than the current dashboard provides. (And personally, I'd like this to be implemented as some kind of API so that you can register tags right from your own apps. But I get ahead of myself.)
Item 3 on your list is interesting...I'd like to see better Linux support, too. I'll be talking to our engineers more about it.
--Ted
Ted Haeger
Touchatag Developer Network
ted@touchatag.com