23 comments

[ 2.8 ms ] story [ 54.9 ms ] thread
"The long tail is still missing in action" - plug&play arriving onto the API scene is going to change that fast.
Yep, I guess that's a big part of what's missing still. Mass use of platforms to make it easy for anybody to do this.
What is plug&play? Is it a concept similar to Microsoft's driver system, or is it a startup, or is it a standard for API designers?
I'd say platforms like stackmob which create mobile APIs + wordpress, drupal etc. which make it easy to add an API to those + then platforms like 3scale, mashery etc. which help manage access.
We need more hard data like this on the API ecosystem. ProgrammbleWeb is a great source.
agreed there is not enough metrics on what is out there, where the opportunities are? I know John and the team @ PW is working hard to track on all of this, but its good to see more detail get pulled from someone external.

If anyone else wants to help slice and dice and make sense of it all, ProgrammableWeb has an API....

(comment deleted)
1000APIs is a small sample size.
Yes, agreed - but the total number of APIs listed is a little over 6000, so it's a reasonable sample to some extent. There are a lot more APIs out there which aren't public / listed - so it's hard to account for those.
That depends on the signal-to-noise ratio you're expecting.
Is great to see a lot of companies releasing APIs but if even better as a developer start a project and being able to easily use external services and even integrate your app with bigger services. Imagine creating an e-commerce an integrating your catalog with stuff from macys, amazon, etc.
I think this will end up happening - Amazon already lets you suck out it's catalogue and push affiliate sales back to it. You could likely build very sophisticated niche virtual retail sites if all the bricks and motar data was there also.

Interestingly in the survey the number of ecommerce APIs was actually a lot lower than we thought it would be. This might be a blip in this "generation" of APIs, or that it's simply hard to do still.

Amazon has a great vision, others are just afraid about people "stealing" their content. Probably companies need to be educated in this area
I guess the model needs to be right so everybody along the way gets paid, then people may step up.
Glad to see the takeup of APIs by the public sector. Will be interesting to watch whether emerging API management solutions can help the open data movement by enabling smarter ways of accessing government data.
The US definitely seems to be leading here - makes more sense to open data than to try to provide more and more different UIs that people need (iPhone, Android, ...). Of the APIs on programmable web in that period almost all were US, a couple were UK and a couple we Canadian.

Further back in time there are other countries also though - and a lot of the Scientific APIs were European.

I think that Twitter has absorbed some of the long tail - it's pretty easy to use a twitter account as a proxy for read and write commands for a low scale device or web site, of course you can't make it a chargeable interaction!
Pretty interesting way to think of twitter - kind of a message bus.
Its an information network for humans and otherwise. Its simplicity + #hashtags makes work well for distributed messaging.
Very interesting analysis and segmentation of APIs! I am just wondering which factor has been so the most prominent for not having the long tail represented (or badly)?