APIs can shut down, bummer. But it's more of a bummer to never use them in the first place.
It'd be rad if you could integrate with Genius' API in such a way that if they ever shut it down, or you want to switch to another annotation service for any reason, you would only have to point to another annotation provider instead of rewriting all the syntax/semantics of the integration.
I am very hesitant to build anything on a free API like this nowadays, after seeing what happened to the Twitter API, and more recently, the Soundcloud API. I don't feel comfortable building an app using their API, and then having them shut you down for whatever reason they come up with.
The problem I see with companies providing an API on top of their service is that their core business comes first, API second. So if someone were to use their API to build something that takes traffic from their core business, they will shut it down. Whereas a company whose sole business is providing an API won't run into that same problem.
And I can't blame the companies for making that decision either, why should the company serve your app if they're not even getting any traffic from it? Which would be a logical argument, except these companies open up their APIs and act like they are ok with it at first, and developers spend time building on top of them, creating even more useful services, making the original company even more popular, and then after those companies benefit from these developers, they shut them down.
If your business model is entirely dependent on margins with other companies' APIs (free or paid), a price increase for those APIs would put you out of business anyways. The concern shouldn't be "What if they charge more for this API?", the concern should be "Why is my business so dependent on this company's APIs?"
It seems to me like this gives the company free ideas and feedback they can then turn around and implement in their core business. Is this too simplistic, or is this simply not practiced enough?
I was super bummed when it happened because I use Soundcloud a lot and their app has some serious issues. I was considering building my own client, but then saw that these guys built a really nice alternative. Sadly it got shutdown shortly after.
Genius is really cool service (founders/owners aside) and I love it for song lyrics. An API into their backend sounds appealing but without some contract that keeps them from shutting down the API with little to no notice or pulling a twitter/soundcloud/netflix/etc I can't even begin to think about building on top of this (for anything but little fun side projects/POC's).
Which is exactly what they want. They don't want people scaling and running huge businesses off their API, they want lots of little experiments by people messing around. They can see what gains traction and then copy it. It's like a free startup incubator and your investment is the time your engineers spend keeping the API up.
Anyone that is not from Genius care to comment how useful this might be? I'm having a hard time thinking about why I should integrate with this service. All it seems to do is take notes. Am I missing something?
I actually think this has a ton of potential. Quick example would be documentation for some framework. Users can comment directly on confusing parts of some documentation, and after some discussion, it can be edited to fix whatever confusion there was. That's just one example, I think anywhere you have big blocks of text this will be handy. Even simple things like marking typos in an article would be much easier if every site used a single annotation service.
Good point. Now, the question is, are the annotations available for download? If not, why bother? Annotations are useful if they can be saved and moved (just like bookmarks).
Nice. Is that a future proof feature? Will I be able to download the data in any point in the future? After being punched in the mouth but other APIs I'm wary of integrating without some understanding of what the future might look like.
It'd be rad if you could integrate with Genius' API in such a way that if they ever shut it down, or you want to switch to another annotation service for any reason, you would only have to point to another annotation provider instead of rewriting all the syntax/semantics of the integration.
Sound cool? Get involved with:
* W3C Web Annotation WG: http://www.w3.org/annotation/
* OpenAnnotation Ontology: http://www.openannotation.org/spec/core/
* W3C Social WG and Vocabulary: http://www.w3.org/TR/2015/WD-activitystreams-vocabulary-2015...