When I first read the announcement about the API changes and recommendations I was shocked and angry. Like many developers, I had yelled when they stopped whitelisting applications, and I thought this new announcement was going too far. Thankfully, my family obligations did not allow me to write this post immediately following the announcement. So, you get the benefit of further review of the announcement and a much more level-headed analysis. First, the money quote in the announcement:

developers ask us if they should build client apps that mimic or reproduce the mainstream Twitter consumer client experience.  The answer is no.

Obviously, there are a lot of people talking about this, but I think this discussion ended a long time ago. A basic Twitter client is a terrible idea in today’s ecosystem. Unless there is major functionality outside of the existing solutions, a new client is a losing idea. There is a high barrier to entry when we already have third-party clients like Tweetdeck, Seesmic, HootSuite and PeopleBrowser. This does not include some of the other applications that focus on team or brand management. So, by saying not to develop a new client, Twitter has saved us and investors a lot of time and money.

The other side of this announcement is their promotion of other vertical solutions. They have asked developers to focus on applications for curation, real-time data signals, social CRM and other vertical and value-add solutions. This is where the real money can be found as well. When you have a third-party client, it would be very difficult to charge people a fee. HootSuite can do it because they have a ton of features in addition to being a Twitter client. You can charge for applications that provide some vertical solution fairly easily.

The Twitter API usage terms are also fairly open. The main thing you cannot do is resell distribution of the Twitter API data, unless you get permission from them. Obviously, there are the potential concerns of hitting API limits, but that is a problem that can be worked around. There are a lot of applications that can still be built on top of the Twitter API. I would still caution building an application that depends solely on Twitter and ensure that you include other social platforms like Facebook and whatever other services make sense.

The main benefit of Twitter finally laying down the law is that we have a solid sense of direction for the developer ecosystem. Twitter has now stated that they want to own the client user experience because it wants to provide consistency. This is something that Twitter should have done a long time ago, but it does raise some important issues. The idea of a federated Twitter is dead. Without development of a new client, there is no way to federate the functionality of Twitter. Identica has developed a similar API, and has promoted the idea of federation before. So there could be a federated microblogging tool in the future, but it definitely will not be Twitter itself. They will also need to improve support of the API, as these announcements will likely have more focus on the currently less popular API features. API support has been good so far, but they may need to increase the resources around it.

Twitter also now owns the platform as a whole and must be as reliable as a utility company. They must provide all of the capabilities that consumers need in the clients. If you look at the Twitter.com usability, it still needs some help. Twitter created lists, but they are awkward to use on the main site. I still find direct messaging to be an almost useless feature, as there is much that can be improved. The idea of the stream of information also has usability problems when the stream moves very fast due to your number of followers. So, if Twitter wants to own the user experience, then they need to make sure they make all of their features as easy to use as possible.

This is what Twitter has asked for. I just hope they are prepared.