There is a significant risk and reward that comes with developing products that leverage third-party application programming interfaces, or APIs. Twitter has used its API to let others spread the word for them; applications like Tweetie and TweetDeck help Twitter reach a broader audience on a variety of devices while making money for themselves. However, downtime for a service offering their API to developers means downtime for every service that relies on it for its API data. In the case of Facebook application developers, continuing reliability issues with the platform have become a cause for concern.
Nick O’Neill of All Facebook reported Tuesday that the API for the 400 million plus member social network had been suffering from some performance issues that day and that the issues were part of a larger trend of poor reliability. “If you are going to build an application on Facebook, you’d expect that users will be able to access your application as long as it’s developed properly, but unfortunately Facebook’s reliability has not been at its peak recently,” writes O’Neill.
On Monday afternoon, Facebook reported that they were “investigating issues that may be causing some Platform apps to fail to load,” and Tuesday said they were “seeing some site performance issues which are affecting [their] API.” According to O’Neill, the reliability of the platform has “been at an all-time low for over a month,” which is not a comforting fact for app developers.
“Clients pay us for their applications and are well within their rights to complain when things suddenly disappear,” writes Facebook application developer Tony Faria in a comment on O’Neill’s post. “Having to tell them ‘Facebook devs are working on a fix’ loses the confidence clients have in us, yet we work so hard to build.”
It is interesting that the latest issues with Facebook were identified at 2:50 pm on Monday, and 2:40 pm on Tuesday. If we assume that these times are Pacific Standard Time, that would mean that the last two delays to Facebook’s API came just before 6pm on the east coast. Could it be that the site suffers a significant uptake in usage around this time as people get home from their 9-to-5 jobs? This could be a strange coincidence, or it could be evidence of the platform’s issues with scalability, either way, developers seem worried.
The danger included with using third-party APIs is not a new topic of discussion here at ReadWriteWeb; in December we brought you the story of Totlol which was forced to scrap its business plan when Google suddenly changed the terms of service for YouTube. While the API providers are ultimately responsible for uptime and content, the developers are tasked with dealing with customers when their service doesn’t work, so be cautious when building your company on the foundation of a third-party API.