After Twitter experienced a third outage in less than two weeks – this time due to the “failed enhancement of a new approach to timeline caching” we are left asking “is it a case of too much too quickly?”
No public service of this magnitude should be expected to be without issues but the recent multiple outages are starting to become a cause for concern amongst social media types. Twitter has a history of having infrastructure issues and the current problems are becoming reminiscent of the torrid time experienced last summer. It currently seems that every time Twitter implements a new feature something else breaks.
Last year, it took Twitter a long time to re-establish trust in the service after a summer of the Fail Whale. At the time many users were considering leaving and never returning – the feeling that the service could no longer be relied upon was that great with many feeling that they could no longer justify investing their time in something that would just let them down.
If Twitter wants to attract advertisers and become a viable platform for both promoted tweets and for selling spots in trending topics then they need to retain trust and prove that they are stable and that we are not heading back to another summer like last year. No-one wants to risk investing in an advertising platform that will not give a decent ROI due to service down time. A failure to get it right first time will almost certainly kill Twitter’s chances of being given a second bite at the cherry.
Twitter needs to be completely open and honest about the issues they have been experiencing. This is one area where they do, at least, have the edge over some other services. They are normally pretty quick with an admission that they screwed up – honesty goes a long way but they also need to show that they are learning from their mistakes.
They also need to publicly demonstrate the steps they are taking to increase both capacity and resilience, not just to meet current needs but to also future proof the service considering the current rate of growth and the potential for further impacts due to new functionality.
Finally, Twitter has to demonstrate effective, intelligent decision making; there has been a issue in the past with Twitter’s ability to handle the load from large events – well, they don’t come much bigger than the World Cup. Is it wise to roll out new features during such a time when you have already been exhibiting problems coping?
Too big to fail?
Twitter is being quite aggressive with updates and feature rollouts of late, perhaps a more softly, softly approach might be in order for a while. Implement one thing at a time and let it bed down for a while before thinking about the next thing.
As the other contenders and pretenders fell by the wayside (identi.ca, Plurk etc.) Twitter became the de facto standard for micro-blogging and short status updates. Celebrities reveal their inner most (ofttimes vacuous) thoughts, election campaigns have been staged and fought, and global corporates have been brought to task. Just like the banks Twitter has become too big to fail and, as a company, needs to do everything it can to prevent that from happening.