We screwed up from a communications perspective this week. When I heard that this change was going out, I rushed to write a blog post. This setting had both product design flaws as well as technical flaws and I did not do my homework. My post came from a product design perspective but the technical perspective was the reason it went away so quickly. Normally, I spend more time understanding the issue before explaining it on this blog but in my haste I made a mistake.
Subsequently, there is now a lot of confusion about what this change actually was. 97% of all accounts were not affected at all by this change—the default setting is that you only see replies by people you follow to people you follow. For the 3% who wanted to see replies to people they don't follow, we cannot turn this setting back on in its original form for technical reasons and we won't rebuild it exactly the same for product design reasons. I've taken some time to break down both the product design and technical problems this feature had.
Product Design Flaws
Since last year we've been hearing from users and having discussions about removing this setting—feedback indicated that it was useful but also created confusion. People would change the setting and then not understand why their timeline had fragments of conversations. From the tweet author perspective, there was an unclear expectation as to who would actually see messages which often lead to trepidation when it came to using replies. Finally, even folks who understood the setting would complain that they couldn't follow accounts with a high volume of replies because the replies overwhelmed their timeline. It was becoming apparent that we had an opportunity for improvement.
Technical Problems
Even though only 3% of all Twitter accounts ever changed this setting away from the default, it was causing a strain and impacting other parts of the system. Every time someone wrote a reply Twitter had to check and see what each of their followers' reply setting was and then manifest that tweet accordingly in their timeline—this was the most expensive work the database was doing and it was causing other features to degrade which lead to SMS delays, inconsistencies in following, fluctuations in direct message counts, and more. Ideally, we would redesign and rebuild this feature but there was no time, hence the sudden deploy.
Hopefully, this clears things up a bit and I promise to be less distracted and hasty when it comes to blog posts. We're still listening, reading tweets, gathering feedback, and working on more efficient ways of doing so. Thanks to everyone for the feedback and for using Twitter.
http://buzzup.com/2g3r
No comments:
Post a Comment