I introduced the Tweet Marker API in 2011. In the 8 years since, I’ve been lucky to have the support of some of my favorite apps like Twitterrific and Tweetbot, plus popular Android apps, with some developers paying a monthly subscription to help keep the service running. But while Tweet Marker hasn’t changed recently, the Twitter world has changed.
Twitter previewed the next version of their API this week. I don’t see anything so far to contradict what I wrote last year.
I believe strongly that URLs shouldn’t change, and that public APIs shouldn’t just disappear. Owning your content and having your own domain name are important parts of Micro.blog too. So I was committed to running Tweet Marker indefinitely, regardless of what Twitter might do.
But two other factors were nagging at me:
- GDPR made me take a fresh look at some of the assumptions in the Tweet Marker API. Changes are overdue. They aren’t simple and would require coordination with client developers.
- Twitter’s third-party ecosystem has changed a lot since Tweet Marker was created. Twitter apps have a very uncertain future. Timeline syncing is often handled without Tweet Marker, such as via iCloud.
Tweet Marker’s time has come and gone. I’ve notified developers using the Tweet Marker API that I plan to wind down the service by July 1st. I’ve also cancelled all the paid subscriptions. (Actually I haven’t billed anyone in over 6 months, so I’ve been running the service for free.)
Daniel Jalkut and I talked about this back in December on Core Intuition 353. I plan to work on Micro.blog for the next 20 years or more, so it needs all of my attention.
Over a million people have used Tweet Marker. I’m proud of that. If you’re one of those people, especially if you supported Tweet Marker as a developer or with a subscription, thank you.
Conversation on Micro.blog