So Twitter did the thing this week where they remind everybody who actually owns the platform, and I've been stewing about it since Thursday.
If you missed it: on August 16th, Twitter's Michael Sippey posted "Changes coming in Twitter API v1.1" on the developer blog, and it is exactly as dry and bureaucratic-sounding as that title implies, which is sort of the point. Buried in the corporate-speak is the actual news, which is that Twitter is finally closing the door on the wild-west era of third-party clients. Tighter rate limits. Mandatory OAuth on every single call, no more anonymous reads. And the part that's got every developer I follow either furious or quietly updating their resumes: a hard cap on user tokens for "traditional" Twitter clients, the ones that just show you a timeline. If your app already has more than 100,000 users authenticated, you're capped where you are. Under that, you get to grow to double your current number and then you're capped too. That's it. Game over, thanks for playing.
I run Tweetbot on my phone. I've run it since it came out and before that I ran Twitterrific back when Craig Hockenberry's team basically invented the whole concept of a native Twitter app. Both of those apps are exactly the kind of thing this rule seems built to strangle slowly instead of killing outright, which honestly feels worse. A ban you can be angry about. A slow choke where the app just can't add new users past a certain point, so it stagnates while the "real" experience gets funneled through Twitter's own apps and Twitter.com, where they can show you ads and cards and whatever else - that's a different kind of annoying. It's the kind of move that doesn't look hostile in a press release but is absolutely hostile in practice.
And look, I get the business case. Twitter's a public-facing ad company now in every way that matters, they took the money, they need a growth story, and third-party clients that show a bare timeline with no promoted tweets are a direct hit to that. I'm not naive about it. But I started using Twitter back in 2009 specifically because it was an open thing, a firehose you could build whatever you wanted on top of. Half the reason it got interesting in the first place was because a bunch of independent developers kept inventing stuff Twitter itself never shipped - pull to refresh, actually - Loren Brichter built that for Tweetie, which Twitter then bought, and now here we are watching Twitter tighten the leash on the exact kind of experimentation that made the product worth buying in the first place. There's an irony in there that I don't think anyone at Twitter HQ is in the mood to appreciate right now.
Unrelated small complaint while I'm in a mood: I also can't stand the official Twitter app on iOS. The Discover tab is a mess of stuff I never asked for, and "Trends" for my area is 90% about a local news story I have zero interest in. So it's not like I'm defending the official app as some superior product they're steering everyone toward. It's the opposite - the client they clearly want to win here is arguably worse than the ones being squeezed out.
None of this kills Tweetbot tomorrow. The token math means smaller, newer clients get hurt fastest, and the established ones have some room to breathe for a while yet. But you can see where the road ends if you squint even a little, and it's not a road where independent Twitter clients exist in five years. I've already started paying more attention to App.net, the paid, ad-free, developer-friendly network a bunch of ex-Twitter and indie developer types have been building this summer as a sort of "what if Twitter but we don't do this" experiment. I don't know if it survives. Most alternative networks don't. But I bought in anyway, mostly out of spite, which is not a great reason to spend money but here we are.
I'll keep using Tweetbot until it stops working. After that I genuinely don't know what I do.