What’s a Twitter client?

Thanks to yesterday’s Engadget report from Karissa Bell, we all know the official Twitter rule that Twiterrific and Tweetbot broke. It’s this bit of Section II.A of the Twitter Developer Agreement:

You will not or attempt to (and will not allow others to)

c) use or access the Licensed Materials to create or attempt to create a substitute or similar service or product to the Twitter Applications;

Of note is that this section of the agreement is entitled “Reverse Engineering and other Restrictions,” and Twitter reverse engineered it just yesterday to add this new restriction.

Michael Tsai had a good question today:

I don’t understand what this means for API users such as NetNewsWire that are not trying to create their own client. What counts as a substitute?

Because you can only read tweets through NetNewsWire that certainly shouldn’t count as a “substitute or similar service or product.” But who knows what’s going through the mind of Twitter nowadays?

On this week’s episode of Upgrade, Jason and Myke (mostly Myke) talk about how it would be a perfectly reasonable business decision for Twitter to shut down its API entirely. That goes too far, and I suspect they didn’t really mean it, because a lot of the API is valuable to Twitter’s most popular tweeters. Large, well-known organizations (like media companies) that tweet a constant stream of links to and excerpts from their own websites need the API to automate that process and keep it running smoothly. Twitter can’t kill the API without pissing off their most valuable content producers.

Which, I assume, is why the new section of the Developer Agreement is written the way it is. Rewriting the API and to make Tweetbot impossible while still giving CNN the freedom to spit out automatically generated tweets every few minutes would be really hard. Adding a provision that allows Twitter itself to decide what is and is not a “substitute or similar service or product” is really easy.