twitter threads and chesterton's fence
This post is a simple example of how it can be valuable to examine situations with the heuristic of Chesterton’s Fence. The author observes that people use Twitter threads and that he does not know a good reason to use Twitter threads, so he concludes people should not use them. Below, I submit to you that there is a purpose to Twitter threads and that they are Good, Actually. First some quotes:
…tter land seem to be jumping on the bandwagon of a very annoying craze. Twitter threads . Please stop; write a blog post…
A “craze” continuously occurring since before the feature’s formal introduction in… 2017.
…othing more than a good thread. I assume it’s because these people seem to get a tonne of followers off the back of these threads, thus perpetuating the whole thing and forcing more people onto that bandwagon. What is a Twitter thread? That’…
This assumption is both uncharitable and not founded in any data I know of.
…ils down to one thing. Context. Using the example above, I became aware of it because someone that I follow re-tweeted Tweet number 47 in this ridiculously long thread. Because it was a random Tweet in a very long thread, I (and I imagine most other people reading the tweet) had absolutely no context as to what the whole saga was about. It isn’t the right platform Sec…
This is a particularly bad example because Cory also has the exact same thing on his blog. He literally links to it at the top of the Tweet thread.
…e new audience for the creator. I’ve some really interesting threads on Twitter that are full of useful information, but as a consumer of that content, it’s a nightmare to follow. Content creators should make content as simple to consume as possible.
Real creative humans don’t tend to “create” “content”. They write essays, sing, do sketch comedy, dance, take photographs, etc. This is a long-running peeve of mine but I’m bringing it up because the corporatized-tech language obscures the nature of this complaint.
Imagine: “I’ve read some really good plays, but it’s not always obvious what’s going on. Playwrights should write out prose between the pieces of dialogue so it’s as easy for readers to understand as possible.”
Anyone can see this would be silly, since the form–drama rather than prose–is part of the intended experience.
If you are consuming Twitter threads equivalently to long-form content, you are missing out on part of their value, since each tweet is open to its own replies. This means that people can chime in with their own stories or anecdotes or objections non-linearly.
Cory’s thread doesn’t enable this because it’s so long it breaks the Twitter UI’s conventions of loading first the thread, then people’s replies. I would probably agree with the author that makes Cory’s Twitter thread relatively ineffective.
However, it can still be the case that someone can excerpt (via RT) a particularly interesting detail they want to highlight without judging an entire thread as worth the time – and not having context is easy to fix if your mouse wheel is working.
So we get back to intent: some writers would hate having a comment section for every paragraph in a blog post. Other writers like to make that form of discourse part of what they’re doing1. So someone writing a Twitter thread isn’t just blogging wrong, they’re participating in a different form. If you don’t engage with what makes that form different, then yeah, of course it looks preferable to just copy-paste the text into paragraphs to read2. But… you can’t therefore conclude that other people are “forcing” each other onto a “bandwagon” because they “get a tonne of followers.”
I even think I’m trying to make blogging more like Twitter threads3 by using hypothes.is to annotate posts like this, to then pull over to my site as a post with quotes. It’s really fun to be able to use the internet able to save or respond to arbitrary chunks of text just as easily as you could a tweet4, sort of reading with a pencil in a hacked-together duct-tape Xanadu style.
It’s pretty common for a thread’s author to further engage with replies to parts of the thread, so the content extends beyond the main thread to a sort of tree. ↩
I’ll digress for a reprehensibly techie comparison: If you don’t use stuff like transclusion and non-linear structure, outliners seem like they introduce noise into what could just be paragraphs of text. But… the whole point is the part you’re not using. ↩
Without properly branching replies, it’s arguable that this is more like inline replies to email… but, you know, I’ve gotten a couple replies to my annotations, so I’m still gonna say it counts. ↩
The Indieweb stuff I have can take the place of the notification of a reply… Well, for responses to things where the author supports it. ↩