How to build culture tech
You can implement stuff in hardware, in software, or in culture
Polls on twitter let you see the results only after you vote. This is a problem when users want to see the results and don’t have a strong opinion, they’ll vote randomly.
How would you fix this?
Twitter could implement a “Show me the results” button, that blocks you from voting after clicking it
If you don’t work for twitter, you can make a web app that lets users make a poll and they tweet a link to it
The other option is to implement it in culture:
This is a great example of culture tech because it’s very easy to bootstrap; it works even if you’re the only one who’s using it. Anyone can adopt it & it gives them an immediate benefit (they get better poll results). Other people adopting it also improves the product for you personally (you get the button you wanted).
This is a very big idea for me because I used to be stuck thinking that if I want to fix something I have to write software to do it. But now I see it more like, in the same way I can prototype or implement something in JS/Python before making a production ready version, I can do the same thing with “culture tech”.
Bonus: you can do this without writing a line of code. It’s the real “no-code” option.
90% of Twitch was developed this way?
I was pretty surprised to read this: “90%” of successful product development at Twitch came from community contributions. And apparently twitter too.
That’s almost as if Twitch had open sourced their platform, and their users could open PRs into it to improve it, and Twitch made a lot of money, and the users got a better platform (win win)! I believe this is the reason why open source software is often much higher quality than proprietary: it lets you take the best ideas from the widest possible funnel.
People keep forgetting this is an option
Samo Burja calls this “social technology”. I love the part where he says it can be documented & taught. As a technologist, I’d like to add: we can also build it.
What would happen if twitter had an option to upvote/downvote/emoji react to posts? I don’t have to wonder, I can just make a poll where the options are “👍/👎/🤔/🤮”. I found this to be super useful to get inline feedback into my writing (especially in cases where my followers are too nice to tell me what I’m saying doesn’t make any sense).
At some point twitter started blocking the preview card for substack links. This really bothered me, so I made a little proxy web app that copies the <meta> tags & hosts it on a different domain to work around this hard-coded block. Thousands of people have used this and still do. I consider this more of a cultural solution than a technical one, because it’s more teaching people how to use a proxy, and how the internet works. It’s very hard for twitter the company to fight this because it’s more of an idea, user behavior.
Building cultural infrastructure
Here’s another piece of cultural tech: what if we created an open API for twitter data? Just like with putting a (results) option in your polls, no one can stop you from building this. You can export your data and share as little or as much of it with the public as you want.
The Twitter Community Archive I’m working on is one implementation of this, but it’s being built on cultural infrastructure: users normalizing exporting & sharing their own data. We’re paving the cultural infrastructure at the same time as we’re building the tech pieces. This is why I was really happy to see that the Washington Post was doing basically the same thing. In their version they were asking TikTok users to export & submit their data. The user data was NOT public, but the analysis & insights are. I consider this part of the exact same cultural infrastructure that is paving the way for other things to be built on top.
The next thing I want to tackle is: “why don’t we have community notes on ALL ads across society?” On Twitter you often see an ad with a community note that says “this company is a dropshipping scam”. And now I ask, why hasn’t this spread everywhere?
When I bring this up to people they throw their hands up and say “because big companies suck and they will never do something that’s good for us”. And to that I say, “do you not realize you have agency? you can build cultural tech, literally no one can stop you (and also companies that create value will totally support this because they will make a lot more money, only the scammers will lose here)”. You can build an extension that shows you what your friends have said about this company, and bootstrap it with reddit data (where most people currently go to search for honest user reviews).
You can just start behaving differently, in a way that brings you immediate benefit. And if it spreads to your friends & beyond, all the better, it’s win-win. All software is open source, whether the authors like it or not.
Man, I'm in awe of your approach to improving things. It's literally just examples of you doing the things you say you want to do/see, and showing how/why they work. It's like "oh wow, things can actually just be this straightforwardly good."
You might enjoy this pieces about how things like Twitch work in terms of “ecosystems” - I’ve found it very helpful in building out product strategies: https://blog.gardeviance.org/2013/01/ecosystems.html?m=1
Note that Palantir’s FDE strategy is effectively an internalized ecosystem.