Live data from Hacker News

Ask HN: How would you design an alternative Twitter

news.ycombinator.com

31–40 of 351 posts

Re: Ask HN: How would you design an alternative Twitter

#31
What is your goal is the first question you need to ask yourself. If your goal is to replace Twitter at its scale, then it's completely absurd, Twitter will continue to be popular for a good decade, no matter what Musk does because network effect.

You shouldn't try to design an alternative to Twitter, you should design a social network that catters to a crowd that is not Twitter or don't care about Twitter.

Tiktok, no matter how much I despise this social network, didn't try to copy Twitter, Facebook or Youtube, it did its own thing, and with the help of massive VC investment and marketing campaign, it became popular, because it was different.

Yet another Twitter knock off isn't going to replace Twitter.

Re: Ask HN: How would you design an alternative Twitter

#34
If you want to design an alternative Twitter then my advice would be to not start with the tech. There seems to be plenty of stuff out there than you can bolt together to get things started, and which comes with options for scaling if you subsequently need it.

Some thoughts:

- Start with thinking about how you're going to maintain engagement without pushing people into algorithmically-generated echo chambers, or into attack/defence behaviour that will just turn your New Twitter into the same cesspit as Old Twitter.

- Also think about how you're going to moderate content - because you're going to have to do it whether you like it or not, and its going to have to be at various scales from post-level to policy-level. There's a very good argument to be made that Legacy Twitter's product is content moderation [1], not software. Human nature being what it is, you may find that eventually it's your product too.

- Also, consider how you are going to deal with state-level and semi-state-level actors who will attempt to infiltrate your platform to use it as an amplifier.

The tech platform will help with the above, of course, but above all I think you need to consciously design the thing that you're building. That thing isn't a software platform or a social network or a community. It's something else. Figure that out first.

And if you want to build something that makes the world better (as opposed to worse, as in Old Twitter), you might consider how people can use your new platform to get stuff done in the world, rather than just shouting and meme-ing. I wish I knew how to do that though.

If you do build something like Twitter then please try to build it so that it doesn't cause harm.

[1] https://www.theverge.com/2022/10/28/23428132/elon-musk-twitt...

Re: Ask HN: How would you design an alternative Twitter

#35
post #5

A thing I truly liked about Google+ was the ability to separate contacts into "circles" to differentiate between real-life friends, family, online friends, colleagues, etc. I wish more social networks used that kind of approach.

I didn't get to use Google+ in its prime as I was averse to all social media at the time, but I totally agree. That was a great concept

Re: Ask HN: How would you design an alternative Twitter

#38
What I've always wished existed for social media was something more along the lines of a protocol (like email or rss) than a platform (like Facebook or Twitter).

It seems like as long as a communication form is tied in to one specific provider by design, then there is always going to be unsolvable problems with deplatforming users, content moderation etc.

That's what I'd love to see- decentralization in the sense of anyone can feasibly host a server, rather than the web 3.0 / cryptosphere sense. I genuinely think that would solve most of the issues we see from monopoly platforms.

Re: Ask HN: How would you design an alternative Twitter

#39
Spot the engineer looking to build something without doing any real empathy research first. This is why the Metaverse is currently flopping, bruh. Take a step back and stop getting erect over tech strategies. Ask if people even want this first, and then concern yourself with user requirements first and foremost.
Post reply on HN