No, the issues are quite technical too, but they cannot be fixed just by changing the feed selection algorithm - they are core to twitter's design.
Twitter is essentially a marketing machine / popularity contest hybrid.
There are no barriers to publishing something on twitter:
- you don't have to be correct.
- you don't have to ellaborate your claims (In fact, you specifically cant - the space is limited enough to disallow it)
- you don't have to provide any references to your claims, and you can easily post screenshots or images from other sources without referencing and frame them in whatever light you wish.
There are even fewer limits to boosting something by retweeting it. You don't even have to read it, and "want to read the article first" sounds like a joke precisely because the rest of the platform never really cared about that.
By themselves, the above properties would not be that big of a deal. However, the retweet mechanic essentially creates an evolutionary environment for ideas. Tweets essentially become brain viruses. Their goal is to replicate by retweeting i.e. by causing a human to click the retweet button on them.
The environment formed by twitter itself as a platform, as well as its users, essentially "decides" which "viruses" are more fit to replicate and which ones aren't.
Now the question is how much of this is twitter's fault and how much is the environment?
Sure, humans can take some of the blame. Humans are faulty and fragile and prone to biases / getting triggered / reacting emotionally. But that's precisely why we build things - to compensate for our faults. You could easily imagine alternative systems which instead of being optimized for marketing (often an exercise in bending the truth as much as possible in a way that benefits someone), are optimized for something else:
- retweeting something could require you to check the references and ensure things are correct
- the character limit is inverted i.e. humans will need to write
- every post is peer reviewed; some basics are automatically checked (e.g. the presence of references)
Or, we could simply get rid of the exponential aspect. Reddit suffers from some of the same flaws, for example, but they're less pronounced - because its more of a linear rather than an exponential environment where best voted things surface to similar number of people (followers of the subreddit).
Its this extreme environment with exponential properties (https://www.science.org/doi/10.1126/science.aap9559) which makes us feel like "all of twitter" sucks and people suck. It doesn't really, but we don't get to see the parts that don't, because they have orders of magnitude less spread - the environment is extremely harsh to things not optimized for it.
I actually have many examples of scientists posting throughly checked studies posting one "interesting" (but poorly done) study involving animals which they didn't think much of, and that study getting 2-3 orders of magnitude more tweets than the rest of their material (because it looks novel), thereby defining most people's perception of them. Some find it embarassing, others that are after the attention more than they're after the science decide to go the dark path or get surrounded by followers that feed their delusion.
Humans are flawed, yes... but we could do so much better.