Hive has a huge issue with the timelines. The posts we make are not tailored to us, but generalized for everyone. First, let me list the pros of this approach:
- It (POSSIBLY) allows new creators to reach more people
- There is less bias and more diversity of content
- You can see all the people you follow in order without some being shadowed by others through bias
Image source: https://sproutsocial.com/insights/twitter-algorithm/
Our algorithm is not without flaws. Smaller creators often get shadowed out here too. We have a small oligarchy of people consistently making hundreds of dollars per post, regardless of what they post, which results on them are always showing up on hot/trending/rising which results on them more consistently making more money! It is a snowball effect! Not that Twitter wouldn't be the same... But I mean, it is already happening here...
Ecency is trying to implement their version of recommended posts, but it is not perfect yet. I praise their efforts though, it is immensely valuable! I myself use Ecency because the posts recommendations are a great value to help me find newer posts and creators!
But if Twitter opens their source code, another front-end for Hive could, maybe, try to implement it! Imagine us having an algorithm that helps people connect with like minded individuals? That would be a revolution for Hive because it would allow people to more easily find content that interests them, would help the consumption of relatable content to grow, and content in general be consumed and appreciated.
Or will it? Maybe using their algorithm would bring cons without pros, maybe it will worsen the gap between the few popular content creators and the many users who are forgotten and rarely make a few bucks on a post only every now and then...
Regardless if an algorithm is good or bad, I think that if they do open the source code, someone will eventually try to port it here. Only then we will be able to evaluate the consequences and what we gain from it.
I am very curious to see how that unfolds.
Talking about descentralization
Don't forget, I am a witness too! Have been for almost one year. Helping me in my campaign is immensely appreciated! It is hard to reach more people and to grow.
I am a small witness, but my node is stable. I have only missed 11 blocks that I were delegated to produce (due to maintenance) while bigger witnesses than me have missed hundreds or thousands! Of course they have their valid reasons, but I am proud of how well my witness node, although small, has been performing.
The value of twitter lies in its user base, not in it's code. Many people have replicated the short form structure. But without the user base, the structure is dull.
The open sourcing will simply expose the tricks twitter uses to determine whose voice gets heard.
I have no idea while anyone would want to replicate that.
What is Twitter other than a huge database and the algorithms? You can buy database scaling as a service. If you combine that with Twitter algorithms, how does that new service differ from Twitter?
It doesn't, but having a front-end with a good recommendation algorithm would give choice to Hive users.
I can say for myself, that sometimes I scroll a lot here without finding very interesting content. That could be the community, of course, but maybe an algorithm would help. Ecency has helped with their "read next" tool.
It's not that complicated to replicate what's behind a tool like Twitter. It's all based on tracking user behavior and creating segments of interest. Rather, the greatest difficulty is finding a way to make users aware of this mechanism and allow them to control it, but this is a problem that not even the largest social networks have yet managed to solve completely.
Some of us are here and not on Twitter, because Twitter algorithm is pure shit, it's harmful and empowers superficial and stupid content over great stuff.
Replicating the algorithm here, will make this place a shit like Twitter.
Let's not make that mistake.