This sounds harsh but not everyone deserves to be heard all of the time- myself included.
One recent example is Piers Morgan in the UK, who randomly decided to criticise Emma Raducanu, a rather likeable British teenager, who just won the US open without dropping a set (including in qualifying rounds)! I can't understand why he would do that, other than to stir up some controversy, and get himself back in the public eye. Either way, it worked.
This can be seen on websites like: https://socialblade.com/twitter/user/rammjaeger/monthly Picture: https://i.imgur.com/poc1QcE.png
Edit: when it comes to Piers Morgan, it does not seem to have been worth it. The last bump of followers happened in March.
Reference: https://socialblade.com/twitter/user/piersmorgan/monthly Picture: https://i.imgur.com/s7145UK.png
"I think when I was younger I wanted to tell everybody everything, because I thought I was so damn interesting. Then I heard the snoring."
The exact same rules don't quite apply to real life, but they make for a nice comparison: in a small group of people (say, 4-5 people) no formal rules are needed. Generally, everyone will have a chance to speak and be heard. Imagine a a much larger group, say, 300 people. There's no way everyone can meaningfully be heard. In the real world no one goes into a group of 300 people to communicate with those 300 people. Usually there is a speaker or performer communicating them as an audience. Even if those 300 people wanted to set aside the time to listen to every audience member, it'd be meaningless: no one would remember anything meaningful from most of the 300 different speakers.
In any case, it should be obvious that group dynamics and the need for structure are modified by group size. A very large online community can never be intimate, and perfectly healthy communities can be irreparably damaged simply be in influx of people.
* Of course, there are exceptions, and I actually use hnreplies.com to get emails when someone replies to me, but there's enough of a delay that I usually am able to avoid getting caught up in a back-and-forth argument. Especially since, by the time I see the replies, someone else has already clarified my position :)
I get it though, there are mental tiers I know where I am. Thankfully I can at least make a decent living/do my own thing.
I'm also aware of effort too. Looking into something/asking.
Engagement stats and views are the wrong KPIs to focus on for social media aites and most of us know it yet Product Managers are just trying to stay n their jobs and collect checks.
If you want to really succeed in creating a community like HN that lasts much longer than a cheap start up like ClubHouse, the key is to stay local, dedicated to a niche, keep ads controlled carefully, and be happy with smaller-scale operations rather than trying to dominate the world. Dominating the world is a very risky and high overhead venture meant only for those who can ask for a small million dollar loan from their father.
There can be automated systems to fix that - to rank contributions, with the aim to boost the signal and reduce the noise, with the final aim to facilitate a lean and productive debate and information sharing; and also to connect contributions, to reduce dispersion, etc.
Such systems to rank contributions are related to the theory of voting systems (quantitative ranking of voters and options).
The aim of many big "social networks" is not «to facilitate a lean and productive debate and information sharing».
No automated system is going to fix eternal September style problems. And before you suggest that I could simply follow the groups I'm interested in, that's still not the same thing because the people in those groups will still be engaging with/distracted by the people I'm not following.
When you upvote an item - you connect stronger to people who upvoted that same item.
When you downvote an item - you connection to those who upvoted it becomes weaker.
The content is ranked based on how strongly you are connected to people you upvoted each item.
That way when a lot of clueless/uninformed people come in and start upvoting useless to you content - they don't create noise in your personalized content list.
The content you see comes from people who have proven to find useful content for you.
This is how my hobby project https://linklonk.com works.
I did a Show HN a week ago and that post has more details: https://news.ycombinator.com/item?id=28405659
No, absolutely not. Voters can be categorized, clusterized and ranked, already as part of voting systems theory (one single example in the latter: election of electors), and more beyond it.
> No automated system is going to fix eternal September style problems
And why not. One example: views filtering contribution by score.
> people in those groups will still be engaging with/distracted by
Not necessarily, if they are given the option to avoid that.
Most social media platforms are simply to big to moderate in any meaningful way, in my opinion. AI has so far show to be ineffective, as show by comments and open letters from Facebooks human moderation staff.
I could be wrong, I hope I am wrong.
Do we want to live in a sophisticated democracy as opposed to a naïve democracy? Some would.
Would you like to be part of an information exchange and debate platform using sophisticated systems as opposed to an open billboard? Some would.
It is very doable. It was done in history and it can be done technically in many ways.
Take for example HN: before you can downvote, you must have consolidated your position as contributor. Very simple - yet affecting the result, the "experience". Much more sophisticated techniques can be used.
Malicious parties could attack a system exploiting its weaknesses: yes, this is why take precautions like testing systems theoretically and practically, and implement defence systems.
> automating moderating
The point is not to automate moderation - it is not about the work of DanG. It is about reducing the noise generated by users "in development" and boosting the signal donated by "developed" users.
In the area of "ranking according to the favouring of favoured users, recursively" is a way to implement it.
It was a thread about an application correcting for some aspects of ADHD. With ADHD persons there is already an information filtering disorder of variable degree; specifically an inability to separate ones-self from a context of conversation, subject matter, or even the logical distance therein. This means that under an extreme case it only takes a single person to mimic the effects of a social network echo chamber of heightened sensitivity. That thread was bizarre in that many comment replies were wildly on a tangent from their respective GP comment and the emotion expressed was completely off the chart for HN.
It really was too much to self-moderate your way through. The volume was just too much. As well, you couldn't sort through the trash to find the pearls.
The volume was a trickle compared to volume today. I don't have a source for this assertion, but I think it is reasonable.
I mean, it’s still better than closed source SDKs with bugs unfixed for years and books written about working around them, but that would be a pretty low bar anyway.
> fauxpen source
> Etymology: a term invented by Phil Marsosudiro at a dinner party in North Carolina
> Date: 2 May 2009
> A description of software that claims to be open source, but lacks the full freedoms required by the Open Source Definition. http://www.opensource.org/docs/definition.php
> synonyms: see open core, neo-proprietary
> antonyms: see Linux kernel, Apache, OpenNMS