Agreed. There are many facets to this problem, so it’s difficult to get in one post, so I’ll try to reconcile the main points.

The core of what I’m trying to say, is don’t kill Linux trying to become Windows. Linux is great because it diverse, but it also has difficulties because of this. We should not change (nor destroy) the ecosystem for people who do not care to understand it.

That being said, we can also make it easier for people who do care and cooperate to make it over. But if we do this we, as Linux users, have to look at this from the right lens. The question is not “Linux users, what do you find difficult?”; this is survivorship bias. The question is “Windows users, why can’t you get Linux on your machine?”. From this framing, the real issues become a lot more apparent:

• Not savvy enough to set up USB stick
• Driver, and other hardware, issues
• Programs needed for work, or general daily usage, are unavailable
• Too much tinkering required (this is related to, but not the same as RTFM and CLI)

The first two points can be solved by purchasing a machine from a Linux OEM (i.e System76). If this is not possible, then you are going to have to do research; if this burden is too heavy, Linux is not for you.

AI has a good and valid use-case here, as it can significantly ease this process (even if it’s only right 60% of the time).

Linux may not have an alternative for your preferred programs; if this burden is too heavy, Linux is not for you.

Developers should follow open guidelines (i.e POSIX). If they refuse to, there is nothing Linux can (nor should) do about it.

The last point can be solved by distro choice, we completely agree here. The problem is finding said distro, which is difficult. For example, I’ve never heard of Ublue until your post. I appreciate distros that handle defaults and don’t push breaking changes. The community can make this better by having a dedicated website (with a decision tree) for choosing a distro, but this has its own set of issues.

No matter, the responsibility falls on the user to pick the right distro; if this burden is too heavy, Linux is not for you.

Linux Mint is a great distro, and I’m happy it works for you.

In terms of mass-adoption though, the fatal point is probably putting a Linux ISO on a thumb drive. Like I said prior, we must be aware of survivorship bias. You don’t care much for the terminal - but you made it through.

The people that didn’t make it through probably failed from the thumb drive step. I only say this from personal experience, because when I first installed Linux, I was very determined and came extremely close to giving up at this step. And I only got through because I happened to find an obscure forum about how Rufus needed a special setting for my machine.

P.S. I also was not tech savvy, but I wasn’t completely lost either - and I still struggled really hard here.

This is actually a really deep rabbit-hole. To avoid typing a novel, I’m going to cut out a lot of nuance.

Windows is installed by default on machines. Since people do not change defaults (many studies have been done on this), this is checkmate. As long as this is true, Linux will not have a major (20%+) market share.

So this has to start from the OEM. Several Linux OEMs exist (i.e Tuxedo Computers, System76, Framework) but they cannot compete with the Microsoft network. Those who are interested in Linux, but are not tech savvy, really really really should buy their device from a Linux OEM.

Driver issues are near non-existent on Linux OEM hardware. So software is the next step; and let me tell you, developing for Linux is rough. There are 2 window servers, 2 graphic stacks, 2 desktop environments, 2 coding standards, 2 C libraries,… you get the point. A lot of this can be abstracted, but it takes genuine work to do - and may be obsoleted in a month; meaning no company will do this.

All to say, creating “magically working” apps - even with a lot of monetary support - is a herculean task. Even Valve (who is FLUSHED with cash) gave up and just decided to make their own distro (SteamOS).

A lot of issues also just require personal tweaks due to open-source software being extraordinarily bad at setting sane defaults. With something like Windows, you can hire people to make this better. Who do you hire to fix the defaults for 300 independent projects? And will the devs even listen to them?

I could keep going, but you get the point, the buck is going to have to stop at the user for a lot of things.

The best solution (in my opinion) is to have specialized distros and have people choose from them. Want to game? SteamOS. Want to dev? Fedora. Want to surf the web? Linux Mint. Creating, and more importantly accurately listing, specialized distros will make lives easier. Leave the defaults to the devs, just download the “vibe” you want.

Windows users and Linux users are not seeking the same thing from their machines. The common mistake I often see from Linux advocates.

From personal experience, when I was a Windows user, I didn’t care (or even know) about privacy, open-source software, nor owning my machine. I didn’t care if I had to sign up for a Microsoft account, and I never changed defaults ever (except for my wallpaper). I just wanted the machine to turn-on, work, and play some games.

Why am I bringing this up? Because Linux requires the user care about their machine and defaults. You need to know your architecture, graphics card, and threat-model. You need to know what your apps are called and where they come from. You need to know what tools you need to troubleshoot (and devs will not help you). This is the biggest the pain-point of Linux. Do not succumb to the survivorship bias of RTFM or command-line.

This issue cannot be fixed from simplifying Linux interfaces (though we should do this anyway!). The soul of Linux is adventure, collaboration, and tinkering. To get the most from your machine, you’re going to have to interact with several communities. This is what makes Linux great, and frankly I do not think we should kill this for the general public - this is how you get enshittification.

The general public needs to understand that incompetence (being brain-dead) will lead to misery. It is simply the rule of the land. You need to care and you need to collaborate. We should not welcome nor accommodate users that refuse to do this.

You make great points, and I will not necessarily refute any of them. This is why I said prior that I wanted a bunch of mathematicians to work towards a solution to this. There are many small and careful considerations that have to be made.

I think a heuristic (simplified model) may work better than trying to flat out solve it. As I said, this is not to refute, just a thought.

First, the problem is fundamentally chaotic (as you’ve said) there is no point in trying to accurately predict (solve) the outcome. Choosing “properties” that tend to be consistent, and then basing “success” off of those may be the more practical option. What these “properties” are would depend on consensus - models have elements you deem important, which may not actually be (as you’ve said). This is just something that needs RFC - hence needing a group of mathies.

Secondly, whatever the solution turns out to be needs to actually be do-able for the average joe. If there is a straight up solution, and it turns out to complex, I think it would be less valuable than a simple-to-do heuristic. If people don’t follow up it’s just worthless - and seeing how long it takes people to do very simple things, we’ll be waiting hundreds of years.

I’ll read the two articles you linked (I’ve read the abstracts) but it’ll be a slow burn.

Just guessing, it likely failed due to:

• Being relatively unknown
• I2P being relatively small
• Java being a non-standardized language
• No data verification (malicious host can corrupt data)
• Poor UI and UX
• The main developer leaving

These are killshots to this type of service as people need to develop/extend/use it - for it to be viable. It is in the right direction, but (similar to many cornerstoning attempts in FOSS) is not handled gracefully.

Yeah the uni-directional relationships are also significant. It also happens to translate well; if Mr.Beast goes to randomcorp.com he is almost guaranteed to pull more people over than if SchmoeJoe went. Those people in turn would cause the website to be a more attractive option (less weight on the edge).

That would mean that there even is nuance within tyranny, which is funny to think about.

There’s also the possibility of cycles! What a fun rabbit-hole. Definitely worth a thesis paper or large-scale open discussion.

P.S. Also agreed that with a “limit” it is not TSP, and is much simpler. It evolves into TSP only when you think about a message originating from a source and making it to everyone - with the same effect for responses.

First off, agreed that monkey brain + internet = unsolved.

Second, I think that this overall is a math problem and what you’re describing is metadata. Before I continue, there are many ways to solve and interpret problems - this is just how I see it.

If you think about this as a graph, it makes a lot more sense as a math problem. People want to communicate and the message has to reach each of them once through the shortest route. In essence, this becomes the “Traveling Salesman Problem”.

Next, imagine the distance between points on the graph become longer (when people group together) and shorter (when people split apart) - we now have described tyranny of the majority.

What you are describing (from my perspective) is the cost of going from one part of the graph to the other. This indeed is a very important part of the problem and directly relates to the tyranny, but does not solve it. Instead to solve this problem, we would have to find a way to standardize the distance between any two points in the graph (i.e it cannot take more than 30 feet to reach any given destination).

I cannot begin to describe how difficult this would be, but my brain is telling me it’s solvable.

The comments (and your github post) helped me think about this a bit deeper. This is why discussion is helpful.

There’s a lot of nuance to be had here, but it’s a conversation for another time.

You bring up something interesting though

IRL you would leave but on the internet you want them to leave.

I wonder if this is because people view these spaces as a home or a “third place”. Like if someone did something offensive in your home, you would indeed ask (or force) them to leave.

People also find it insanely difficult to “leave” because all of their friends are on the platform. Since it’s almost never open-protocol, that means being locked to said space - so you can only get people you don’t like to leave.

We generally agree the moderation has become overbearing. I would argue most of it is straight up ineffective and performative. We need actual data and science backing moderation policies, not just “this feels good”.

This is interesting perspective.

If I’m interpreting what your saying correctly, this becomes an alternation of the “Traveling Salesman Problem” - where people are the nodes, sending information is the destination, and medium of communication is the weight. The goal being finding the shortest path for two-way communication (go to destination and return).

If this is the case, “tyranny of majority” is indeed a very difficult problem to solve. This phenomena causes the weights of the graph to become change based on the number of surrounding nodes. Higher when less nodes (i.e Lemmy) and lower when more nodes (i.e Reddit).

To go even further, companies are manipulating their weights (creating closed ecosystems, etc) to make is so two-way communication is only viable within their bubble (think an edge of infinite weight). And it would also mean that it truly is unreasonable to expect laymen to “memorize the graph” (know a forum for everything) - it indeed would be just easier to know a subsection (i.e Reddit, Facebook, etc)

I’m just spitballing here, but a lot to interpret if true.

Reddit was is a peculiar position, flooded by the masses isn’t really what I would call old reddit. Known by them yes (for answers), used by them no.

I think that development is fairly recent, maybe from 2019? And around that time, Reddit started to go to shit.

I agree with the government portion. I’m not really sure what the solution would be, which is why I want math folk working on it - but discussion is always helpful.

The problem is engagement - which in turn means building a community.

If someone goes to your forum, hijacks your posts, and them puts them on Reddit; there is no incentive to use your platform, which stifles and dies since no one is communicating on it. This is also part of the tyranny, as people can figuratively walk into your house, eat all your food, and then expect you to restock.

On the old internet, this was not really an issue since there was a better culture. But if all your friends use Reddit, and all the content is reposted there, why use the internet forum?

I agree that we need solid alternatives, but this doesn’t really tackle the tyranny of the majority problem. We need people to use the platforms for communication, otherwise it has not solved the problem.

For example, if you use Signal but every single one of your friends use WhatsApp and refuses to switch (which is common), then you are forced to use WhatsApp. This is why it is tyranny.

EU can facilitate thousands of platforms, but if the masses don’t use them it’s pointless.

Federated-platforms are kind of a step in the right direction, but they’re extremely weak to internal bad actors. If lemmy.world gets one million normie users, then cuts off the entire federation - then Lemmy has effectively been hijacked and set back 10 years.

It’s frustrating but true.

To use an extreme example, if I saw someone just spamming the hard-R I would want their comment immediately removed. The rhetoric makes the space becomes completely unserious; just not a good environment.

The funniest part is that this mirrors real life. If someone did that IRL, I would just leave.

I am not going to argue in terms of right/wrong because I’m just not equipped to. But in terms of platform result, I do not want to participate on a 4Chan clone - because it always leads to unserious discussion, bad faith, and death.

I think this is an XY problem.

People keep trying to bring back the old internet ; This is an broken and outdated solution.

The root problem (in my opinion) is that we need to share critical information to the masses, but the masses introduce “tyranny of the majority”. It’s a really tricky problem to figure out, and I really really really want mathematicians working on this.

If you live in the states, the Electoral College exists because they were looking for a practical solution to this problem. Considering the outcomes, it did not work - but there is no shame in this, as I think this is actually a really hard problem to solve.

The only known solution is to not share information to the masses (a.k.a keeping the normies out). In essence, this is what the old internet was - and a large part of what made it great. But this is not correct as it does not meet the criteria of the problem. Nor does it translate well, since your neighbors are apart of the masses.

If anyone has any thoughts on this, please share. If you do math for a living, please gather your friends and make an open-thesis about this.

EDIT

After some discussion in the comments, I have a general hypothesis:

• One platform, one name.

People must be able to distinguish the resource they are accessing - highly recommended this process be easy. This provides consistent “edges”.

• Open protocols only.

Looking at “tyranny of the majority” from a different perspective, one answer is to standardize how people communicate. This means no closed ecosystems nor convoluted protocols. This provides “standard weight” while preventing “infinite weight”.

• Server-wide censorship cannot be allowed.

This eliminates every platform I know of. Servers should not be given any tools to prevent incoming nor outgoing data. People should handle moderation individually - sane UI can of course be made available (BlueSky block filters could be inspiration?). Blocking should only be handled by the “nodes”, this also prevents “infinite weight”.

I find it really funny that this conclusion kind of alludes to the early internet in a lot of ways. Maybe it wasn’t the internet-forums, but the internet itself that has changed.

I’ll leave the privacy policy here and let people decide for themselves.

They keep two types of logs. An identifiable one which is deleted in 24-48 hours (dns0 and quad9 also do this) and an anonymized one. There is no mention of “business partners”; and it also says explicitly that the information is not used to target ads.

As the privacy policy and service reads, it is not a honeypot. However, Google generally does not act in good faith, so there’s no telling if they have 100% adhered to the policy.

No matter, to make calculated and informed decisions, we should have all the facts in order.

Wouldn’t you just use AFS, CEPH, NFS, or 9p?

I really don’t want to be that guy, but isn’t SSHFS (FUSE) actually a terrible option when compared to an actual file-system? MacOS isn’t really missing out on much there.

The most painful part of MacOS (which makes it downright unbearable for me) is that system configuration files are XML. It’s an absolute nightmare.

Like I said prior, there is nuance to be had here.

We agree that Google products are generally a honeypot (good products that lure you in), but which products are honeypots are important.

You very likely want to avoid Chrome, Gemini, and Google Search - but 8.8.8.8 is not a honeypot, it is a loss-leader. You will be lured in from 8.8.8.8 if you say “huh. this is a great service. is there anymore?”, but 8.8.8.8 itself is not a malignant service.

Google does not automatically mean bad. It is dangerous precedent to blanket ban and remove nuance.

8.8.8.8 is an excellent service, and provides genuine privacy gains. The largest downside being that it is such a massive target for bad-faith and ignorant actors - like the Italian government.

Tor itself has a pretty good routing scheme that seems like it could replace DNS entirely. There are obvious (but surmountable) UX issues and there may be scalability issues - but it is 100% worth investigating.