Hacker News Reader

Comment by waldopat | Hacker News Reader

Moxie Marlinspike nailed this in his web3 critique from a couple years ago: "People don't want to run their own servers, and never will. The premise for web1 was that everyone on the internet would be both a publisher and consumer of content as well as infrastructure... However – and I don't think this can be emphasized enough – that is not what people want."

That said, the discussion seems stuck in a false binary between the control of self-hosting and the convenience of corporate services, but I think what the market wants is a third way that provides both control and convenience.

And to be honest, public libraries already do this, y'all. GO GET A LIBRARY CARD. You can stream from Kanopy at home.

https://moxie.org/2022/01/07/web3-first-impressions.html

View on HN

Exactly! Here in Spain there is a network of web libraries that are proxies of your corresponding local library that allow lending as long as you have a library card. You even have magazines and newspapers, I know because I developed such network!

That's amazing. Do you have a reference to it? I'd love to learn more. I also have some extended family in Spain.

Hell yeah! Thank you for your work!

People don't want to "actively spend effort and mind power" to run their own servers

But purely outcome wise, many people want the benefits of hosting their own servers

Totally. You see this happen a lot. Centralization happens for a reason, even if it's a bugbear of a concept these days. It's because the market is demanding it.

How come that a public library, one of the earliest examples of centralized information infrastructure, is not an example of outsourcing and relinquishing control? Instead of your own (small) books collection you get to use some external (huge) book collection. But now you only can borrow a physical book, or some recorded media. You have to return it, and making a copy for personal use only is still a bit problematic.

Either you own and control something, or you do not, there's no third option. A best, you can outsource your stuff piecewise: run your own software on a cloud VM, or bring your own furniture into a rented apartment, or give a valet the keys to the car you own for parking, etc. But there's always some relinquishing of control in exchange to some other aspect of efficiency / comfort.

It's also easy to mistake what most people want for what everyone wants, and miss an important market.

> Either you own and control something, or you do not, there's no third option.

I think there's a full spectrum you're missing. You can own something with other people, and your level of control can be continuous, not discrete & binary. For example, my public library is funded by my local government, which I can influence with lobbying and voting. I can join the board of the library, and I can just go and talk to the librarians in charge to influence their decisions.

In an individualist consumerist mindset things are pretty stark : full self-hosting or full submission. If you reject that mindset there are many more options.

Couldn't agree more – thank you!

Yay civic engagement!

As a public institution you, the citizen, own it. What you are talking about is hoarding access. You want complete unfettered access to content without barriers and without friction. Typically the only way to do that is via pirating.

Let me remind you of the open source credo about free as in freedom not free beer. You are right that there may be exchanges or compromises at play, but it was a bit shocking to me when talking about what is essentially the digital commons that no one mentioned a library, which exists.

I'm also saying from a practical perspective if you want to stream movies without giving money to big tech, you can literally do that tonight with a library card. The infrastructure already exists.

> As a public institution you, the citizen, own it.

Nominally, yes. In terms of that meaning anything, no. The benefit of ownership is not exclusivity, but control. If the library doesn't have a book (or other piece of media, of course), I have no power to influence them to get it despite that theoretical ownership. If the librarian decides a book is offensive and removes it from the collection, I have no power to influence them to keep it. I have to live with someone else's decisions about what the library does and does not contain, just like with a commercial service. So my nominal ownership really means nothing at all.

You should go talk to your librarians, you can totally influence all these things!

Right. You want free beer not freedom.

> That said, the discussion seems stuck in a false binary between the control of self-hosting and the convenience of corporate services, but I think what the market wants is a third way that provides both control and convenience.

If I were to run my own version of Google Photos and the like, I'd probably go with the hybrid option:

Run all the software I'd run if I was self-hosting, but in the cloud, possibly with a backup in a second cloud. ie, put my photos in Backblaze B2, with second copies in S3 or something.

Personally, half the reason I use Google Photos is so that if my house burns down, I don't lose my pictures. A self-hosted server running under my desk doesn't carry that guarantee. Backups are off-site for a reason.

Though maybe self-hosted at home with a single cloud backup would be good enough.

When running your own backup server, you're forgetting about scenario(however less-likely) when Google Photos will loose your photos, or if your google account gets banned with no ability to call anyone in Google to dispute that. In this case you can safely rely on your own backup to have those files at hand.

I was skeptical about this scenario until one day Gmail lost 1 year worth of my emails. It's just gone. All other emails are there, but not this particular year. And there is no person who you can call to talk about that.

I want something easy to set up that lets me easily backup things like this within a user-chosen circle of family or friends. Build my own trusted "micro cloud."

have you tried CasaOs or Zima board? It's their premise your own micro cloud

They're missing the key feature I'm looking for: Decentralized backup to the same devices owned by people I choose. That's the "someone else's computer" part of what I want in a "cloud."

I can already easily run such things on my home computer. It's having remote (encrypted) backups and redundancy if my own system goes down that I'm looking for.

I don't feel like most people even know self-hosting is an option.

I wasn't aware of how most ingredients were made and what effect they had to my body. Once I learnt I started being way more careful with how I source food.

Same goes with technology: people don't understand that what they upload can be used against them or arbitrarily taken away from them, what they buy can be removed, that companies spy and abuse them. Once they do, many look into self-hosting or at least other alternatives from big tech. I've met many who want to self-host, but they lack the technical literacy to do so.

Moxie is wrong, he likes to project his own ideas as wisdom and always factually correct. P2P networks have flourished. Bittorrent, bitcoin, Tor just to name a few successful ones

The size of the whole pie has grown, and, yes, so have P2P networks.

However, centralized networks of various kinds have grown too. From application-level networks (e.g. networks messengers from Signal to Discord; payment networks from Visa to AliPay) to transport- and network-layer networks (e.g. cloud-oriented client-server topologies; Tailscale), you see growth in the centralized world too. While I don’t have hard data, I’d casually assert that that growth is comparatively far larger than the P2P growth.

Also, if you’re referring to Moxie’s classic post about centralization vs. decentralization[0], I think you’re mischaracterizing it; it doesn’t claim that P2P is altogether inferior, wrong, or any such thing. I’d summarize his point as being that anchoring your application to a decentralized P2P protocol makes your application development less agile. In other words, there are engineering tradeoffs there, between centralization and decentralization. When he weighed the tradeoffs for making Signal, he came down on the side of a centralized architecture.

[0] https://signal.org/blog/the-ecosystem-is-moving/

> When he weighed the tradeoffs for making Signal, he came down on the side of a centralized architecture.

And as a result, Signal has a single point of failure and hacker/governments attacks.

Due to its (afaik) unrivaled privacy-preserving architecture, the Signal servers have systematically less info on you than, e.g. a Matrix server. The risks of an attack should therefore be much lower.

Oversimplifying a little bit here, but I believe the most likely outcome of a Signal hack would be downtime, rather than a data leak. Contrast that with e.g. matrix.org (where the majority of Matrix users are).

> Contrast that with e.g. matrix.org (where the majority of Matrix users are)

This effectively confirms that federation is important, i.e., you shouldn't use the same server as everyone else, for your security.

> Due to its (afaik) unrivaled privacy-preserving architecture

https://news.ycombinator.com/item?id=29888228

https://news.ycombinator.com/item?id=39445976

https://github.com/signalapp/Signal-Android/issues/13842

https://news.ycombinator.com/item?id=42788647

This. There's a good reply to Moxie's article here: https://news.ycombinator.com/item?id=21936929

Thanks! There's a lot to chew on here, but it's a very important discussion.

Exactly! Thanks for your comment! I actually propose what you recommend being provided through libraries at the end of the article. Libraries rules!

I don't agree with the premise that people don't want to be part of the infra. The real problem is that gate keeping is a great business model. It is so profitable to create a wall garden that companies compete ferocely to take care of you content.

If home networks easily let you have a public server I bet they would be more common. They could of been built into modems.

Reply on news.ycombinator.com