Hey! Thanks for the comment!

> But why should a (public) library be interested in providing such services? For funding? What about costs?

Public institutions like libraries are usually funded through government mandates. We as citizens decided that having free access to books is a good thing and nations, states, and municipalities dedicate tax dollars to fund those programs. So, if we decided providing internet-based services through the library was also important, we'd enact mandates for that, too.

Not saying that's likely, but it is possible.

> At-cost service[s] provided by the libraries will probably collapse as soon as a for-profit company comes up with a cheaper plan.

At-cost actually means it couldn't be cheaper (at least if economies of scale are equal). That gets a little hairy because companies like Google can provide services like Photos and Drive for "free" because they make so much money selling search data, but generally speaking that's the deal.

> Also, host by a library still creates centralized service, which comes with all problems that a centralized service inherits. It only shifts the problem, not solving it.

Totally agreed – if there was only one library. But, there are tons! And as I mentioned, if the services are based on interoperable standards, you could easily move your data between services and have them talk to each other so there's no vendor lock-in. Think ActivityPub for files.

Thanks again for reading and engaging in the discussion!

> Not saying that's likely, but it is possible.

I'm afraid that's not how things work.

For example, it is possible for everyone to self-host their own service, it's true and everyone can do that right now. However, that's not what happened.

In reality, people oped in to use these cloud services, with full understanding of it's downsides, in exchange for convenience and low cost for themselves.

And as I've pointed out in my last comment, the companies has optimized their services so well, it made it very hard for a library, which is a "outsider" in the service field to compete. The library and it's lacking of technical know-hows, political resources etc will eventually doom the service, making it a product people only buy with higher-than-market price as a show off of their goodwell. That's not a sustainable business model or any model, really.

You must know all these things if you want to change the world for the better. Dreaming on vague an idea is easy, making things actually work is hard.

Also, you need to reconsider the meaning of the term "self-hosting", because unlike what people widely believed, "self-hosting" is not equivalent to "setup and running a server by yourself". The definition is much wider.

In fact, if you ever downloaded something from the Internet and storing it on your computer for later use, that's self-hosting, you just not sharing it with remote access. And if you copied what you've downloaded to multiple computers, then you've just created redundant and distributed backups.

But if you taking that into consideration, that downloading is self-hosting, then there's even less reason to use self-hosting service provided by a library, because why should you pay extra for all of that if you have already done it with no added cost? Just for a expensive remote backup maintained by people who has other jobs?

    - "if there was only one library. But, there are tons!" - This is just empty dream. Not many will be welling to do it in reality.
    - "if the services are based on interoperable standards" - Designing protocol is not easy, and by simply defining the protocol you also restricts what the protocol can do. A file sharing protocol? Good luck getting it to also do bookmark or chat. So, empty dream based on empty dreams?

I do feel you have a good heart and wanted a good change, but you look inexperienced. I would recommend that you learn the industry or simply work in it for a few years, then maybe you'll come up with something that actually works.