A New Focus for SporeStack

Foreward

Just a heads up, this is a bit more of a ramble of some thoughts than a news post. And SporeStack will stay as it is for the forseeable future. In fact, I think the current offerings will remain the same and perhaps expand at times -- those will still be available. I just wasn't sure how to word the title in another manner explaining that there is a new focus, but the old focus is unchanged.

Post-foreward

When I started SporeStack, I felt it was silly to require an email address and/or a credit card to rent a server. Inspired from a website that was a torrent downloading service which completely skipped the whole login step -- you paid with Bitcoin and got to use the service. The legality of that service aside, which I think is no longer around, it made me think the same for servers. And why should anyone need to know your email, name, phone number, and so forth just to host a blog? Of course there's many other uses for servers than just blog hosting, but it's one that comes to mind.

SporeStack has proven quite successful, now in its 8th year. I didn't know where it would go or if I'd have any customers. In fact, I had started a few other Bitcoin services before. One had one customer and the rest had none. And the goal of providing servers without Know-Your-Customer has proven successful. It does that. Another concern of mine was over free speech. And to my knowledge, no server that we host at Vultr or DigitalOcean has been removed for free speech reasons. This made hosting on Vultr and DigitalOcean an easy option. They both provide great services with a good API, that SporeStack is able to work with. The easiest way for a lot of people to use SporeStack.

I did pursue self-hosting, and did so very briefly. But it was overcomplicated in design and not used, so I moved on from it. There seemed to be no reason to pursue self-hosting at the time. And it seems that SporeStack meets the needs of the market pretty well.

Though it does these things, I find myself feeling like there is a next step that ought to be taken.

Next steps / A new focus

I really don't know what most do with SporeStack. Since hostnames are in plaintext, if someone says "blog," I assume it hosts a blog. Or if there's a token message that explains what it is. Now of course there's the AUP-violators who I get an email with a description of what was being done, like torrenting some TV show or a phishing site trying to con someone. But my default remains on the "blog" focus, or around communication. A Jabber server, email, blog, etc.

I think those things because original content and communication are the best parts of the Internet. Yet year-by-year I've watched the Internet become more and more commercialized. Search engine results have gotten worse and worse. It's like "the good stuff" has vanished, is drowned out, or is hidden now. I don't like money for the sake of money, or progress for the sake of progress. It seems as if every opportunity possible to profit on dopamine and attention has been found and exploited to the maximum. "Big Tech" companies know things about their users that the users themselves probably aren't even aware of. Clever tracking tractics from cookies to screen resolution, typing style to keywords, IP addresses to credit card numbers. We are the product in a dystopian reality. We are gamed and corned all of the time.

It wasn't always the case. Reddit used to be a great place to collaborate. Now Reddit is one of the top sources for AI to train on. Our ad-free, commercial-interest free conversations are being used to train our replacements. All of that code on Github, once a great place to share with your resume, now enables fancy code autocompletion and rewriting for people too lazy to use documentation and peers at their own direction. The system does it for you now, whether you understand it or not. And I'm sure it can be faster at times, and despite the mishaps, it will offer an edge to some. But what is the edge for? It's going to be harder and harder to be competitive without using AI. AI using workers will replace workers who don't (a lot of the time.) And AI agents will eventually replace them. Finally, Universal Basic Income will probably step up, and in my opinion it's going to greatly increase depression and suicide rates.

Though I've felt a number of jobs have been meaningless, I do like work. I like working on something challenging and meaningful. Society appears to be shifting to a work-is-bad mentality, trying to avoid it at all costs. And the costs are great -- the biggest tech companies being even more intimately involved with your lives, files, and thoughts.

I'll admit that I wonder if we're better off without technology at all. It's two steps forward for every ten back. But certainly, there are less bad ways of using it than others. And there can be some wonderful collaboration and sharing of knowledge with it.

I am just not on the train of wanting everything to be bigger, faster, stronger, smaller, more intelligent, etc. I don't want an AI superintelligence to be better than me at everything I could ever do. I don't want to compete with an AI model that I've unwittingly helped train. What's the point? Sure, there is some progress that I am good with. But websites built by AI for SEO purposes that offer no real information, I am not happy with. I am not happy with the TikTok and Instagram model of constant dopamine hits to drown out the rest of life and make us discontent.

I am also not happy with the environmental disaster that technology has brought, and is increasing. Datacenter power use has doubled or tripled in the past couple of years just over AI! Something already high -- too high, managing to go up so much higher.

What the new focus actually is

From what I've already said, you might fairly conclude that I am a proponent of less and going backwards rather than forwards. This is indeed the case. I am also increasingly environmentally concious after a number of experiences. Scrapping/recycling is another new interest of mine, helping me realize that not everything needs to end up in a landfill. There's a lot of lesser evil options out there.

All of this to say, Vultr and DigitalOcean are great, but I've been a bit lazy to use them as a crutch. I accomplished one goal -- privacy minded servers. But how should hosting look, if it ought to exist at all? I don't think we should be chasing faster processors, more memory, and using computing in ways that don't really benefit humanity in the long run.

I think that we should be using less power and more second hand equipment. I think that when harddrives go bad, they should be scrapped to be recycled. There's a surprising amount on them that can be, even on motherboards. The more that is recycled, the less that is mined. The less that is mined, the more nature we can enjoy.

So I've been figuring out what hosting would look like, if I felt it was done "properly." Older, low power gear that uses very little power and no equipment designed for AI training/use. My last post was about Monero mining and I have to admit that proof of work isn't the best thing. Certainly not for the environment. But it's the most viable option I'm aware of, at this time, for people to be able to pay for SporeStack privately.

I just don't want to be building bigger and bigger datacenters. Nor do I want to pay DigitalOcean and Vultr to do the same. And I'm not sure what my users will want exactly, but I at least want to be able to offer hosting that is deliberately less, rather than always chasing the moon.

For now, I'll try to in-house a few infrastructure servers and go from there. But I would like to offer low power VPS servers. Probably 256-512MB minimum for memory. IPv6 only for now, maybe I can get a IPv4 block at some point. I'd like to offer SNI-based proxying for IPv4 backwards compatibility. And I'm not against renting out IPv4s -- they're just hard to get and I think, ironically, it's time to move on to IPv6. IPv6 doesn't really require any more of equipment -- it just makes addressing easier. So yes, it is progress which I often argue against, but I don't feel that it offers nothing.

I'd like to price the smallest flavors cheaper than the DigitalOcean or Vultr offerings. I have one datacenter in mind to start with, but would like to have more than one. It's still nice to offer the option of redundancy, of course. I also think that some degree of solar powered hosting is possible, using lead acid batteries (US-sourced, better in the cold, and very recyclable) and used solar panels. I'd like to do this without active cooling (other than possibly fans.)

SporeStack does have an ASN and a IPv6 block to work with, but we will see how it plays out.

I don't know how many of you have read this far, and how many of you will actually care about this. It may just be extra work on my part and not a profitable venture, but at least getting my own stuff hosted this way, not feeding into "Big Tech" as much, would make me feel a bit better about my small part on the Internet.

Thanks for reading!

-Mycelium

PS: If you're tired of the "web 3.0," Gemini is very interesting.