Comment by erulabs | Hacker News Reader

Comment by erulabs | Hacker News Reader

Parent Full thread

That’s essentially what a Mastadon instance is. If it’s hosted at home or on someone’s cloud account - the difference is only in connectivity speed. I’m extremely hopefully, perhaps naively, that a new internet isn’t going to be based on blockchain or owning data or any foundational change - it will be based on the ability for people to host their own server-side software again. The App Stores democratized client side apps, but they forced a huge centralization into platforms.

What we need is a cloud platform that allows people to deploy code to computers on their home network, or any cloud, or any data center... one that encourages and assists with cloud native design and aims to reduce load rather than hope you architect incorrectly and happily charge for the waste.

Perhaps I’m being naive, but it feels like a tooling problem, not a social or political or economic one. Anyways, I’ve quit my job at Stripe and just finished YCS19 with the goal of closing the gap between Facebook/Mastadon, Medium/yourBlog, Minecraft.com and a docker image...

And you’re totally correct about renting home server “boxes” too - except most house holds already have plenty of compute and connectivity - it’s just that installing / managing server software is completely out of reach to most people (and some developers!). It doesn’t need to be grandmas running router software - just the family nerd able to create on the internet again - and reliable enough software you’d actually invite grandma to signup!

> What we need is a cloud platform that allows people to deploy code to computers on their home network, or any cloud, or any data center... one that encourages and assists with cloud native design and aims to reduce load rather than hope you architect incorrectly and happily charge for the waste.

Sandstorm tried this, and they got very very far.

I think it was ingenious, but the sysadmin in me was very skeptical about the HA/scaling story (there was none). I guess it's fine for self-hosting for yourself&friends, but there should be an easy way forward - after all, Kubernetes commoditised clustering (I mean - once you're past setup).

Where can I follow your efforts? You mention KubeSail in your profile, but that seems developer-oriented?

We (Sandstorm) had an HA/scaling story, which we called "Blackrock"[0], and it was (and still is) the basis for the Sandstorm Oasis hosting service. I actually think we spent far too much time on that part too early. It turns out we could have operated Oasis by running regular single-machine Sandstorm on a beefy (though not that beefy) VM and been fine -- honestly, we would have had fewer outages and better performance. We probably should have spent our time making the platform work better for developers and end users instead, and built out the scalability later when customers really demanded it...

[0] https://github.com/sandstorm-io/blackrock

Oh, I've somehow totally missed this - sorry!

> We probably should have spent our time making the platform work better for developers and end users instead [...].

I keep a number of folders / bookmarks / stickies, where I save all the quotes like this one, so that I can then stumble upon them randomly. This is the single most important thing to keep in mind when building any kind of a platform. Developers particularly, are often the most important users of your product, since they will over time help drive the momentum and keep on bringing value.

Thank you for this reminder.

KubeSail is the company, yes - and developer oriented only in the sense that it’s not “Enterprise” - the idea is to make cloud native accessible to everyone - to make hosting your own apps fun and easy and cheap again! We have a long way to go and haven’t launched anywhere yet, but you can join the mailing list or just signup now for a free cluster namespace :)

"What we need is a cloud platform that allows people to deploy code to computers on their home network, or any cloud, or any data center"

Ease of deployment is one thing, but managing and troubleshooting these deployments is another. Ordinary people generally don't want to do any of these things.

There are some people who have both the technical aptitude and desire to do these things, but they'll likely always be a small minority. An even smaller minority will be willing to do these things long-term without getting paid.

Mastodon is dead because Americans can't spell it.

Reply on news.ycombinator.com