Announcing Community Satellite Pilot Program

If you are a business and you provide IT services it may make sense to run you own satellite to provide your customers with cloud storage. Let’s say your business is in germany. As a german business you are way more trustworthy in your local area than an american business. Also your satellite would be in the local language. Also as a local business you know how to advertise the service in local language better than an international business, since you know what are the quirks of the local population.

Community satellites are also multipliers of the network. Each provider has his own bubble to address.

I have thought about something like that too: If one satellite goes out of business, make the data accessible through another satellite. But while technically this could probably be solved, I have no I have no idea how the non technical stuff could be organized in a proper way.

Well the question is, if the satellite operator has enough data to keep the SNOs entertained. Either his own data or data from his customers. Plus he has to manage the SNOs like required legal and organizational matters (e.g. GDPR) tax issues, he has to do the payment on his own (cryptocurrencies??? Nobody really wants that) with all the fear that the SNOs will just leave and destroy his or his customers data without a warning or to be able to hold anyone accountable for. I am German and I cannot imagine a serious business in Germany who would do that, either running a satellite or becoming a customer of one unless maybe they are operated by some really well known provider like Deutsche Telekom. But even then I doubt it.

1 Like

The satellite operator doesn’t really have to keep the SNOs entertained. There are other satellites as well. The sum of all will keep them entertained.

Yes that has to be managed of course. I don’t really see an issue there. If the satellite has customers the provider can also just spin up his own nodes to keep his customers data save.

I’m german too and I believe the opposite. It’s all about marketing.

By the way Deutsche Telekom offers Telekom Cloud which is Microsoft Azure running in Telekom Datacenters managed by Microsoft employees from US remotely. Basically a black box setup.

Yes exactly, customers with a satellite also have a contract with that provider, this handover has to be legally managed. I would believe if a satellite shuts down it will take care about those thinks by himself and remove all the data or advise his customers to migrate to another satellite. There would be no need for a builtin way.

I see it that every satellite operator has to make sure that the SNOs that have chosen his satellite will remain with him and not leave. If it is not worth it for them they will not add his satellite or go away.

That’s good but I don’t see it. As I have said it is beyond my imagination for serious businesses.

And I don’t think Deutsche Telekom would seriously consider running and managing their own satellites and SNOs. But maybe they would pay Storj for it.

It doesn’t cost anything for the SNO to follow a satellite. If it’s a paid satellite and the SNOs get paid what reasons would there be to unfollow?

The Satellite doesn’t have much traffic. Does it hurt? No! There are others with traffic. If there comes traffic and it’s paid properly you would miss out on that if you’re not following.

The only reason to unfollow a satellite would be if it doesn’t pay.

1 Like

Or you are full and you have only cold data from that satellite, so no earnings from egress.

I’d say that depends on the satellite. They could impose any requirements on SNOs as they wish. And if there is a better satellite for me, that either costs me less or earns me more, then as SNO I know what data I would delete.

1 Like

I wonder how many SNOs will really hunt for every cent. You also have to take into account that the amount of space has to be filled again first. Earning more on a satellite can only happen if there is also more traffic.

I just provided another possible reason :slight_smile:

And am not advocating for it. But simple calculation would say how much more one could earn from starting a new node vs keeping really (c)old one. So if I want to upgrade to bigger disks, I might just start a new node instead of continuing on that one, cause I suspect that any ingress to it would mostly be cold data again anyway.

And if identities were possible to be changed, one might even give away/sell such old node, lol. That actually seems quite useful for the network, doesn’t it? :thinking: Now enlighten me why not :laughing:

That’s one of the questions a serious satellite operator should ask himself. I don’t know it but I’d say if as a satellite operator you simply assume that anyone will follow you regardless who you are and what you do under almost any circumstances, then that’s your first mistake. So as satellite operator not only do you have to know your customers, you have to know your SNOs as well and keep an eye on them. How hard it is not to keep everything well adjusted shows with the announced changes of the economic model where we still have no official word from Storj.

And if some XY company from whatever country which language you don’t speak or can’t even read and you have never heard of is going to start a satellite would you immediately add it as SNO? I would not.

So don’t get me wrong, I really would like to see the biggest IT companies spinning up their satellites and even better every DAX and DOW company too and let them move Exabytes of data onto our disks. I just don’t think it will (easily) happen. It should be carefully and thoroughly assessed what would be required to get this adopted.

Why not? Honestly, I do not care if the satellite is pay for data and bandwidth, since I never know, what’s their customers store on my node anyway.

From a technical and monetization aspect, I agree. But there are definitely companies/organizations who I would not want to support if they ran a satellite. Even if it would be very profitable for me. Now they could have an account on any satellite… I don’t get a choice in that, but I do get to choose if they ran a satellite and I’d want to know a little bit about the company/organization behind it if I’m going to offer storage services to them.
I’m not going to name examples as I don’t want to start a political/moral discussion, but I’m sure we can all think of some examples.

On the other side, I might be willing to take a loss for companies/organizations I’d want to support. The internet archive comes to mind for example.

3 Likes

I see one problem here. If a community satellite changes the source code to use something different than the Storj-provided encrypted erasure codes and suddenly my node hosts unencrypted data with undesired, illegal content, I may become liable for hosting that content.

The satellite doesn’t get to choose that. Uplink encrypts and erasure codes the data.

That’s under assumption the satellite doesn’t invent their own fork of uplink, telling their customers to use that one instead for that satellite. Again, not something I can control.

Hate to break it to you, but that’s already possible on the current network. Uplink is open source as well. This can already be done.

1 Like

But right now the satellite will detect these cases when auditing the pieces. If uplink doesn’t upload with the Storj-approved erasure coding scheme, then the repair process will mangle all data and the audits will start to shine bright red.

First… If someone intentionally does something like this, why would they care if that happens? Second, there’s nothing stopping people from crafting their data in such a way that the erasure coding is valid so it won’t run into this issue.

But I feel like we’re getting a little side tracked here.

1 Like

that’s a different story. However, I believe that satellite should not be added to the semi-trusted list of community satellites.
And, by the way, it’s not controlled by a satellite operator anyway, this is a client-side code…

1 Like