How to make sure Cardano does not end up as an AWS Blockchain?


#1

I am pretty sure this has been already covered, nevertheless I will still put the questions and please point me to the coverage.

Following the Shelly realease when we will have maybe around 1000+ stakepools around the world creating blocks, how do we make sure these stakepools are not running their nodes purely & solely on AWS infrastructure?

I understand EOS ended up in this terrible scenario, but what will prevent Cardano from it?

With full decentralization we should never have more then a certain percentage of nodes running in the cloud of a particular infrastructure provider. It might even be a bit harder to express the limitation, as we should avoid any cartel scenario, where a few parties control the infra (like we have for BTC today).

How is this gonna be accurately & safely managed?

Will stake pools have to acquire a license during registration and confirm their infrastructure setup, having somehow the license tied to their infrastructure so they won’t be able to change it without required to apply for a new license? I would even think a policy that they must run on-premise and no clould operations are allowed is reasonable.

What happens if we have 30% of pools operated via AWS and another 30% via Azure or whatever and at some point in the future Amazon acquires the other cloud provider?

How is it possible to sustainably ensure the inrastructure level of decentralization of Cardano, without any manual / human supervision, BUT complete trusted automation?

Or we do expect the Cardano Foundation to continuously monitor & ensure this aspect of the decentralization? Or perhaps we will have the Cardano Community in their best interest governing this decentralization quality via voting for Stakepool licenses (revoke and grant scenarios)? (BTW I think for security reasons the Cardano Forum should also move on the Cardano Blockchain at some point when the Sustainability features are released to the Community.)

We just don’t want another AWS Blockchain, having Jeff even the slightest chance to pull the plug.

I am pretty sure this has been deeply thought about, but did’t find anything specific yet.


#2

It won’t, it’s a job for the community to either run proper nodes, or vote for pools that run proper nodes and require transparency.

They won’t. It’s an open participation system. Anyone can run a node or a pool on any infrastructure they want to - whether they get enough vote to actually produce anything or not is the job for the community.

Then we will have 60% of nodes running on Amazon servers.

It’s impossible.

We don’t.

Kinda. There are no licenses. If someone runs a pool on AWS and people vote for it with they delegation - this pool gets to create blocks. It’s equal and open for everyone.

Vote for pools that provide their own infrastructure :fist:


#3

I don’t think there is a real issue here as you can easily count the next few years as the baby step years. Lets say Cardano is 90% run on cloud services over the next 5 years while Cardano gets built out, develops and matures. The good thing is you have 5 years of server management and know how, and switching the underlying hardware would be the easy part.

When the time comes that the Cardano ecosystem values more robust and decentralized server topography, incentives can be introduced to make that happen.


#4

So we will end up being an AWS Blockchain, trusting that Amazon will never pull the plug or attacks the infrastructure anyhow.

As AWS might be the cheapest way of running the DPoS infrastructure? I don’t see pools incentivized otherwise. Proof of Work was a bit different in this particular context as you could tackle the extreme hashing power with dedicated hardware better then with a cloud service.

I don’t think the community will be able to verify or will have the motivation to investigate what infrastructure their stake pool will be using, rather make the decision on other aspects (marketing, reputation, etc.).

Well Amazon is already running multi trillion USD businesses on AWS so why wouldn’t they run Cardano. They are too big & too profitable to have a financal incentive to interfere. But will this be the case in 50 years from now? It doesn’t feel like the complete decentralization & trustless System we have been striving for. On the other side I do not see an automation solving this scenario (sure you could initially reject nodes with Amazon addresses), but only the Community or delegated members of the Community may effectively guard this quality.

I somehow feel there is a need for “Guardians of Cardano” voted by the Community (following the principles of liquid deomocracy) with the mandate of overlooking certain essential qualities (such as decentralized infrastructure mentioned above) of the Cardano Ecosystem and being compensated in ADA from the Treasury. This incentivezed, well-meaning and expert external viewpoint by human intelligence may be an essential part not only for further developing the platform (voting on features & improvements), but also continuously auditing, supervising & ensuring that it’s qualities are met. This basically means we should not only vote on CAPEX items / developments for the future roadmap, but also on on-going OPEX “maintenance” services such as the Cardano Guardians. Is this something aligned with the strategy or too far from the expected operations of the Sustainability capability of Cardano?


#5

Where I am trying to push the conversation is that I believe a fully decentralized cryptocurrency such as Cardano requires a fully decentralized dedicated organization from the members of the Community to manage the sustainability over the next decades. Not sure whether Cardano Foundation is meant to transform into such an organization one day?

Such an organization would be incentivized via regular compensation from the Treasury (OPEX cost), it would be elected via liquid democracy, have a clear structure, agenda & mandate, goals & targets and corresponding measures & metrics. Maybe even some empowerment to influence / configure the Platform in some extent via multiple approval cycles (including the Community Approval itself).

Members of such an organization would take certain roles such as Guardian (monitor & supervise some quality of Cardano such as infrastructure decentralization), Priest (evangelist of Cardano bringing the message to new communities), Doctor (providing quality support / aid for any problem someone might have with Cardano), Fighter (fighting the trolls, fake news, hackers or any enemy of Cardano), Scientist (monitoring the platform qualities and suggesting configuration updates for optimization), Strategist (proposing “next best” items to the Cardano roadmap along the Product & Adoption Vision in the context of the continuously changing environment), Inspector (reviewing the operation & performance of members of the Cardano Organization with the previous roles), etc.

Members would have their reputation to manage within the Cardano community providing regular transparent, verifiable & regular updates of their operations and justification of the value they bring to the Ecosystem.

The challenge would be to make the Organization efficient, incorruptible and transparent to the Community.


#6

I’m thinking of starting a staking pool. If I do it will be run locally on hardware I own. I’m glad to know that could be a selling point.


#7

I like that idea.


#9

literally the blockchains will have a superior human power, known as the base to manage that decentralization, I hope that with the minimum vote of the community, and that the requirements and laws are fulfilled by all the nodes and foundation.


#10

It should be trivial to provide some network information such as IP range and geolocation in a ‘details’ or ‘properties’ window when selecting a candidate staking pool.

This may reduce the security posture of the staking node but is like already available / easily identifiable through information gathering techniques.