I was just thinking about the challenges of the impending testnet launch with vaults-from-home (yay).
It occurs to me that without the economic incentive in place to be a reliable node and with many just wanting to ‘test it out’ there could be a lot of, well I want to say churn, but I’m computer illiterate so I’m not sure if that’s the correct terminology. Basically there will be a lot of very unreliable nodes until there is a good reason for there to be more reliable ones (competition for safecoins).
How big does maidsafe’s contribution have to be to mitigate this slightly unusual situation? Do we actually want ten’s of thousands of people suddenly getting interested and testing it out for 20 minutes? Or, whilst we are pre safecoin, should we really just try to focus on providing reliable resources to the network ourselves rather than encouraging everyone to play with it (to make it a bit more realistic)?
Apologies if these seem like silly questions, I’m just trying to work out if Maidsafe’s early contribution to farming will start with the launch of vaults-from-home and whether the scale of it will really mitigate the unrealistic situation of relatively few reliable connections while everyone plays with it? It is naive for me to think that providing my own rubbish little 2mb upload speeds reliably will make any difference to anything, or should I actually be quite diligent about leaving my pc running even during the testnet stages? Does every little help with just a few hundred or thousand testers?
Good points I think but hard to give reliable answers to - though MaidSafe might be able to give more based on… their knowing a lot more than me
I think that while they maintain one stable and one test network is not a problem, and that when the incentives are added it may give us extra confidence in certain respects (though if course with that comes the incentive to attack as well).
I think it’s important to provide a slow node! I certainly will, the resource proof check will gauge what task you will undertake. You and I will likely just be relaying messages and that helps with security. I’m sure we’ll be serving up data too just a lot less likely to win the farming attempt. There’s a lot to learn but we shouldn’t shy away as there will be a lot of underresourced nodes in the wild so we shouldn’t take it easy on the network as we wouldn’t be doing the end product any favors.
I see your point about churn and people being all Willy nilly without incentive but it’s still something to learn from. It could provide an edge case for not having high enough incentives in the balancing algorithms. Just my take on it
Not sure how many nodes will be needed for a reliable test? I could run several nodes on different machines for a longer period of time without a problem. My internet connection is pretty good so I don’t think running multiple nodes would be an issue bandwidth wise. Would be good to get some advise on this from MaidSafe once “at home vaults” launch.
I think incentives for people to run nodes should be considered only if not enough people are willing to run one just to help out. I think there’s quite a few people with investments in MAID or MaidSafe (via BTTF) that would be willing to help out by running (a) node(s) to further the development of the network.
I would not worry. I think having no stable nodes they would just fire up a number of stable VPS to put a percentage in the node network… much like they done before but to a lower level.
Yes, I suppose that’s true. Given the small scale at the start I guess a lot of us will have a hidden economic incentive to leave our pc’s on all day and dedicate a reasonable amount of space and bandwidth to the network, even before safecoins are an issue. Not to mention the ‘enthusiasm for the tech’ incentive and desire to see it succeed.
It’s good to hear that it would be that easy if the network starts creaking.
I guess I have my answer then, it’s worth providing a reliable node myself (for testing more than resource requirements), and it is also probably fine to encourage lots of random people to test it out even though they’re likely to just join and try it out for a few hours.
Thanks folks I feel like I have a better picture of how it might play out now.
