I thought it was mentioned by the team that anyone can copy the code and create their own network?
i think he means that the code of safenetwork can be forked and be edited to create other networks. in my idea, I am not talking about parallel networks that are separate as in the forks but multiple networks that are interconnected that allow multiple gameplans and if one fails the other continue
But you still have to specify what you mean. A fork could be an improvement that still uses the one network. OR it could mean a new network
ye i mean a new network
people say it would be better to have a free market, other people that a central algo should be there, why not test them all in parallel? I argue there should be some kind of a test system that can test lots of gameplans together
edit: but in this thread I am proposing a safe network that is a multinetwork system that can have networks fail and others thrive and that way we can add a new network that has a new idea and networks that dont go well should be either updated or let die
Of course that is possible but will not have the data or coins obviously. Data is stored on peoples disks in vaults. Not in a database that everyone copies.
And parallel networks will suffer from the problems I mentioned above as well as not being the first mover
In addition to all the problems I have already mentioned, you need a separate coin for each network and how are users to work out which coin to buy, how are cross network payments to be done. This is a real mess. It will not work.
Yes that for sure is true
what does that mean, I dont understand?
If the safe network has been live for 3 months with 10PB of data and people earned coin then a parallel network starts with nothing, no coins and no data
imagine a safecoin and each network having each own âcoinâ where a user buys safecoin with fiat OR farm the âcoinsâ and in a market can get safecoin or even directly exchange with other âcoinsâ. all this can be very simple as in âI want to do somethin on the other network (e.g. access a file in the other network) then a simple system does everything automaticly and you can do that on the other network.â
imagine a bank that is in eu with euros and a client wants to buy something in america with dollars, he buys the item and the bank pays in dollars having taken your euros and did all the necessary steps to convert it and take it from your account and give it to the sellers account
ok but in my idea, there is a way of moving things. if you are a vault you can just switch network and then the network automaticly adjusts.
if a new idea comes that is revolutionary even if some vaults leave from the 10PB network and go to the new one, the data will adjust to the new situation and will be transfered if the system thinks it is good
edit: if though the network has the logic of if you dont like the network leave, then if people leave and dont come back or massivly people are like safe is bad and disable all together 1tb of storage vaults it would be chaotic.
with a choice of moving you get people to vote on a better network.
networks then have a system connecting each others with balancing out the costs of exchanges gets etc so the best network as in best implementation gets to help safe network stay alive in bad days
i guess this is very possible to happen, considering the greed that surrounds us. There will surely be some devs who will fork the code make their own coin and then make interoperability between safe and the new network.
I dont think that is possible⌠safe network is safe, if one version is at the most pcs and another version tries to participate it cant
Iâm sorry there is much basic misunderstanding here and I am off to bed. I have tried to outline the problems without being too technical and it seems that is not enough. It seems a superficial idea of yours and you need to go back and put together my answers to get a fuller picture of the issues and maybe then you will get a better idea of why it is not feasible.
That is why Maidsafe will be running plenty of testnets before they hone in on the solution that will be used. Splitting the network from the start is an absolute mess in simple terms including your idea of many coins with one safecoin to exchange them with.
Please just spend more than 5 seconds and think of the amount of programming that would be needed
- just to implement all the different vault implementations
- then add to that the programming to have them talk to each other
- then the programming to allow a user on one network to talk to a vault on another network
- and that is assuming this is contained in the one vault. (which of course is harder)
- then the programming to have multiple coins and an exchange system with the universal safecoin
- then the documentation
- then the UX implementations
- then apps that have to implement different apis because the underlying vault implementation is different
- then the debugging time is multiplied by the number of implementations and additionally the time to sort out why one implementation acts different to another and try to fix that
- then the education of the public to understand the different networks
- then the confusion of the public as someone tries to explain why there are so many.
- then the battles as different groups try to get users to use the network they are using so the data is on the one network and their version of the coin is worth more.
- Then how does rewarding work when I GET data from another network
- Then what happens when one network dies and the users go **** off I am going back to google youtube or back to facebook or whatever.
- and I could go on but I am tired and off to sleep very soon.
Your proposal, assuming it could even be done would Explode the time taken to get to even beta let alone release.
5 Likes
to answer to the complexity of the system the user would not feel a thing, the people that got the vaults would just have half of the storage divided for all networks and half for their choise of network given that those people are the people in here discussing which is the best implementation.
I am really scared about one network that if it fails all data is automaticly lost and it needs to be reimplemented and rebooted
are you not scared that bitcoin will fail? everything is possible but the probability is very low for bitcoin as for safenetwork should be same.
Who on earth do you expect to pay for all the extra coding?
Do you really want to delay us for another 10 years?
I like how youâre already counting in AGI into your calculation. Without a super intelligent AI this would take more than 100 years and we would still just end up with one system (if itâs interconnected; so whatâs the point of all this).
1 Like
Good point. Well presented.
Down the road, there might be value in having multiple network mirrors that launch at the same time, with the same data, and compete in a mutually beneficial way. Consider it a metaphor for two hemispheres of a giant safe brain.
The ârightâ network might have one set of pricing/control or optimization algorithms, variations on a theme, etc. whereas the âleftâ network has another set of parameters. Every PUT from the client code gets sent to both network âmirrorsâ at the same time. Every GET is requested from both network mirrors at the same time.
The reason for doing this is simple redundancy, and extra checks on Parsec consistency, possible security benefits, and a possibility for self-optimization. This only works well if there is a decent level of communication between a section and itâs mirror, though⌠parallel subnetworks instead of multiple independent networks
1 Like