If thats how its gonna be there better be a very easy way for folks to stand it up and register it(similar to an iphone/android app install/register), and it begs a few other challenges:
So if my computer is turned off what happens to emails sent my way? If there is no host running the technology to recieve the message at that time does the client sending them email get an error and the message sits in like a local outbox until the other person comes online and they are both online to facilitate the send(is there like a local persistent send polling then)? Its a design that kinda works but is not really how email in clear net has worked(with 3rd party central servers) nor is it great because both parties have to be online at the same time to facilitate pending messages. #serverlessproblems I suppose.
I don’t think it’s suitable either. It adds complication to what is already quite complex and would only discourage possible users. In fact the biggest problem, of the network, will be gaining enough users to reach critical mass. Complicating the entry of new users would be a huge mistake.
Two tier coin isn’t implemented by most Blockchains, you could risk some sort of black market and scam trade… You could see adverts up for an account secret and password that has these new coins.
Etherium does have a separate gas coin doesn’t it… Probably similar to what you are trying to do here, but I can at least buy more gas.
What about issuing a 30 year certificate? More that enough time for a human life span, and short enough to recycle domain names?
On 30 years, most probably if anyone cares enough about the domain name they are probably a large corporation … who could probably afford to keep the domain by paying a network calculated network fee.
If the fee isn’t paid, the next person to buy the domain gets it? This would be a way around the network fee, but what multinational wants to rock the boat and have to transfer all their content to another account?
Only problem I see here, is the desire for information for ever? What if a really useful data resource disapears? If the safe address expires and doesn’t point to original pages? We would need some sort of waybackmachine service built in to the network so we can browse the content of previous owners?
With no network clock… how to we measure 30 years.
… I just wanted to mention that in the dev forum after some discussion agreement seems to be possible that a good solution would be to enable name resolution as plugin for the browser which would
Take the burdon of discussing/programming different variations off maidsafe the company
Enables new ideas that might work very well without going through hoops and convincing the whole community (or forking and maintaining a whole own version of the browser)
Let’s people choose the name resolver that fits their needs best (for example someone can have a whole own universe of names for a project where it is of need/benefit of they can just use a petname system with priority for family and friends +then default to the public system so they can make automatic use of their address book without app developers needing to care about public name exhaustion)
I don’t see why the network should know about naming systems at all …
For example right now if you type in ‘safe://antifragile’ the browser does a sha256(‘antifragile’) requests the mutable data with that xor name and if it exists looks where it points => that’s the website to display
… The network doesn’t know about name systems and just knows data and manages that… Very elegant imho and not introducing additional complexity at the lowest level
Maybe not possible, but what about you choose a domain name but you don’t get it right away. It comes on a global ‘auction’ list and everybody has a certain period to bid. The highest bid gets the domain name.
Because then people have to wait (how long???) for their mail address and cannot transact with another via mail till that is finished. Who is to administer the jerk who for one hundred safecoin takes all the new names just for kicks.
The time to wait for another to bid would have to be significant for it to be any use for that which its purposed for. A week, or do you think you need a month to allow enough exposure for all interested parties to see the new auction come up.
Should be possible as service on app level if name resolution is a plugin (probably not the network benefits in this case but still possible - and the applicant can use the name until end of the auction period)
I was more thinking about a week and maybe only for shorter words. I was hoping this could be automatic without administrating (or a minimum). At leas you know, if you checked weekly, what domain names are to be sold.
I’m with the others here that Maidsafe should preregister the big domains. If some idiot claims a big company name and puts crap on it because he doesn’t like the company, then you have a false start at day one with the network. Also the companies we don’t like should have a fair chance to get their domain here and use it. Otherwise adoption by the mass will be hard.
Yes really, really important and should have special attention. An idea I had was to reuse the bitcoin address format so you could perform small airdrops to existing bitcoin holders, just enough to play around a little with the network for a small upload or domain registration. See this topic:
Very interesting. Creates lots of possibilities, pros and cons as you say.
One thing I’m interested in is if names really can be non transferable? Making this hard or expensive is something that could be very useful, but I’m not sure it can be done.
For example, names are owned by accounts and I think accounts will have to be transferable, and your example does rely on this (at least so non farmers can obtain accounts created using fresh Safecoin). If I can transfer the owner account, surely I can transfer a name this way? So far I’ve not been able to imagine a way to avoid this.
A side point - I don’t think we can assume that people won’t rent names - many people do this now, even though it is often/usually a bad idea, same with email etc - but especially if we make it hard or expensive to create them.
Name rental may or may not be a bad thing, but I suspect it will happen regardless of fresh Safecoin (I mentioned this possibility just now in the name resolution discussion).
Anyway, it’s great to have such original thinking to chew over, particularly in this area.
I hope we don’t shoot it down too early! [Edit after reading the replies… ] Ahh.
Maybe someone could add a ‘name service’-plugin/app where you get e.g. safe://cn.yourwebsite.co.uk for the Xor address you give, when you can prove that you are the owner of the site yourwebsite.co.uk. This with a similar (one time) procedure to get an https certificate (like certbot from Let’s Encrypt) for your website.
I meant a separate naming system, done by a ‘central’ entity (app/service) on top of the Safe Network. So better to call it cn://yourwebsite.co.uk then (cn: clear net) instead of safe://cn…, where the system translates your cn://yourwebsite.co.uk to the corresponding Xor address. Or something like that.
Maybe that is also not possible. Certainly against the principles of being completely decentralized, but you can’t prevent people to come eventually up with solutions like that if the completely decentralized naming solutions are insufficient for some.
Edit: This cn:// naming system isn’t aware of domains of course. But you can’t omit it if you want to be able to port for example both www.yourwebsite.co.uk and www.yourwebsite.com.
To ensure perpetual accessibility to perpetually stored data I would argue for having more than one kind of sdns. Both centralized and distributed ones. Let the best one win.
Edit: Including all variants, paid, unpaid, etc.
