The whole PublicID and service thing is still very confusing to me how this would look like
I have the following issues:
As a starting point, in case squatting becomes a thing, which I hope it won't be,
but if it does, amazon, facebook, microsoft, baidu, apple, pentagon, nsa, nasa, us, sex, eu, npc, sgcc, stategrid, sinopec, a, b, c, d, etc,
I'm going to jump on them as many and as soon as possible.
Every idea I get I'm thinking...
can't that be part of an aggregate site/app?
Like reddit and github, how would this look on MaidSafe?
I just don't know what names I want to give my public IDs.
I have at least two things in mind that I don’t think would not be necesarily part of an aggregate site.
One, in case it’s needed, a public ID for e-mail address.
Two, a site that I want to be about a self-constructed Esperandito, called Esperango for now, adding some words from Japanese anime and be closer to French instead of Spanish as a fun attempt to make a unified European language.
That needs a totally different URL in my opinion, because it’ll differ from other constructed languages due to potential politics and pages about culture.
I don’t want the second one to be under esperango.folatt.
What if I want it to be under esperango.language, language.eu and language.european all at the same time?
Is that too much?
Should I make a good single choice instead?
Three, my opinions on the EU and my political viewpoints maybe?
Can I delete Public IDs?
What if I want to use diacritics or non-roman letters?
This will probably be a thing, pretty simple first come first served
You can decide to do whatever you want, you want:
reddit.folatt, google.folatt and github.folatt it’s all possible
As far as I know, this will not be possible. More important is that you link a service (a website) to an publicID and if you don’t do that, it’s just a registered name in the SAFE Network universe.
If two people register the same publicID what happens. Say I registered as “Maidsafe” just to prove a point. What would happen? Or better yet look at a phone book sometime. There are reams of people with the last name. What if you ha a mess of people with the same last name. Even if only a fraction of them decide to make a family website there are bound to be some conflicts.
Also if Im understanding this correctly @folaht and if my experience with the testnets is any indication. You can’t have like “www.microsoft.com” or whatever. it goes "publicid.service.safenet or whatever the scheme is now. But the key here is publicid.service. So I could register the service microsoft.com and you could register the service microsoft.com and microsoft could register the service microsoft.com but it would all look different. It would look like blindsite.microsoft.com and folatt.microsoft.com and microsoft.microsoft.com (which honestly looks really weird).
Also iirc aren’t public ids supposed to have some kind of numbering scheme attached once mutable data or safecoin comes into play? So that everyone gets a unique public id no matter what they name it and so they can’t guess someone else’s public id via process of elimination?
In my perfect world, PublicIDs would be transferable. I don’t know if will be possible, but I would like to perhaps, transfer a PublicID from the account I created it in, to another account created later, or sell a business to somebody else, and transfer the PublicID associated with that business to them.
They’ll all indeed be able to register their own microsoft.com service as you explain, but in your example, the URLs would instead be microsoft.com.blindsite, microsoft.com.folatt and microsoft.com.microsoft, i.e. safe://<service name>.<public id>
so it’s more like the service is your domain name and your publicid acts as the suffix. So instead of .com, .ca, .us, .net, .org and so on you have .blindsite, .folatt, .yourpublicid. This would kind of make domain squatting useless.
How does the network combat publicID squatting tho? One could easily grab up every word and combination of words in the English (and any number of other languages) and then try to resell them to people. If I want the public ID “Bob” but there is already a “Bob” what happens?
Yes, and the same problems arise, and there have been long threads and much deep thinking on this forum trying to figure out better ways of doing this. I had another go on my own yesterday and concluded that, much as we don’t like some aspects of the current domain system, it does a pretty good job and is hard to improve on IMO.
I still couldn’t see a way of having the benefits of a domain like system without the obvious downsides (mainly squatting), and think the benefits are well worth it: memorable addresses for any online property or app.
Thinking about it lead me to think that if we don’t charge a rent for public IDs (as there is for domain names) it will make hoarding them worse.
What I realised from my thoughts yesterday is that the key qualities that make domains / public IDs so useful are that they are:
memorable
universal & unique (ie like a street address)
easily obtained
These qualities make them both useful and valuable.
Making them tradeable makes them easy to obtain and creates a market which enables squatting. It is very hard to limit any of the downsides significantly without messing up the benefits. Anyway, I don’t want to derail this topic, the place to discuss ways to do public IDs is here:
See also this post for links to six (yes six!) proposals including five RFCs in different ways to do public IDs:
For those unfamiliar, with ENS, the names go up for auction (which lasts 3 days and is blind). At the conclusion, the losers get their bid back less 0.5%, which is burned. The winner pays nothing and gets control of the name. In hindsight, the developers wish that the winner also had to pay a fee to make squatting less profitable.
The winner currently just has their bid locked up with the name for at least a year, and until they relinquish the name. It is possible or likely that in the future there will be renewal fees. The current implementation of ENS was like a temporary beta, with plans to move to a permanent implementation after learning lessons from this period.
The alternative is basically having multi layered names. That is there could be an infinite number of happybeing names registered. The downside is that you need another way to tell them apart so it becomes more complicated. You’d need to have some metadata to display to tell for example who registered the name etc, it would be very difficult to make it easy for regular users to tell a scamming website apart from where they want to go though. On Wikipedia if there’s multiple articles with the same name then all of them gets a topic in parantheses behind the name, that’s one way. Not sure if it is possible to make such a system of many of the same name with some kind of disambiguator work well, but I wouldn’t completely rule it out either.
Yeah I think we’re over thinking this. In real life we have several pieces of simple easily rememberable data to identify ourselves.
Given name.
Surname.
Birthplace.
Family lineage
Age
Current location
Contact Information etc
Bob of Someplace
Alice out of Mary by Bob
Mary#1 is 20yrs older than Mary#2 and Mary#1 is mother of Alice and mate of Bob of Someplace.
We don’t need to know all this but couldn’t a dynamic hash be determined based on one’s activity on the network? This hash would not be readable as something people could use to track activities. When you wanted to add someone you’d add both public ID and their hash.
