RFC - Decentralised Naming System II - continuous auction (by Seneca)

Features
61

Good points @Seneca. I think we are going to struggle to understand the behaviours of either system on SAFE.

Risks Of An Untried Scheme

Your scheme sets up possibilities for attacks that don’t yet exist, and commercial practices that could well be more damaging than squatting.

I think it’s impossible to predict how often domains will be taken over.

One way to approach these unknowns this is to ask if a similar approach has been used in any other area. If so, what happens, if not, why not?

I can’t think of any situations where once an asset has been secured, anyone can take it off you without your consent just by paying more for it than you did. It’s like an auction that never ends, where you don’t actually have control over the asset, you just get to hold it until someone else, someone richer than you, wants to take it off you.

The nearest I can think of is a hostile takeover, but here it is only the minority stake that is “bought out”. A majority of the owners have to have sold out first. So it’s not the same, but might give some clues as to how this might work out.

Can anyone think of a situation that uses @Seneca’s approach?

There are risks in devising a new scheme, so I think the aim needs to be evaluated more fully to see if change is needed, and if the risks are worthwhile.

Access For Everyone

As I already mentioned, it seems out of step with trying to make SAFE accessible to all. Your proposal gives wealthy people the ability to take from poorer people, simply because they can afford it, and no compensation. The loser is left with nothing!

Make Squatting Harder

If instead we just make squatting very hard, we can have all the benefits without these risks.

The obvious way to do this is to limit the number of names per account (as already planned), and to make it hard to create or maintain multiple accounts. This would be useful for other reasons (making it hard to attack the network) so extra points if we achieve it.

Both are hard problems, over to you @divine :wink:

3 Likes
What is your favorite name for a SafeCoin analogy to BTC’s Satoshi? (reboot) divs = irvines, troons, and ayrs?
62

An alternative would be to sell the other party the domain instead too. It could end up being cheaper for the bidder and the money would go to the current owner, rather than the network.

An aggressive take over would likely be costly and may not be well received by the public/customers either. Much like in business, it is usually better to trade, rather than fight.

63

@seneca The more I ponder your idea, the more I like it! With the rough edges taken off, it is a truly excellent and distributed solution, IMO!

64

I think it is hard to say what will happen or how often. Remember, the thief doesn’t have to use the domain, they may just wish to disrupt a competitor, or an organization they dislike, to enforce copyright, to censor a news site etc, and with complete anonimity.

1 Like
65

I really don´t see how this could ever be an appropriate solution. I´m still not really sure if I got ir right because it seems so counterproductive to me, but reading the comments over and over again, it appears to me that @Seneca proposes to institutionalize the possibility to highjack domains based on economic power. It´s clear who will be the winner: the one to which the additional expenses are peanuts. Decentralization certainly causes chaos where centralized systems provide clear structures, but that doesn´t mean that the whole game needs to be turned into an All-pay-auction.

Expiring domains is certainly worth to think about, but in general the idea that I presented above would allow the network to decentrally develop different DNS which serve as both: a discentive to massively grab domains and a potential to react to exploits.

1 Like
66

I agree, but the cheapest option is usually the one chosen. The exchange could still be anonymous too.

Moreover, given that the bidder may not get the domain for, say, a month after becoming the highest bidder - there will be ample time to make adjustments (redirects, new name, etc).

67

There are a near infinite number of alternative names. If someone rich is desperate to get your valuable name, sell it to them, then rebuild with another name. Who is the loser here?

1 Like
68

You keep mentioning measures the loser can take (adjustments) but I don’t see what they can do to mitigate losing their domain. Can you explain what you think they can do, and what this does for cars such as:

  • a user typing in their domain for the first time (or on a new device), which they found on your expensively produced brochures at a show, or one of the postcards you spent hours designing, printing and handing out, or was printed in that article you sweat blood to get written about you etc etc
  • visits from other users who have bookmarked their website
  • visits from websites, search engines etc that have links to the old name
  • embedded links to resources provided by another site whose domain has been stolen (embedded scripts) when you had no idea this could happen, so used the short name, not the private key name

I can see that a properly designed browser might be made to handle the change of a bookmark whose domain has moved, but with a significant delay no? Surely this has to be checked for every URL accessed, which is gong to take time.

I don’t understand how the others could be handled within the grace period so please elaborate.

It would also mean that all branding, on and offline would be invalidated. Fixing this, even just on the website, could be very costly both in time and money, even for a very small business. And for most businesses they’d have to get their web designer to do it, and he’s not necessarily going to be able to do this at short notice.

I think we’ve barely touched on the problems domain theft creates, and that this is being glossed over somewhat.

2 Likes
69

Another way to look at my proposal is to consider typing in a name in your address bar a name search query. The ranked list is the search result list, where a higher rank puts you at the top of the list.

The browser-side option to default to the top one is similar to the “I’m feeling lucky” functionality found in Google Search. The browser-side option to default to the most recently visited one can be seen as an auto-complete of your address bar.

Since that functionality is browser-side, the communitiy can figure out what should be default in browsers. Defaulting to the highest ranked one seems to be the main problem here, so maybe that’s not a good idea.

2 Likes
70

Thus ignores the impact on someone who has lost their domain. Their choice is to enter and win a bidding war, completely rebrand at great expense (and risk the same happening again), or slip down this list.

Describing it like Google search might make it seem familiar and therefore more acceptable, but it isn’t search. This is an unexpected pop-up placed between any user requesting a specific website and instead getting a complicated list from which to choose.

If this happens frequently it its a bad user experience.

If it happens rarely, it is unfamiliar and confusing, also a bad user experience.

For user, and for website owner, this to me seems far worse than the problem you are trying to solve.

1 Like
71

i really don’t see why bidding wars should happen … what do you gain from buying an occupied domain …?

and yes - loosing an important domain is horrible … I used a private owned link management website for more than 10 years … i had EVERYTHING bookmarked there (literaly Everything)
Within these 10 years i lost contact to the guy who knew the other one who hosted this site … i still don’t know what happened … i believe the server and all the backups or he died … because there was no info-letter in advance and no one after … suddenly the site was down … this was really a huge shock for me …
but still … these things happen … and if a website owner forgets his login and can’t change the content of his/her website anymore we have nearly the same result as if the site was highjacked … the only difference that in one case we do have lost one domain forever …

ps: yes of course it is pure speculation … we don’t know what will happen and maybe i’m wrong … i just don’t see it being such a huge deal …

72

It doesn’t have to be complicated, the list can contain metadata, including a short summary for example. We can supplement this system in all sorts of ways.

It happens when you visit a domain for the first time, so it depends on the type of user how often you get them.

I consider a DNS where major organizations can’t be found under their own name to be far more problematic than the fact that multiple identities use the same name, and that the user is asked to pick one when first looking up the name.

3 Likes
73

Poorer people and rich people will lost there name because stupid people that try to get all the name already. The seneca’s approach let everyone free to use the same name. I would add to it that the addon remember which address you have visited first and stay on it until the user itself change the address. And you can bookmark that address (Including the random ID). Adding to that a special url to see the list of the same url and there value should be available for everyone (Like safe:urllist.www.microcash.com). And if it’s an organization not lucrative, everyone can give some coins as a gift to keep the site at higher priority. Everyone can pay to keep a site they like at high. Everyone will still try to get good name for there site relevant as do what they do.

4 Likes
74

okay - now we really are talking about coexistence of different websites with the same name :smiley:

that feels strange but is pretty awesome to be honest

3 Likes
75

That’s fine. But isn’t what you’re suggesting a search engine, not a DNS? Can it be DNS is not needed at all because a search engine + user bookmarks are sufficient? The bookmarks would point to .onion addresses and .onion addresses will be used in hyperlinks and web service endpoints?

Lastly if what you have described is a search engine then it wouldn’t be right that only one search engine can exist, right? The user should have a choice?

3 Likes
76

The arguments in favour of this change are using situations and usage patterns chosen to suit the desired outcome. They are not addressing the issues raised. This isn’t a rational approach to design, which needs to consider all stake holders, usage patterns, and potential behaviours within this new system.

This is a call for a radical change to a fundamental and proven system. What we implement will bear on the entire project, so to fly, this needs to be worked through thoroughly and evaluated objectively. I don’t see the points I’m raising being taken seriously and addressed systematically on this thread, just set aside with what seem to me weak or inapplicable rebuttals. So I’ll leave my comments at this for now.

I think this is a radical and exciting idea, certainly worth exploring, but IMO it has a long way to go to become a serious proposition.

EDIT: It does appear that you are describing a search engine where ranking is determined by the highest bidder. There’s no place for that on the existing web, because there’s a DNS.

1 Like
77

The dns system, anyone can implement a new one in the safe network and have it’s one rule set. The current one is the default one and can easily be superseded be a custom one. It can have multiple dns system. The most popular one will win. With the current design of the safe_network as I understand everything it’s possible and can’t be prevented. But I’m not talked about share older here but donation for a popular site. But it can be done with a different dns system through the safe network.

EDIT: I was not talking with the existing one but in the safe network itself.
That can’t be prevented with a custom one. People will adopt the one where the site they seek is the more relevant one.

OUPS: It should have a function to be able to burn safe coin for a particular purpose for this to work. That would be a great feature.

1 Like
78

There is a near infinite numbers of letters and numbers…yeah (btw. with emphasis on “near”). You may not share the idea of valuable strings of letters and unvaluable due to a given social context. Most people do and pricing of domains are a reliable empirical proof.

The proposal wasn´t about selling, it was about rich or simply annoying people being able to hijack domains.

Everyone. Domains need to provide some sort of stability, otherwise they are senseless. Sure, users can stick with address books, back to the stone age, weehee. Besides, there are alternatives that leave people with stability and measures to deal with exploitation

79

I think that’s a valid perspective. Just add that those addresses do have an official (non-unique) name.

I’m sorry you feel that way, I do take them seriously. We just seem to differ in perspective on what impact the different drawbacks of the different options would have. It’s a matter of anticipations, which are always subjective. Funny, because I feel domain squatting is underestimated in this thread. :slight_smile:

Well, that’s what we’re doing in this thread right? It’s not like I’m calling for immediate implementation.

For what it’s worth, I find the Delegated DNS concept interesting as well.

DNS is actually one big address book. SAFE itself runs on one big address book. I don’t see how a digital address book is related to the stone age…

3 Likes
80

yes - but a smart businessman might buy it instead of trying to steal it and risking bad press +confusion …
And yes if bad people like to waste money on annoying people that is a bad thing when they succeed … but the ones with a very low budget in mind for a domain (like me) might simply not care that much because there are many other cool domains out there … and you really would need some money to mess with a real company …

ok how can we manage the domains without some central organisation involved and without loosing valuable addresses long-term?