BTS: .P2P .DAC - Decentralized DNS using the BitShares blockchain!

[kenCode]

@mdw
This very detailed post by @hadrian is nice, it mentions the leasing thing too:
https://bitsharestalk.org/index.php/topic,9346.0.html
 
Squatters get dealt with. This is one of the nicest writeups I've read so far.
This entire thread and everyone's ideas will be taken into consideration for the DDNSWG roadmap.

[hadrian]

I posted a topic - "Ideas for an alternative DNS model"- a while back.
It didn't get any comments, so it's probably crap, but someone may find something of value hidden in there?

[Pheonike]

The idea of a retainer fee on registering is good. You have to lock away a certain amount for a set period of time before it is released minus the normal registration fee. If a rights holder challenges and win, you lose the retainer.

[kenCode]

Yes, I am Mike Ward. One of many, and I hope to get my name registered before others do! I missed mikeward.com by only a few weeks back in the 1990s, and the registrant of the same name has never let it expire

I didn't want to make too big a deal out of that point about squatting, except to suggest that whatever you consider it to be (and definitions vary) there are easy ways to mitigate it to some extent. Main point was supposed to be that choosing the anti-squatting mechanism as the feature to distinguish this product from others does not provide incentive to the people registering the names. Again, people have long complained that all the .COMs are taken, then they proceed to register one anyway.

Unique features like "instant registration" will make this more competitive IMHO. Better anti-squatting measures, not so much.

There are plenty of good ideas about how to mitigate the big land grab that takes place if you offer all-you-can-reg domains for dirt cheap. Registry operators in the existing system are a good model to look at. They typically hold back the best names, and auction them off later, after the value has (hopefully) appreciated. People taking registered service marks like "Apple" is a slightly different issue, and is traditionally dealt with by offering pre-launch sunrise periods, where trademark holders can get first rights to register, as well as a binding arbitration process to challenge "bad faith" registrations. They work somewhat, but not so great, for various reasons.

Holding back some names from general registration is an easy way to avoid most issues. Specifically for dealing with bulk registrations (like registering loads of common surnames)

• diminishing registration prices over an extended launch period help curb excess registrations,
• higher renewal fees discourage holding large numbers of names over time waiting for high resale prices
• requiring a certain amount of assets to be deposited or kept in an unspendable state for each domain registered might help (?)
• perhaps only the first "premium" domain name registration per user account is cheap (where premium is defined in some arbitrary way like length of string < 5 characters and it's not on reserved list) For strings beyond that length, no limit, but more premium strings cost more and more.

And so on. There are probably as many ways to disincentivize bulk registrations as there are users on the forum. Some are easy to implement and fairly effective.

Aah cool, thanx Mike
MikeWard.com - seized by yet another squatter. You feel the pain more than I do then, having someone actually take your name.
 
I think the people that the squatters will annoy the most will be the larger brands and names, the ones especially that want to display their name on some form of advertising, like a tv commercial. Their domain has to be short and sweet. Like ddns:web.mdw
 
If registered on the BitShares blockchain, I can forsee names being snagged in 10 seconds flat (our block times). In the OP, I mentioned too that there could be a consensus mech involved where the LESS people that vote for you, the more expensive your reg will be. The land grab you mentioned. Domains, userid's, electrical outlets, light switches, aircraft and vehicles of all kinds. They should be routable and nameable too IMO. I think the "bad faith" you mention could be mitigated with that after-the-buy consensus vote.
 
Registering loads of names, that's a squatter. I can't see any reason why one person would want to register more than a few "TLD's", ie:
ddns:web.mdw
ddns:mail.mdw
ddns:land.mdw.home.oven (see this stuff above)
fact is, mdw becomes your new TLD for everything. Then it just child-nodes down from there.
 
I like your fee structure above. It becomes increasingly expensive for each new "TLD" the user wants to register.
 
What do you think of my list of names above? Toast, Vitalik, You, Oleg... Once we get our list of names picked out (kinda like the list used in the movie "Gone in 60 seconds) then we just need to schedule the g+ hangout, write the roadmap, get the project funded and knock it out.
 
Thanx for joining us Mike!
  -ken

[mdw]

Yes, I am Mike Ward. One of many, and I hope to get my name registered before others do! I missed mikeward.com by only a few weeks back in the 1990s, and the registrant of the same name has never let it expire

I didn't want to make too big a deal out of that point about squatting, except to suggest that whatever you consider it to be (and definitions vary) there are easy ways to mitigate it to some extent. Main point was supposed to be that choosing the anti-squatting mechanism as the feature to distinguish this product from others does not provide incentive to the people registering the names. Again, people have long complained that all the .COMs are taken, then they proceed to register one anyway.

Unique features like "instant registration" will make this more competitive IMHO. Better anti-squatting measures, not so much.

There are plenty of good ideas about how to mitigate the big land grab that takes place if you offer all-you-can-reg domains for dirt cheap. Registry operators in the existing system are a good model to look at. They typically hold back the best names, and auction them off later, after the value has (hopefully) appreciated. People taking registered service marks like "Apple" is a slightly different issue, and is traditionally dealt with by offering pre-launch sunrise periods, where trademark holders can get first rights to register, as well as a binding arbitration process to challenge "bad faith" registrations. They work somewhat, but not so great, for various reasons.

Holding back some names from general registration is an easy way to avoid most issues. Specifically for dealing with bulk registrations (like registering loads of common surnames)

• diminishing registration prices over an extended launch period help curb excess registrations,
• higher renewal fees discourage holding large numbers of names over time waiting for high resale prices
• requiring a certain amount of assets to be deposited or kept in an unspendable state for each domain registered might help (?)
• perhaps only the first "premium" domain name registration per user account is cheap (where premium is defined in some arbitrary way like length of string < 5 characters and it's not on reserved list) For strings beyond that length, no limit, but more premium strings cost more and more.

And so on. There are probably as many ways to disincentivize bulk registrations as there are users on the forum. Some are easy to implement and fairly effective.

[kenCode]

@.yoshi
It's less expensive to code in the squatstop(first!) now. Not only that, but once mass-adoption begins, then we are barraged with people saying how we fucked it up, we didn't even try to stop squatters, etc. The labor to defend a short sighted decision is more intensive than if we had just took an extra day or week to figure it out now and code it in from the beginning.
 
@mdw
Here in Germany, businesses rely on their family name. Your name is everything here. In America, brands are used more than anything. Both can be seen as one in the same however, if you are Apple, then you want ddns:web.apple right? If squatters can snag it, how much more strife is created then? A lot.
 
Code it correctly the first time.
I'm hoping Mike Ward will chime in here, I got another email from him just now so will try to get his thoughts publicized in here as well.
 
Thank you everyone for your input on this! Our collaboration on this will be world changing

[pc]

but I also believe that the emphasis on a complicated auction-style system to deter squatting was ill-advised. Not having a predictable annual cost is a negative for the business community,

[.yoshi]

solid insights mdw. I think you're absolutely right;  no need to waste time and resources at this point over-architecting the product over potential non-issues. even if domain-squatting does become a problem it can probably be dealt with down the road.

build the mvp leveraging our existing strengths and improve it as we go along. just my .02 bts

[mdw]

OK my $0.02 - for a Bitshares DNS offering, it's important to think up front about what the distinguishing features should be. My personal opinion is that the emphasis should be on the blockchain and DPoS. Domain Name resolution is the 800 pound gorilla, but it's a universal problem, and a proprietary solution from one chain will not satisfy the general need for all blockchain-based DNS. Be a different DDNS, highlight your inherent strengths through creative feature sets.

I would assert that since you have here a fundamentally different chain than Namecoin's "traditional-style" POW chain that should be the focus. Namecoin is very well served by the SHA-256 POW system they use, due to the unique merge-mining situation they have with Bitcoin. But making a weaker copy of that is, well, weak. Bitshares is a very fast, efficient chain with a different consensus model and should leverage that.

You have the potential here, for example, to get a new domain name registration up and resolving before even the first block is added in a 10-minute block system! Contrast that with Namecoin, where not only does each block take a while, but new registration requires two consecutive operations, so realistically you can't expect your name to resolve for people an hour after registering it.

Feel free to flame me to death for this next part, but I also believe that the emphasis on a complicated auction-style system to deter squatting was ill-advised. Not having a predictable annual cost is a negative for the business community, and I believe that "squatting", however you define it, is not an actual problem.

I've been involved in the domain industry for years, and people have complained since the late 90's that .COM was so completely squatted. Yet they continue tor register them, instead of going for shorter .WHATEVER domains. People find names, and the business has continued to boom with xxx,xxx,xxx .COM registrations. It continues to grow even this year, with hundreds of new TLDs introduced.

I don't expect many to agree about squatting being a fictitious problem, but my point is that designing feature sets around those capabilities that are unique to Bitshares chain is more appealing, no? You might prefer to think of it in terms of being a competitive advantage.

Near-instant registrations, as I alluded to earlier, would seem to be one such interesting differentiator. Ever look at Dash's (was Darkcoin) InstantX payments, where it happens almost immediately? This, for domain registrations (and subsequent website setup) is a killer feature that's very tough to match by any would-be competitors.

[kenCode]

Thanks Ken.
Maybe that'll include some offering from this group - I hope so. So count me in for whatever, I'll be glad to do what I can. - Mike

Thank you mdw, your vote of confidence helps my sanity, sincerely. Sometimes it just feels like people have their priorities all screwed up, like I'm the "chicken with his head cut off" as fuzzy put it. Our work here will be utilized by generations of people!
 
I've skyped with Oleg at Emercoin as well, he has some very intelligent things to contribute.

[mdw]

I think you should contact EMC Dev (emercoin.com) to cooperate for decentralized DNS.

Oleg from Emercoin is one smart and friendly guy, I interviewed him for a podcast a while back.
https://soundcloud.com/mightbemike/episode-01-oleg-khovayko - just past 15 minute mark

I'll ask him to check out this thread.

[mdw]

Thanks Ken. I think your thread here proves that interest in blockchain-based DNS is as strong as ever. There are many small missteps on the path, but the basic underlying need remains. There have been, and continue to be compelling reasons for decentralizing this core Internet infrastructure. Round up @toast and the gang and let's have some fun.

My personal belief is that Namecoin is going to really come into its own soon, after such a long period of stagnation. I also believe that other, different, blockchain-based DNS will come to market in quick succession, and mainstream folks will finally start to realize they have real choices. Maybe that'll include some offering from this group - I hope so. So count me in for whatever, I'll be glad to do what I can. - Mike

[bitcoin42]

Wow. This sounds amazing. All I was able to think of with this topic was this http://www.new-nations.net/

[kenCode]

Mike Ward, meet the BitShares community.
BitSharers, meet Mike Ward.
 
Mike Ward is a Technologist, Coder, and Writer for The CoinTelegraph amongst others. He's been following my comments regarding DDNS.
 
Mike I know you have some questions regarding DDNS and may like to discuss your past work in it as well.
So, without further ado, Mike you have the floor...

[kenCode]

PLEASE UPVOTE THIS, IT BRINGS EXPOSURE & LEGITIMACY:
http://cointelegraph.com/news/114149/adopting-the-chain-how-tesla-could-protect-its-website-and-its-cars#comment-2008708292