Author Topic: DNS design decisions: 50m shares, one TLD (.p2p), DNS "family parent"  (Read 10449 times)

0 Members and 1 Guest are viewing this topic.

Offline bitbadger

  • Full Member
  • ***
  • Posts: 95
    • View Profile
Sounds good to me.  Although I might have made different decisions, your reasoning is sound, and this is a decision that I can stand behind.  +5%
Pei5BrnEUqcCuUdffNZmBPL3rg6duj3vnU

Offline santaclause102

  • Hero Member
  • *****
  • Posts: 2486
    • View Profile
plus.p2p would stand out because of the number which makes it unique and would thus be great as a brand and symbol for the p2p movement and the technology used within BTS DNS...

Offline toast

  • Moderator
  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
Why only one TLD?

There is a pretty good discussion here you might want to read first: https://bitsharestalk.org/index.php?topic=3652.0

The short answer is, if you look at what distinct TLDs actually do in real life, it's to establish different *rulesets* for how the namespace is managed. The fact that it can be used to avoid name collisions is a *secondary effect* and an artifact of how traditional DNS is structured. AAPL.com and AAPL.org aren't just avoiding a collision - AAPL.org is utilizing the fact that it won't get sued on the .org namespace, while it would on the .com namespace. Sure you can make it so that two TLDs use the same ruleset and are on the same blockchain, but then the only thing that it accomplishes is to make an aesthetic difference between typing "yoursite-org / yoursite-com" and "yoursite.org / yoursite.com". One blockchain is one namespace, no matter how you slice it or mask it.

So the idea is, future DNS derivatives who want to have different rulesets for their namespace, or who want to prove me wrong and show how people are just willing to avoid collisions (would you register "yourname.dac" if "yourname.p2p" is registered and popular? maybe) can launch a DNS derivative a la BTS X.

So .key might have a ruleset that fits the supply/demand characteristic of having a namespace for public keys better than domainshares.

.intr could be the first namespace to support a full character set.

There might be another blockchain that handles migrating .com, .org, etc onto a blockchain.


Why .p2p?

Here were the good suggestion, IMO:
* .bts
* .dac
* .blk
* .we
* .p2p
* .key
* .dom

I don't want to do .bts or .dac because I find them slightly more awkward to say than .p2p, and I have a really strong aversion to people trying to sell me stuff (I know that's not what's happening, but I can't help but think .bts -> "BTS STANDS FOR BUY BITSHARES(tm)"). Furthermore I think .p2p does a much better job of marketing the advantages of a blockchain-based TLD than the others - people will instantly recognize what's up versus just thinking it's another TLD.

.key shouldn't be wasted on anything other than a namespace designed specifically for keys.

.we is risky because all two-letter TLDs are reserved by ICANN for country codes.

.blk is ok but worse than .p2p, fewer people know what a blockchain is

.dom is ok, maybe the first derivative with different namespace rules can grab it


50,000,000 shares:
It's just aesthetics. Open to alternatives.
« Last Edit: March 21, 2014, 02:40:12 am by toast »
Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.