Devcon workshop notes

Notes from pre-devcon ENS workshop


I’m not the only one who took notes, so please whoever else did add them here as well. I apologize for any incompleteness in the following.

ENS Notes

Re-Branding

  • ENS seems like just Ethereum
  • Option 1
    • Lean into it
  • Option 2
    • ENS but change E
    • Everything Name Service
  • Option 3
    • Completely different name
    • Unrelated to blockchain?
    • Triangle (Zooko’s triangle), ZNS (Zilliqua), TNS (Tron?)
    • DNS (decentralized name service)
    • UNS (universal name service)
  • Votes
    • lots of ppl want to keep ENS
    • very few want to keep Ethereum as the E
    • many want Everything or recursive ENS

Universal Login

  • hadriencroubois.com/enslogin
  • let ens determine what your preferred is
  • rate limit
    • CAPTCHA as earner for enough gas
    • watch an ad
    • doesn’t scale when you pay for the users yourself
    • BAT
    • Shipl
    • coinhive = 11 days for $0.05
  • New standard spearheaded by authereum

Multisig

  • new schemes possible to redue overhead on mutliisig users
  • either optimistic voting like moloch (only vote when it’s contentious)
  • should key holders be paid?
  • move power away incrementally
  • Who are part of the multisig
    • removed from the website…
    • give them kudos? or should they be semi-anonymous?
  • multi-sig group is same for .eth TLD as for root
  • will th multi-sig exist if ENS expands?
    • would like to reduce but won’t be able to replace without getting rid of upgradeability
    • or move to DAO
  • True Names Ltd, Singapore
    • grant proposals from true names go to multi-sig people
  • Multisig capabilities don’t need to be bundled
  • Hopeful funding to improve for UX of multisig prticipants
  • key list is a closed list, should maybe turn it into a public list or google group
  • what about metalevels with js library and browser / dapp developers?
  • Summary
    • make time delays
    • separate out roles and rights
      • auction rates seem like easiest to distribute rights similar to MakerDAO
      • negative feedback loop, registration cycle should modify the price to register
        • easy to game
        • ppl might be inelastic in their preferences

—Lunch Break—

Resolver Innovation

  • Topics
    • Aliases
    • Default resolvers
    • Wildcard subdomains
  • Wildcards
    • Currently
      • any custom resolver can be made with custom functionality
      • Alias would only need a little code to reproduce equivalent of CNAME
    • pre-generated key pair can be used in tandem with the wildcard keypair in order to recover
    • can allocate names for free and keep track of them off chain and transfer them later or let a user use them in the mean time
    • generic xpub is blocked by a sha512 precompile
    • Questions
      • do we change any of the interfaces?
        • the label won’t be present in a lot of situations but in a lot of situations it wont matter
      • want to include the label or risk changing every type of profile that already exists
    • improves privacy
    • still a risk of custody
    • could cals a separate resolver instead to make sure that the resolver supports recursion
      • makes it backwards compatible
    • pass a resolver as a parameter
    • Action Item: Start a discussion thread and pool ideas about how to implement it
  • Default Resolver
    • not a fallback system cause it might break if you finally set a resolver
    • when registering domains the default resolver and address should be set in the registration process
    • will need to update controller after short auctions
    • Requested features
      • set binary?
  • Aliases
    • brainstorm how and then just write it
    • mapping from name to alias, if there’s an alias to that and if not go to normal resolver
  • PKI
    • not to go to ICANN but when integrated w browsers they might have a lightnode running within and then there’s no gateway nodes in between
    • no intermediate certificate or rootchain

Using DAI or alternative currencies to pay for domains

  • dai could be converted in time to pay
  • ether is
    • unstable
    • you’d rather put ether in a CDP and use the dai to pay for the domain
  • how do you keep track of tokens allowed or not, adding removing, their price oracle
  • currently uses the maker price oracle to be ~$5
  • could be replaced with a feedback mechanism that isn’t gameable
  • encirca.com - ethsimple.com?
  • adding DAI is easier to do on the client side or a middle contract
  • ENS has less risk on chargebacks if they offer visa purchase
  • encirca charges extra fr the convenience and to cover their risk
  • adding contract in between would allow multi currencies as well as meta transactions
  • using ENS for keybase-like verifications
    • like twitter
  • text records
    • need standards for the different social profiles
    • minimize text on chain
    • just IPFS hashes
  • should ENS be more like whois or like identity resolver?

Does ENS compete w DNS?

  • there’s the name space and then there’s the infrastructure
  • don’t want to compete with the name space
  • want to provide alternative infrastructure
  • argent.xyz
  • .eth names is not ens

Multi-coin

Things you’d like to see ENS focus on

  • Governance of the funds
  • Batching transactions / accepting DAI
  • How to handle homograph attack?
    • IDNA rules
  • enforced best practices for normalizing text
  • login work
  • bully all the wallet operators to always have compatibility
    • multicoin was a request from wallets
    • “unofficial” repo of shame
  • verify the registration contract on etherscan
  • docs don’t mention controller
  • ens login / more wallets and usernames
  • ens logins / alias
  • batched request
  • default resolver
  • recursive resolution
  • anything that gets users using ENS
  • login work
  • adoption / integration
  • everything

Competitors

  • Handshake
    • replace the DNS root
    • managing and signing top level domains
  • overly complicated?
  • cash grab?
  • FIO, IOV, Handshake have their own blockchains
  • ENS advantage can interoperate with other contracts
  • ENS can piggyback on Ethereum infrastructure like metamask
  • Ethereum is highly secure, other chains will bootstrap and be low
  • Handshake is competing with ican in the name space (not the infrastructure space)
  • Working with ICANN is better than working against
  • IOV, idk
  • FIO is serious competitor, wallets and exchanges would be the block producers
    • issuing their own names
    • contacted them and said we don’t want to compete
    • they do a lot of other stuff besides naming
    • let ENS do naming and they’ll do other stuff
    • it’s own chain based on EOS
    • VC funded
  • ZNS, zilliqua name service basically a copycat
    • VC funded
    • directly competing and wants to win
  • Question about Brave integration
    • Branden Eich working on ENS and IPFS integration
  • Opera beta browser resolved ENS as IPFS
1 Like