The UX problems is mainly faced around network economics and the complexity that comes with that.

• now, there is an extra layer of economics that need to be done in order for your things to work. you need to go through the documentation to actually understand how things work
  • the problems happen both at onboarding and as active users
  • once you are on the network, you don’t know how much curation you need to actually get picked up
• indexers, curators, delegators
• investor perspective
  • delegation is another issue in deciding what
• this problem is more for developers.
• if you think from a higher perspective, the navigating economics of participants on the network are a wider problem for all people.
  • every participant has doubts about other participants on the network
• most people (developers) have a problem with the subgraph gateway
  • this leads them to having non-consistent data returned from the same query
• two causes for attrition of subgraph developers
  • DIPs solves curation
  • people leave because of the costs. curation is an added expense that they do not realize why they need.
  • queries are quite expensive for some dApps.
• query costs can be reduced with better patterns
  • but it seems to primarily be a byproduct of the protocol setting this value at $40
• since the past year, most focus has been on retention
  • they were 1-2B queries per day
  • now we are ~500M queries per month
  • retention is something the team is totally focused on
• past sunrise, there can be an increased focus on acquisition
• we are in a very cost sensitive industry
• it is hard to change from public good to
• solve the broader protocol issue
  • people are primarily used to bad ux
  • the bottleneck that people have faced is the query costs
• the query cost is more of a protocol economic decision
  • query costs would also help costs
• we are at the best UX today for most of the
• indexers: allocation optimization
  • auto-allocation, auto-reallocation
  • for users, they have done the best they could do
  • subgraphs are tooling, they do not have network events
    • postman to apollo: no one cares
    • networks, when changed, cause problems
  • tooling it
• network effects: was a fan of agentc. this was their initial push into web3 with AI
  • if the top 10 dApps could have AI attached to it, that would be nice
• from what he understands, running an indexer is a hard task
  • the graph network solves as enabling the network effect of compute
• there are components that need to be optimized to enable this type of composability
  • the gateway is not well positioned to enable this
  • a lack of modularity does not enable this
  • if we open this up and have a more modular framework, then you can have these people come in and utilize this infrastructure