# LabDAO Tech Team Meeting 18 May 2022 ## CLI * Estuary * `file push` and `file pull` use this as a service on the CLI for IPFS pinning * Important esp when running compute on remote machines, need to pin entire subdirectories * CID is default for locating content * Abstraction above CID which provides a virtual filesystem (think user directories for each wallet address) * Instead of pushing a single file, we can push entire directories to IPFS (ex. `some_folder/subdirectory/file_a.json`) * Articulating namespaces allows us to create public (or private if they're encrypted) datasets * Whitelisted providers --> Known datasets * Can fetch content using the namespaces via the CLI * Eventually, we can append local data to these datasets * [Nebulus](https://nebulus.dev/) - "IPFS-less IPFS" useful for staging prior to IPFS pinning, useful playground * MVP for application developers * CLI tests ongoing * Application developers won't be able to hook up their apps to the CLI yet. Alpha release will be `revcomp` for now, but iterations for adding new applications should be quick * Add application code to template repo * Add API endpoint for available apps so we can run applications' code * orbit db ### Q&A * stability of https calls? still * changes to directories can be in sync with a daemon directories - *cookiecutter data science* integration with makefile possible ## Contracts * Rename `client` and `provider` in our smart contracts * proposal: `user` & `service provider` linguistic styleguide - lily (will be on docs.labdao) * `greenlist` instead of `whitelist` * `client` will be used for the software enabling interaction with the protocol * Modularizing which blockchain users are operating on. We're developing on Polygon for now, but need submodules for whichever chain we're using (impacts contract addresses, subgraph endpoints, RPC endpoints, etc) ## Graph Team * Met yesterday, more to come soon :D ## Laboratory services * ligand detection NMR * input * small molecule structure * amino acid sequence of protein * parameters * output ## ML inference * collaboration with TalentDAO (?)