06 / THE DEPENDENCY MAP

Web3 applications.Map the whole system.

The chain is one part of the experience. Account for the files, names, endpoints, publishing access, and maintenance processes around it.

Name what the visitor depends on

Trace a page from the address a visitor opens to the files, images, application data, and status information it needs. Identify the operator and failure behavior of each dependency. This inventory is more useful than calling the entire project decentralized without saying which components that description covers.

Separate content identity from availability

For IPFS publishing, plan retention rather than assuming that one successful upload establishes long-term availability. The IPFS documentation explains pinning as protection against garbage collection on a node. Record who retains the release, how it is retrieved, and which public entry point identifies the intended version.

Keep publishing authority visible

A team may distribute frontend files while retaining centralized control over the domain or release pointer. Record who can make those changes and how they are reviewed. This is a design boundary to understand, not something a hosting label automatically resolves. Keep an auditable release record and a tested recovery procedure.

Make partial failures understandable

The documentation might load while an application-data service is unavailable. Explain which feature lacks current evidence rather than displaying a generic claim that the network has failed. Keep conceptual pages readable without wallet access, and make recovery guidance available through a route that does not depend on the failed action.

IPFS distinguishes persistence, pinning, and permanence and discusses the responsibilities involved in retaining content.

Read the IPFS persistence guide

Questions to settle early

Is IPFS the same as guaranteed permanent hosting?

No. Plan persistent retention and verify retrieval. The delivery network does not remove the need for someone to retain the files and operate the surrounding publishing workflow.

Can ordinary hosting be part of a Web3 application?

Yes. Describe its role accurately in the dependency map and assess the resulting trust and availability assumptions rather than hiding them behind the project label.