Why Custom Attributes in .NET give me Nightmares
Summary
The post argues that .NET custom attributes rely on a bloated blob-based storage for arguments, making enum values and type references costly to resolve at runtime. It highlights how extensive type resolution across assemblies, type forwarders, and multiple core libraries inflates attribute blobs and complicates parsing. The author contrasts this with a more efficient token-based metadata approach and offers reflections on tooling and potential improvements.