H.3 Workspace
A clean machine exposes the real local contract. If the first successful change depends on an engineer's full history, hidden SDK, pre-generated files, and a wrapper whose fallback nobody knows, the repository has documented a happy machine rather than a supported workspace.
From Clone To Useful Evidence
The path begins with H.3.1 Git Workspace. Partial clone, sparse checkout, shallow history, indexes, and maintenance change different costs, so choose the combination from the operations the developer must still perform. A fast first checkout that makes blame, generation, or on-demand fetch unpredictable has moved cost rather than removed it.1
H.3.2 Bootstrap then selects repository-owned tools, credentials, configuration, SDKs, and the smallest first operation. Its checkpoints should separate network, authentication, tool, workspace, generator, and build-graph failure rather than end in one opaque “setup failed.”
With that state established, H.3.3 Local Feedback walks through edit, index, generate, build, test, and review preparation for a real team scope. The important comparison is not perfect local/CI identity; it is whether documented differences let local evidence predict the relevant review lane.
Keep A Repair Path
Break the normal wrapper in the walkthrough. H.3.4 Degraded Workflow names the raw tools, bounded bypasses, owner lookup, and restoration condition that remain. The emergency path should preserve evidence and authority rather than begin with destructive cleanup or permanent policy bypass.
Use Git Workspace for transfer, history, path, or status symptoms; Bootstrap for clean-machine failures; Local Feedback for slow or misleading daily loops; and Degraded Workflow when the paved route itself cannot perform the repair.
A coherent workspace carries one developer from source acquisition through review-predictive evidence and still leaves a bounded repair path. Optimize the specific cold or warm operation, then verify that discovery, generators, IDEs, and recovery still work for the affected cohort.
Footnotes
-
Introducing Scalar: Git at Scale for Everyone — Git workspace maintenance and performance dimensions at repository scale ↩
Sections in this chapter · 4
Choose Git transfer, history, populated paths, and local maintenance techniques for the workflows being optimized.
Move from a clean machine to repository-owned tools and a coherent first successful operation without hidden workstation state.
Keep editing, indexing, building, testing, and generators coherent for the scopes developers actually use.
Preserve a bounded emergency path when wrappers, generators, indexes, or the build graph itself are broken.