Dev Diary #6: Update July 28

Dev Diary #6: Update Since May 8

Since May 8, ProjectXL has moved from architectural convergence into operational alpha.

The last developer diary focused on the journey state machine, the four-region shell model, and the work required to make the application feel like one guided product instead of a collection of separate windows. That work is still central, but the story has changed. The architecture is now starting to pay rent.

The major shift in this period is that core planning and execution capability is now operating on top of the journey-driven shell. Integrated planning, financial operations, schedule interaction, readiness feedback, snapshot management, performance hardening, licensing, and in-product learning have all advanced from design direction into working product surfaces.

ProjectXL is now live in alpha testing. That does not mean every planned assistive or teaching layer is finished, but it does mean the product is no longer just proving out the shell architecture. The alpha surface now covers a broad operational workflow, with onboarding and learning layers coming online around it.


Highlights

  • Integrated planning workspace capability matured across schedule, cost, labor, work package timing, and activity visibility

  • Cost overview, labor planning, work package timing, and cross-lens completion state moved onto a more consistent planning workspace model

  • Financial workspace functionality became operational, including charge mapping, actual cost import, cost reconciliation, contract performance, funding versus actuals projection, and earned value surfacing

  • Schedule interaction deepened with more mature Gantt and network linking behavior, drag-driven updates, event and non-labor date movement, and project start milestone handling

  • Scope completion behavior was pushed toward parity across Gantt, schedule planning, and work package timing so the same planning state is reflected consistently across lenses

  • The readiness and feedback model was substantially hardened through layered validation, manifest-driven producer routing, readiness code separation, chip severity coloring, and unified feedback routing

  • Region A through Region D moved closer to a governed evidence pipeline rather than a set of parallel local feedback channels

  • Chip drill-down behavior was tightened so workspace feedback is more honest about what is ready, what needs attention, and what is blocking progress

  • Large-workbook performance became a first-class deliverable, with incremental recompute, save-flow optimization, cold-boot logging cleanup, fingerprint-based compute, and Parquet-based storage and transport work

  • Snapshot, compare, restore, and recovery paths became more coherent, with improved semantics and cleaner handling of projection caches, workbook ownership, and stale-state risks

  • Multi-window behavior was hardened so planning, financial, schedule, and execution-state surfaces can operate with more predictable shared context

  • Licensing and startup flows were strengthened for live alpha distribution, including native JWT licensing, activation hardening, retry behavior, and more durable task-pane preferences

  • The welcome and project-start experience was redesigned to better support first-time users entering the product

  • Orientation, tutorial, and in-product learning moved from concept into shipped product surfaces, including tutorial shell work, orientation content bundling, a local concepts window, first-launch gating, and cross-launch behavior between orientation and tutorial experiences

  • Documentation, build notes, and release tracking continued alongside the code so the alpha surface is tied back to explicit product and design intent

Overall, this was the period where the new architecture stopped being the headline by itself and started enabling the product.

The journey state machine, shared shell regions, validation routing, and manifest-driven state model still matter. But they matter now because they support real planning work: building and adjusting schedules, connecting cost and labor data, importing actuals, reconciling financial performance, reviewing readiness, preserving snapshots, and helping new users understand what the system is asking them to do.

The product is now much closer to the intended ProjectXL experience: a guided planning and execution environment where schedule, cost, work package structure, validation feedback, and user orientation operate as one connected workflow.