doc: trim self-promotion and over-defensive phrasing; tighten §1, §5, §6

README.md

@@ -47,23 +47,17 @@ configs:
 
 # PowerZoo Dataset
 
-A curated, harmonised collection of **real-world power-system and data-centre traces**, plus a library of **canonical electrical-network case files** (transmission and distribution), used by the *PowerZoo* / *PowerZooJax* benchmark suite for reinforcement-learning research on power systems.
+Real-world power-system and data-centre time-series, canonical grid topologies, and JSON manifests linking the two — packaged for reinforcement-learning and forecasting research with the PowerZoo / PowerZooJax benchmark code (released separately).
 
-This release is the data companion submitted to *******. Source code for environments, baselines and reproducibility scripts is released separately under a ***; this dataset repository is self-contained and can be loaded standalone.
-
-> *****.** Author names, affiliations and contact information are intentionally omitted during the review period, in compliance with ***.
+Submitted to the ***. Author information is omitted during ***.
 
 ---
 
-## 1. Why this dataset?
-
-Power-system RL benchmarks have historically suffered from three gaps:
-
-1. **Synthetic load curves** that do not capture the multi-scale structure of real demand (sub-hour ramping, weekly seasonality, event-driven spikes).
-2. **Inconsistent grid topologies** scattered across `MATPOWER`, `pandapower`, OpenDSS, IEEE PES test feeders, etc., each with its own conventions.
-3. **No canonical link** between high-fidelity demand traces and a fixed set of grid topologies, making cross-paper comparison difficult.
+## 1. What's inside
 
-PowerZoo Dataset addresses all three by shipping (a) cleaned, parquet-encoded traces from public regulators and cloud providers under a single schema, (b) a unified Python representation of standard transmission and distribution test cases, and (c) lightweight JSON manifests that wire the two together for reproducible experiment configuration.
+1. Eleven parquet time-series files ingested from public regulator and cloud-provider releases (GB / AU electricity load, generation by fuel, day-ahead forecasts, market mid-prices; Alibaba / Azure / Google data-centre utilisation).
+2. Fourteen electrical-network case files (Python classes with bus / branch / generator / load tables) covering transmission systems from 5 to 2383 buses and distribution systems from 33 to 533 buses.
+3. Eleven JSON manifests that map each parquet's raw columns to a shared canonical schema (e.g. `OPERATIONAL_DEMAND` → `load.actual_mw`), so traces from different sources compose cleanly in one experiment.
 
 ---
 
@@ -237,21 +231,24 @@ This repository **redistributes derivative datasets** built from publicly access
 | Google cluster data 2019              | Google                                 | [https://github.com/google/cluster-data](https://github.com/google/cluster-data)                                                                                                                     |
 
 
-The packaging artefacts authored for this benchmark (the manifest schema, case-file Python representations, schema harmonisation logic) are intended to be released under a permissive open-source licence at camera-ready time; the underlying parquet data inherits the upstream licence in every case. **Do not assume any single licence applies to the data themselves.** When in doubt, redownload from the upstream URL above and treat that copy as authoritative.
+The packaging artefacts authored for this benchmark (the manifest schema, case-file Python representations, schema harmonisation logic) are intended to be released under a permissive open-source licence at camera-ready time; the underlying parquet data inherits the upstream licence in every case. The canonical copy is the upstream URL above.
 
-No personally identifiable information is present in any trace; demand and generation series are aggregated at substation, regional or national level.
+All series are aggregated at substation, regional, or national level; no PII.
 
 ---
 
 ## 6. Intended use & limitations
 
-- **Intended use.** Reinforcement-learning research, load / generation forecasting, market-clearing simulation, OPF benchmarking, distribution-system control, data-centre power-shaping research, demand-response studies.
-- **Out-of-scope.** Short-term operational decisions for real grids, retail tariff design, settlement of real markets. The forecasts and prices included are historical educational artefacts, not signals for live trading.
-- **Known limitations.**
-  - Each manifest's `date_range` is the inclusive UTC date span of the matching parquet file at release time. Upstream sources (especially AEMO and GB BMRS) continue to publish new intervals; redownloading from the source URL will yield more data than this snapshot.
-  - The Ausgrid trace is imputed; the imputation procedure is not documented in this repository.
-  - The data-centre traces are stored at 5-minute (300 s) resolution. The upstream releases provide additional fields and finer resolutions; this repository ships only the columns mapped by each manifest.
-  - The grid-case files use parameters consistent with the cited source systems but are not guaranteed to be byte-identical to any specific MATPOWER / IEEE PES / PEGASE release; they should not be used for compliance studies of real networks.
+**Intended use.** Reinforcement-learning research, load / generation forecasting, OPF benchmarking, distribution-system control, data-centre power-shaping, demand-response studies.
+
+**Limitations.**
+- Static snapshot at release time; upstream sources continue to publish. Redownload from the URLs in §5 for live data.
+- Geographic scope: Great Britain and Australia only.
+- The Ausgrid trace is imputed by the upstream publisher; the imputation procedure is not documented here.
+- Data-centre traces ship only the columns each manifest maps, at 5-minute resolution; upstream releases offer more fields at finer cadences.
+- Grid-case parameter values are consistent with the named source systems but are not byte-identical to any specific upstream release.
+
+Not intended for real-time grid control, retail tariff design, or settlement of real markets.
 
 ---