docs: add rroadmap.md documentation file

README.md

@@ -8,7 +8,7 @@ pinned: false
 ---
 
 
-**README** • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
+**README** • [Roadmap](roadmap.md) • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
 # 📊 Traders-Lab — Accumulated Financial Time Series
 
 Traders-Lab publishes **public financial time series datasets** with a deliberate focus on **long-term accumulation**, **structural consistency**, and **historical depth**, rather than short-term freshness.
@@ -102,4 +102,4 @@ If you're training AI models or building quant pipelines with the TroveLedger da
 Thank you for helping grow the trove!  
 
 ---
-**README** • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
+**README** • [Roadmap](roadmap.md) • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)

history.md

@@ -1,5 +1,5 @@
 
-[README](/Traders-Lab) • **a brief history** • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
+[README](/Traders-Lab) • [Roadmap](roadmap.md) • **a brief history** • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
 # How TroveLedger came to be
 ### A brief history
 Over the past year, TroveLedger has gradually taken shape — not as a sudden idea, but as the result of a series of practical needs becoming increasingly clear.
@@ -49,4 +49,4 @@ TroveLedger has since continued to grow through regular updates and the steady a
 The ledger remains open.  
 
 ---
-[README](/Traders-Lab) • **a brief history** • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
+[README](/Traders-Lab) • [Roadmap](roadmap.md) • **a brief history** • [TroveLedger](/datasets/Traders-Lab/TroveLedger)

roadmap.md

@@ -0,0 +1,33 @@
+
+[README](/Traders-Lab) • **Roadmap** • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)
+## Roadmap
+
+At its core, Traders-lab's `TroveLedger` remains focused on a specific task: accumulating high-quality historical market data over long periods of time. The emphasis is on consistency, completeness, and reproducibility — properties that are essential for serious analysis and for training machine learning models on real market behavior.
+
+### What remains stable
+
+While details evolve, several principles remain constant. TroveLedger favors durability over novelty, structure over breadth, and long-term usefulness over short-term completeness. It is built to preserve market history as it actually unfolded - consistently and without interpretation.
+### Dataset expansion
+
+The primary form of growth continues to be the gradual addition of new assets. Stocks are introduced in structured batches, typically aligned with established indices, rather than as isolated symbols. This approach leverages external professional curation and avoids the randomness that characterized earlier Traders-Lab preliminary datasets.
+
+Beyond equities, other asset classes are actively considered. Indices themselves, foreign exchange pairs, commodities, and selected cryptocurrencies each come with their own structural challenges. They are evaluated individually, with no guarantee that all of them will be included, and no fixed order in which they will appear. Expansion follows feasibility and relevance, not completeness for its own sake.
+
+### Metadata and structure
+
+While OHLCV data forms the backbone of the dataset, additional metadata is an area of ongoing consideration. Certain information — such as dividends, splits, or earnings events — is inherently discrete and time-bound, making it suitable for structured inclusion.
+
+Other forms of fundamental data change slowly and unevenly. Capturing them at high frequency would add volume without adding insight. Determining which metadata is useful, how often it should be updated, and how it should be stored remains an open design question, approached conservatively to avoid unnecessary complexity.
+
+### Tooling and openness
+
+TroveLedger is supported by a growing collection of internal tools used for data ingestion, validation, and maintenance. Some of these tools may eventually be published, but only where doing so is clearly useful and sustainable. Preparing code for public release requires additional effort, and that effort should be justified by real demand.
+
+As the dataset matures, attention naturally shifts toward how it is explored and consumed. Rather than developing a bespoke interface, TroveLedger is designed to integrate into existing analytical environments.
+
+In this context, OpenBB has emerged as a particularly compelling point of interest. Its modular, provider-based architecture aligns well with TroveLedger’s role as a long-lived, curated source of historical market data. Exploring whether TroveLedger can function as a data provider within OpenBB or how OpenBB might leverage it as a persistent historical backbone is an active line of thought. The exact form such an integration might take remains open, but the conceptual fit is strong enough to warrant deeper investigation.
+
+### Cadence and rhythm
+
+Updates are intentionally released at a steady but non-mechanical pace. Regularity matters, but TroveLedger is not designed around a rigid release calendar; instead, updates are published when they are ready and meaningful. This rhythm reflects the long-term nature of the dataset rather than short-term attention cycles.
+[README](/Traders-Lab) • **Roadmap** • [a brief history](history.md) • [TroveLedger](/datasets/Traders-Lab/TroveLedger)