Build a charting tool, a market data site, or a quant terminal — however elegant the code, three questions remain: where does market data come from, can you display it to others, and how should you disclaim when you do. This article approaches from four angles — data rights and licensing, permissions for real-time quotes, the boundary between tools and "investment advice", and common compliance points for open-source projects — to give developers an actionable checklist.
⚠️ Risk Warning
This article is an objective compilation of public knowledge, for study and research only, and does not constitute legal advice. Data licensing terms, exchange API policies, and financial marketing rules change at any time; before launch, defer to data vendor contracts, the latest exchange terms, and licensed counsel's opinions.
1. The "Rights and Licensing" Problem of Market Data
1.1 Data Has Owners
Market data (quotes, candles, trade ticks, order book depth) usually comes from exchanges or data vendors, bound by contract terms, and in some jurisdictions may also implicate database rights. "Findable online" does not mean "free to use".
| Data Form | Common Licensing Source | Common Restrictions |
|---|---|---|
| Public web quotes | Exchange websites/portals | Usually personal non-commercial viewing only; bulk scraping and redistribution banned |
| API data (REST/WS) | Exchange API terms | Rate limits, use restrictions, resale banned, attribution required |
| Aggregated data (paid feeds) | Data vendor contracts | Billed per seat/use; redistribution and cache abuse prohibited |
| Delayed vs real-time data | Different license tiers | Real-time licenses typically cost far more than delayed |
1.2 Common Pitfalls
⚠️ Common Pitfalls
- Resale and redistribution: packaging scraped quotes into a "data service" sold to third parties violates most API terms outright — a breach of contract risk.
- Long-term caching: some terms require market data to be used for immediate display only, banning long-term storage followed by offline redistribution.
- Attribution: even when display is allowed, prominently marking "Data source: XX" is typically required — both contractual duty and liability protection.
2. Displaying Real-Time Quotes: Permissions and Boundaries
2.1 An Exchange API Being "Free" ≠ Free for Commercial Use
Many exchanges offer public market APIs for personal learning, but terms commonly state:
- Personal non-commercial use only (commercial products need applications/payment/agreements);
- No high-frequency polling or endpoint abuse (exceeding rate limits may trigger throttling or bans);
- No redistribution to third-party terminals.
💡 Practical advice: check the terms before commercializing
Before building a public-facing quote site/app, review the target exchange's Terms of Use to confirm whether "commercial display" requires authorization; if unsure, email the exchange officially and keep the written reply.
2.2 Delays, Disconnections, and Disclaimers
- Real-time data arrives with latency; futures/derivatives last prices and order books may already have moved — UIs should note "data may be delayed";
- Reconnections are normal; the presentation layer needs "connecting/reconnecting" states so users don't mistake frozen prices for current ones;
- Disclaimers should cover: third-party source errors, network delays, local computation errors (indicators/drawings based on historical snapshots).
3. Tools vs "Investment Advice": The Boundary
3.1 Tools ≠ Advice: Guard the Content Form
| Safe Practice | Dangerous Practice |
|---|---|
| Displaying indicator results (MA/BOLL/MACD) | Popping up "System verdict: buy/sell" deterministic signals |
| Providing drawing tools, replay, multi-chart sync analysis | Promising "follow the signals for 100% profits" |
| Publishing educational content (knowledge bases/tutorials) | One-on-one targeted signal calling for specific users |
| Attributing data sources with disclaimers | Hiding delay and error risks |
3.2 Special Regulatory Attention to "Marketing" in the EU/Hong Kong etc.
- EU MiFID II / MiCA: marketing financial products (including crypto assets) to retail users has explicit rules — promotional material must not mislead and must highlight risks; pure tool/information display usually isn't "marketing", but adding "recommendation" flavor can trigger reclassification.
- Hong Kong SFC: licensing and content constraints apply to "investment advice/promotion"; tools serving HK users that include personalized advice need assessment against licensing requirements.
- General principle: the more specific the recommendation (specific instrument + timing + direction), the closer to regulated advisory activity; tool apps should position themselves as "information and learning tools that do not provide investment advice".
4. Compliance Checklist for Open-Source Projects
For charting/trading-tool open-source projects specifically, common compliance points:
- Data sources and terms: note quote sources and license scope in README/settings pages; re-check API terms before commercial deployment.
- Disclaimer: show "for study and reference only, not investment advice" in both UI (footer/chart watermark) and docs.
- Risk warnings: features involving leverage/derivatives/futures must prominently warn "you may lose your entire principal".
- Geographic restrictions: if your exchange/data sources bar certain jurisdictions, restrict access accordingly and document it.
- No return promises: example strategies and backtests must carry "past/simulated results do not represent future returns".
- User data: self-hosted services collecting user info (watchlists, alerts, positions) must comply with local data protection laws (e.g., GDPR/PIPL) with privacy statements.
- Trademarks and logos: confirm trademark boundaries before using exchange names/logos in promotion (factual statements like "supports XX data" are usually safer than official logos).
Summary
The core compliance posture of market-data tooling: licensed data, attributed display, explicit disclaimers, advice never overstepped. Personal-study and commercial products face different requirements; spending half a day checking data terms and target-market rules before launch beats receiving takedown notices or lawyer letters afterward.
💡 Tools provide information; humans make decisions
Tools provide information, people make decisions; data needs licenses, disclaimers must be in place. The core compliance posture of market-data tooling: licensed data, attributed display, explicit disclaimers, advice never overstepped.
In one sentence: tools provide information, people make decisions; data needs licenses, disclaimers must be in place.