An attempt at creating a silver/gold standard dataset for backtesting yesterday & today's content-extractors - View it on GitHub