I'm no fan of Fox News, but my understanding is that lots of the climate data /is/ disputed, because the raw data is gone and all we have is either irreversibly

I'm no fan of Fox News, but my understanding is that lots of the climate data /is/ disputed, because the raw data is gone and all we have is either irreversibly 'corrected' data or summarized reports of the data. That's not trustworthy or useful, because without the original raw data you can't reproduce any of the analysis. Without the ability to reproduce the analysis, you're not doing science anymore. To connect back to the article, if the claim is true that 90% of the thermometers for one of the datasets are positioned too close to structures that alter their readings, and there's no information available about exactly what those influences are for each thermometer, then we can't model and correct for the influences. Again, the data is useless. It can't even be used to say what the temperature is 100 feet away from the thermometer, let alone globally.