Fossil Box for Schools including Special Educational Needs. I met up with Mark Baggott yesterday from the Earth Heritage Trust. He has been doing an absolutely remarkable job sourcing British Fossils in labeled boxes free for schools. So far over 150 schools have benefitted.Reading the handwritten letters first hand, from children who have had the... Continue Reading →

Open access geoscience data: First global scale submarine landform dataset driven by terrain knowledge at 15 arc second.Paper, code and dataset published by Yu et al (2025).This dataset is reported to be “the first global scale submarine landform dataset at 15 arc second, which offers a new perspective on submarine landforms, providing key insights into... Continue Reading →

Enjoyed listening and giving a talk to the AAPG and Mining Young Professionals network yesterday. Excellently organised event and thought provoking discussions.Great session from Graeme Bagley on soft skills, then an inspirational talk from Promise Ekeh .This was followed by geoscience digitalisation talks from Shashwat Verma , Peter Kiss and myself on unstructured data and... Continue Reading →

As a follow up to the free webinar I gave on how Large Language Models can help geoscience, a write up of the event is now published in Digital Energy Journal. https://www.digitalenergyjournal.com/home/

The Common Pile: The largest dataset ever built exclusively from public & openly licensed text.The authors Kandpal et al (2025) also release “Comma v0.1-1T and -2T, two performant 7-billion-parameter LLMs trained on text from the Common Pile”.“produces models comparable to those trained on an equivalent amount of unlicensed data. Ultimately, we believe that the Common... Continue Reading →