Stephen Gadd

@docuracy.co.uk

Technical Director & Lead Software Developer, World Historical Gazetteer @whgazetteer.org Research Associate, Unlocking Upcycled Medieval Data @ihr.bsky.social GIS Consultant, Layers of London @layersoflondon.bsky.social

Today I start my research trip sailing down the River Seine from Paris to Honfleur (via a few stops) in a reconstruction of the Klรฅstad cargo ship built in 998. Itโ€™s the penultimate stage in the shipโ€™s journey from Rome to London throughout the summer. Iโ€™ll post updates as we go #Saga25 #MedievalSky

BildBildBildBild

Thanks, @drpda.bsky.social, this looks like a really useful source, as it includes dates and waypoints for a number of long-distance voyages. ๐Ÿ‘ I'd love to get hold of Paula Martin's transcript, though: any suggestions of how I might do that?

Philip Allfrey@drpda.bsky.social ยท last yr.

This is impressive work, I wish I had a use case for it! For a paleography course I took a decade ago we looked briefly at a journal of a 17th century Scots mariner. It's been digitised at collections.st-andrews.ac.uk/item/journal... Paula Martin made a transcription, but I don't know if it's online

๐Ÿช„Historical Sea Routing is now much quicker to load, reload, and compute seasonally-plausible sailing routes, thanks to some behind-the-scenes browser magic. Find a 6500-mile return trip in under 1 second! ๐Ÿ›Ÿ Do you have historical voyage data which might help with calibration? [1/3]

๐Ÿ‘‡Results from my latest version of desCartes, a hybrid machine learning system for detecting roads on old maps. Its 0.76 IoU score for roads is 31% better than the closest benchmark. ๐Ÿคฏ Preprint forthcoming on arXiv.

The robustness of the desCartes Ilastik-Segformer-b4 model's road segmentation is illustrated in these images, which demonstrate accurate discrimination between main roads (red), minor roads (green), semi-enclosed paths (orange), and unenclosed paths (yellow) in diverse urban and rural environments.The robustness of the desCartes Ilastik-Segformer-b4 model's road segmentation is illustrated in these images, which demonstrate accurate discrimination between main roads (red), minor roads (green), semi-enclosed paths (orange), and unenclosed paths (yellow) in diverse urban and rural environments.The robustness of the desCartes Ilastik-Segformer-b4 model's road segmentation is illustrated in these images, which demonstrate accurate discrimination between main roads (red), minor roads (green), semi-enclosed paths (orange), and unenclosed paths (yellow) in diverse urban and rural environments.The robustness of the desCartes Ilastik-Segformer-b4 model's road segmentation is illustrated in these images, which demonstrate accurate discrimination between main roads (red), minor roads (green), semi-enclosed paths (orange), and unenclosed paths (yellow) in diverse urban and rural environments.

Really enjoyed working on this with Stan Hinton and would love to get some feedback on methods and findings from anyone out there. All criticicsm welcome, this is what WPs are for!

Cambridge Working Papers in Economic and Social History@cwpeconhist.bsky.social ยท last yr.

โ•New WP from @alexislitvine.bsky.social and Stan Hintonโ• ๐๐ž๐ฒ๐จ๐ง๐ ๐ฌ๐ž๐œ๐ญ๐จ๐ซ๐ฌ. ๐€ ๐ฌ๐ฎ๐›๐ฌ๐ž๐œ๐ญ๐จ๐ซ๐š๐ฅ ๐š๐ง๐š๐ฅ๐ฒ๐ฌ๐ข๐ฌ ๐จ๐Ÿ ๐ก๐ข๐ฌ๐ญ๐จ๐ซ๐ข๐œ๐š๐ฅ ๐จ๐œ๐œ๐ฎ๐ฉ๐š๐ญ๐ข๐จ๐ง๐š๐ฅ ๐๐š๐ญ๐š ๐š๐ง๐ ๐ญ๐ก๐ž๐ข๐ซ ๐ซ๐ž๐ฅ๐š๐ญ๐ข๐จ๐ง ๐ญ๐จ ๐ž๐œ๐จ๐ง๐จ๐ฆ๐ข๐œ ๐ ๐ซ๐จ๐ฐ๐ญ๐ก

๐Ÿ“ฃNew blog post alert! ๐Ÿ“ฃ Who had access to common land in the past, and how extensive were common rights? Our latest blog post by Leigh Shaw-Taylor demonstrates that it was not such an open system as we might assume, even before enclosure... www.campop.geog.cam.ac.uk/blog #skystorians

The Cambridge Group for the History of Population and Social Structure, Cambridge Top of the Campops: 60 things you didn't know about family, marriage, work, and death since the middle ages

campop.geog.cam.ac.uk