Talk:Christian Astrology

Well, this is a dog's breakfast.

I figured that what would be needed here was to go through here and add basic formatting, and perhaps splitting it by chapters, similar to what I've already started doing at Tetrabiblos. I knew this was a daunting task. Part of the chore would have been to upload Lilly's illustrations and tables to Commons as images, and add them to the text here.

I gather instead that I am supposed to cut bits out of the text pagewise, to match a set of image scans that I didn't know we had. It did not show up in any of the expected categories at Commons, and the file is in a format that displays here but does not download. It might be easier to just display the scanned images as the book. That way all the illustrations and figures would be available to the reader, and this text can be discarded. I'd not know how to make this so, though. - Smerdis of Tlön (talk) 04:22, 5 October 2011 (UTC)

I downloaded the PDF yesterday, converted to images, split and cropped them, combined to DjVu and applied OCR in order that we could have the reference available, since this work apeeared to have only one "real" scan source, and it was it a badly formatted PDF. We prefer works to be matched to the pages so we can see that the transcription is correct. Instructions on doing that are at this page. We do need the text here, so it can be searched, copied and used. The page images are already available in the form of the DjVu file (you might need a DjVu reader program to view it on your own computer, just like you need a PDF reader to read PDFs)

The current text, while poor, is better than the DjVu's raw OCR, so it can be "matched and split" to prevent you having to copy-paste each page. I'm going to try that now.

Match and split is not working very well (I managed to split off 80 pages) due to the atrocious OCR I did (the scan quality is shocking, but its the only scan I could find). I'll try to improve the OCR quality, but in the meantime you can copy-paste pages. Sorry about that. Inductiveload—talk/contribs 05:53, 5 October 2011 (UTC)

The images do need to be extracted and cleaned, but a lot of the symbols are available in Unicode. A small selection: ☼☽☾☿♀♁♂♃♄♅♆♇♈♉♊♋♌♍♎♏♐♑♒♓. Inductiveload—talk/contribs 04:38, 5 October 2011 (UTC)