OK, help me out here. Is there some aspect of human society that I'm not understanding, where information is considered more official/trustworthy if it's presented as a PDF report and not as a web page?
-
OK, help me out here. Is there some aspect of human society that I'm not understanding, where information is considered more official/trustworthy if it's presented as a PDF report and not as a web page?
Today's frustration is this report from @osi about Delayed Open Source Publication - a fascinating document, but why is it a PDF?
-
Simon Willisonreplied to Simon Willison last edited by
Here's my first attempt at a Markdown conversion of that document (using Gemini 1.5 Pro against an image of each of the pages in the PDF), created so that I can copy and paste from it and read it on my phone https://gist.github.com/simonw/7b913aaaff8278d2baaed86e43ece748
-
@simon what did you use? In the past I had decent results using https://github.com/VikParuchuri/marker
-
@mseri I used my own little in-development JavaScript app that sends each image to Gemini in turn for conversion to markdown - it's an extension of https://tools.simonwillison.net/ocr
-
@simon @osi I don't see anything in their doc that benefits from being a PDF. That said, you can version a PDF and call it a complete work so maybe that's what they are going for.
I thought it was weird when the PSF did the same thing earlier in the year.
Most foundations do this with financials too and docs one might share via email.
-