I'd like a #GitHub feature that extracts/offers the text from images/screenshots (so that we can replace the images) to improve issues and bug reports where people ANNOYINGLY insist on using images for text.
i would have crawled across broken glass for this when i was doing tech support for a FOSS project. windoze users were the worst offenders at the time.
I mostly got such issues from new computer science students. Most of the time not a screenshot but a photo of their laptop screen showing the piece of code they‘re struggling with.
interestingly, using AI for this would be useful and could probably work quite well. Instead we have AI bot for reviews that you can't disable... Oh well...
78% of developers claim AI makes them more productive. 14% say it's a 10x improvement. So where's the flood of new software? Turns out those productivity claims are bullshit.
Paul_IPv6
in reply to daniel:// stenberg:// • • •Molly
in reply to daniel:// stenberg:// • • •Ricardo N Feliciano
in reply to daniel:// stenberg:// • • •SuperIlu
in reply to daniel:// stenberg:// • • •Gabriel Viso Carrera
in reply to daniel:// stenberg:// • • •Baudouin Feildel
in reply to daniel:// stenberg:// • • •Vinicius Fortuna
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to Vinicius Fortuna • • •Teixi
in reply to daniel:// stenberg:// • • •issues with auto-ocr, then auto-generate minimal reproducible example
en.wikipedia.org/wiki/Minimal_…
'delegate like a boss' w/o dogfooding? mikelovesrobots.substack.com/p…
Where's the Shovelware? Why AI Coding Claims Don't Add Up
Mike Judgetoadjaune
in reply to daniel:// stenberg:// • • •GitHub - dynobo/normcap: OCR powered screen-capture tool to capture information instead of images
GitHub