4 ms·
The problem with Naptha is that the OCR quality of Tesseract (used by Project Naptha) is not very good for screenshots. Historically Tesseract has been optimize
by eastendguy 5y ago
The problem with Naptha is that the OCR quality of Tesseract (used by Project Naptha) is not very good for screenshots. Historically Tesseract has been optimized for text documents. So the OCR results of this new macOCR app are significantly better.
Another alternative is Copyfish. It is cross-platform and uses cloud ocr:
https://github.com/A9T9/Copyfish https://github.com/A9T9/Copyfish
- duckfang 5y agoThere is 0 reason to use the cloud, unless exfil and selling data is your thing. The cloud == someone else's computer. Never forget that.
- eastendguy 5y agoMy personal reason for using "the cloud" in this specific case is that the OCR results are significantly better. > The cloud == someone else's computer. I think everyone around here knows this and can make their own decisions, if and for what data they use "the cloud".
- duckfang 5y agoBe careful speaking about "everyone". You never know who the 10000 are. https://xkcd.com/1053/ https://xkcd.com/1053/
- bori5 5y agoNot everyone’s in the US
- stjohnswarts 5y agoSure but it's fairly assured that 80% at least are from the USA, possibly higher. Also it's fairly easy to extrapolate the premise of "the 10000" to include the rest of the world