3 ms·
I'm confused. According to the article, several days working results in a 90kb compressed DB. If that's the case, it can't be capturing all the text you see, in
by ternus 2y ago
I'm confused. According to the article, several days working results in a 90kb compressed DB. If that's the case, it can't be capturing all the text you see, including web pages - perhaps "working" here doesn't involve general web browsing? What am I missing?
- tsujamin 2y agoPresumably the snapshots it takes and summarises are periodic, but at what interval I'm not sure. Alternatively (and this is a technical possibility) it could be hooked into WebViews and HWNDs to get notified when it's worth taking a screenshot. Given that you can exempt particular website (in addition to private browsing modes in "supported browsers" already being exempted [1]), that implies some integration between a browsing context and Recall. [1] https://support.microsoft.com/en-us/windows/retrace-your-steps-with-recall-aa03f8a0-a78b-4b3e-b0a1-2eb8ac48701c https://support.microsoft.com/en-us/windows/retrace-your-ste...
- gigel82 2y agoText compresses very well. Also, it wouldn't be capturing the entire web page, just the text you actually see on screen. Obviously that 90kb doesn't include the separate folder with all the jpegs.
- xcv123 2y agoThe full screen OCR will capture the jpeg file names
- wombat-man 2y agoYeah I'm also wondering what kind of perf drag it would be. I'd probably just turn it off. Like when would I ever want this?