I haven't tried on macos, but yeah, the biggest problem right now seems to be implementation differences with how embedded spaces are handled. Earlier, the biggest cross-browser issues were pixel alignments within the QR codes, but those seem largely resolved.
Fair warning though: this was designed and implemented by an LLM, as an experiment to see if it was possible. I only guided it to a working solution by pointing out problems, and never dug deeply into its inner workings.
It is not a tricky problem because it has a simple and obvious solution: do not filter or block usage just because the input includes a word like "gun".
My dad wears smart glasses because he's nearly deaf and the classes show captions for the person he's talking to. They're great. He doesn't use or care at all about the camera. Having the captions would be very useful to him in a courtroom setting. Collateral damage I guess.
It's not enclosed in the final product. It is used during manufacturing. For example, you mechanically compress helium to get liquid helium, then when it depressurizes back to ambient pressure, it's -269 C, which is pretty close to "as cold as possible", and colder than any alternatives.
> > some folks want to use lossless cut
> In that case I would encourage you to ruminate on what the following in the post you're replying to means and what the implications are:
You may have misunderstood the comment: "lossless cut" is the name of an ffmpeg GUI front end. They're not discussing which exact command line gives lossless results.
I was curious what the protein picture was showing:
"Figure 1 Example predictions of SimpleFold on targets ... with ground truth shown in light aqua and prediction in deep teal."
and now I'm even more curious why they thought "light aqua" vs "deep teal" would be a good choice