This is actually a really good idea and I can’t believe nobody thought of it before.

It's a very good question. It looks like it was mainly failures to generate a result within a predetermined time. Some of the failures were due to cryostat hardware failures (a fridge went out during a NIST campus closure); some due to fiber + interferometer polarization drifts; and so on. It also appears that [perhaps?] a few of the misses are due to latencies in the timetaggers to record a common timebase. I can't quite tell from the arXived version of the paper: https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Farxiv.org%2Fabs%2F2411.052...

All in all, it's a marvelously good overview of the impressive experiment!

"AI" is a marketing term for these types of tools. They are just specialized for a specific type of task, it's just that task can be generating images, video, audio, or text that can cover as wide a variety of topics as you can train it with data. Ultimately, text generating AI is only concerned about putting the "most right" words together in sequence to follow up your prompt. These can be wrong if there is no really right words (eg it doesn't "know" the answer, so you get "halucinations") and it certainly isn''t designed to play a good game of chess.

It's also happening tomorrow, according to the summary. Hmmmmmmm...

Hm. Perhaps this is a use of 'grifter' I wasn't aware of.

How is Geerling a "grifter"?

So you're saying that neither of the groups of people (one of which includes you) keep their eyes on the road. This directly contradicts your previous message. Cute.

Eyeline != focus. You're not keeping your eyes on the road.

My fifth speed is an overdrive.

Everyone keeps complaining that everything is too expensive, but almost no one is buying the low cost options anymore.

Browsers already blocked access to file:// by websites, they will jus treat local or LAN addresses the same way. You'll need to have a compatible origin as the top-level address to access. Should be an easy fix, no legitimate site should be doing this. Any sites that are should be using a custom protocol to invoke some action on the user's device with their permission instead.

It will be nice to have a built-in markdown reader in windows. Then I can write text files in markdown and know anyone can open and read them properly with a fresh Windows install. This does seem a bit odd, but not as bad as the AI integration, and honestly I'd say it's only bad if it interferes with using Notepad when you're not interested in writing markdown (for example, increasing application start time or making it easy to accidentally add formatting).

1, Microsoft should add an API to allow apps to declare content to be private/non-indexable. At the same time, they should add overrides in Recall so users can choose to force index or force ignore specific apps regardless of how they mark their content. The power should be in the hands of the user. If the user indicates they want to index their Signal activity, it should be done in an unblockable way. Conversely, the default would allow for ignorant users' privacy to be respected. And Signal would not have to engage in unsupported behavior that could potentially break down the line (imagine a future Windows where rendering to Recall and the screen is treated the same to simplify the code base and reduce bugs).

2. When I run software on my PC I am granting it permission to use my property to perform some task. If that program goes off and starts doing other things I may decide it is violating that trust and terminate it. That is my right as a PC owner. Signal is interfering with Recall's operation which doesn't really fit in with its basic premise as a messaging app and if it is not configurable this can be seen as sketchy behavior. Since they have announced this behavior publically it indicates it is not meant to be hidden behavior which does make the behavior more legitimate.

3. Apps should not claim that content that users can see cannot be permanently stored. This is at best naive and at worst a bold-faced lie. For example I could take a picture of my computer screen and Signal would never know. And of course the person being spoken too can remember what has been said. Users who are under the impression that what they say will have no consequences may end up behaving differently than if they were aware of the reality that anything they put "out there" may very well never go away.

I suspect the first questions the court will want answers to are a) why is filing a lawsuit a more reasonable option to you than just turning the feature off and b) why don't you just run one of the many free alternative OSs available, many of which can run a wide array of Windows programs through translation layers? It's a free market.

You can always take a picture with your phone of your monitor and Signal would never know. I think this is the real problem with apps that claim to deliver messages without a trace... they try to hide the reality that if the recipient can view messages, they can save them permanently and there's no way to prevent that. Misleading users isto believing otherwise is unethical.