tinsuke

joined 3 years ago
[–] tinsuke@lemmy.world 10 points 2 weeks ago

I think that this would be too rational and scientific to cause an irrational market to correct.

I'm leaning more towards a money related confirmation. Maybe an OpenAI or Anthropic internal shared numbers.

Or, one of the four horsemen of the AIpocalypse?

[–] tinsuke@lemmy.world 10 points 3 weeks ago* (last edited 3 weeks ago)

https://github.com/chenxiaolong/BasicSync#permissions

ACCESS_WIFI_STATE, ACCESS_COARSE_LOCATION, ACCESS_FINE_LOCATION, ACCESS_BACKGROUND_LOCATION, FOREGROUND_SERVICE_LOCATION

Optionally used for stopping Syncthing unless connected to specific Wi-Fi networks.

And location isn't a permission granted by default on install (unlike Internet access), the user has to approve of it explicitly.

[–] tinsuke@lemmy.world 29 points 3 weeks ago

Nice!

https://github.com/kristapsdz/openrsync

This is an implementation of rsync with a BSD (ISC) license. It's compatible with a modern rsync (3.1.3 is used for testing, but any supporting protocol 27 will do), but accepts only a subset of rsync's command-line arguments.

But also:

The actual work of porting is matching the security features provided by OpenBSD's pledge(2) and unveil(2). These are critical elements to the functionality of the system. Without them, your system accepts arbitrary data from the public network.

rsync has specific running modes for the super-user. It also pumps arbitrary data from the network onto your file-system. openrsync is about 10 000 lines of C code: do you trust me not to make mistakes?

[–] tinsuke@lemmy.world 34 points 1 month ago (7 children)

Nanite, Lumen... Both flagship UE5 features, IMO, haven't realized the potential (we've been sold by Epic), stutter struggle plagues UE5 titles.

The Witcher 4 is supposed to collaborate and stretch UE5's legs in Open World and (sigh) foliage rendering (!!!).

And instead of fixing all of that, Epic is working on UE6? Lol, feels bad for all the developers who tried/are trying to make UE5 work.

[–] tinsuke@lemmy.world 6 points 1 month ago* (last edited 1 month ago)

Oh man, I totally get you. But as others said, it is relatively easy to publish apps on the Play Store.

And I honestly think that the fight is far from over. I think it only serves Google's purpose to already admit defeat.

If not in the US (whom I believe to be highly interested in tracking down "rogue" app authors), then at least other jurisdictions might outlaw this anticompetitive behavior.

[–] tinsuke@lemmy.world -1 points 1 month ago (1 children)

Rao’s low revenue word choice matters here. He said “exceeding $5 billion” and not “nearly $6 billion,” not “approaching $6 billion.”

In a filing where Anthropic was trying to impress a federal court with its commercial scale, Rao is expected to use the biggest number he can. “Exceeding $5 billion” tells you his real figure is much closer to $5 billion than to $6.

Look, I have a genAI aversion as much as anyone on this community. But...

Is that what was stated in a federal court arguably true? I think so. Trying to define what someone is "expected" to achieve with a statement and using that as some kind of proof? That's too rich for my taste.

[–] tinsuke@lemmy.world 16 points 1 month ago (2 children)

Why?

Google added that it doesn't believe its own Gemini models were used, but still has "high confidence" an AI model was part of discovering the vulnerability and weaponizing an exploit.

No, really, why? If Google itself or their models didn't discover the vulnerability, how would they know genAI was used on the discovery of the vulnerability and weaponization (interestingly, not "creation") of an exploit?

[–] tinsuke@lemmy.world 14 points 1 month ago

The headline (as in AI replacing jobs) is as real as the CRAP (Computer Rendered Artificial Picture) it uses.

[–] tinsuke@lemmy.world 14 points 1 month ago (1 children)

I think it might be very hard for AI-pilled leadership to back off from their claims and AI mandates and it would not help the hype that they might be banking on/profiting from.

[–] tinsuke@lemmy.world 29 points 1 month ago (3 children)

Since they're using (now expensive) genAI and productivity isn't improving, because it almost never does, rhen they need to do something to appease stakeholders and offset the cost of using genAI, ofc.

That and keeping employee's perception of power and self worth down, so they keep working their asses off in the hopes of not being the next in line.

[–] tinsuke@lemmy.world 5 points 3 months ago

For yt-dlp specifically, there is a stable backport that is fairly up-to-date: https://tracker.debian.org/pkg/yt-dlp

It doesn't yet have yesterday's version, but the previous version was made available 3 days after it was released.

 

Just a guide on how I got MariaDB working instead of SQLite for my PhotoPrism instance running on a FreeBSD jail.

view more: next ›