TheHobbyist

@TheHobbyist@lemmy.zip

Just a stranger trying things.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

TheHobbyist,

Unfortunately, I’m afraid it will only enforce a two tier privacy system: one where people of importance get granted a privilege when others don’t…

TheHobbyist,

Fedora is still pretty frequently and recently up to date with respect to packages and kernel, not sure you’d be losing much over arch.

But the debate to me is also not that important, I’ve been running fedora and have at some few occasions gotten some instabilities due to updates (mostly Nvidia with Wayland) so I can totally understand someone wanting stability and reliability over bleeding edge).

TheHobbyist,

Is there any case in Switzerland where someone was caught and fined for having distributed copyrighted material?

TheHobbyist,

Is this an exception? Having heard in France and in Germany how easily and often people get caught, this is the first time I hear of such a situation in Switzerland.

Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data (www.404media.co)

ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more....

TheHobbyist,

This is not the case in language models. While computer vision models train over multiple epochs, sometimes in the hundreds or so (an epoch being one pass over all training samples), a language model is often trained on just one epoch, or in some instances up to 2-5 epochs. Seeing so many tokens so few times is quite impressive actually. Language models are great learners and some studies show that language models are in fact compression algorithms which are scaled to the extreme so in that regard it might not be that impressive after all.

TheHobbyist,

Are Meta even committing to stop tracking when users pay? Or are they simply not showing targeted ads but still totally tracking?

TheHobbyist,

Where do they claim that?

The article from Facebook I found about the subscriptions is this one: about.fb.com/…/facebook-and-instagram-to-offer-su…

The only relevant thing I saw related to the topic was “while people are subscribed, their information will not be used for ads”. It does not say that information will stop being collected. Just that it will not be used for ads.

So by all interpretations, there is in fact no suggestion that they will stop tracking paid users.

TheHobbyist,

More notably, what it also does not mean is “we will stop collecting it”…

TheHobbyist, (edited )

I’m not sure you’re attempting good faith communication, but in the case you are, I think most people’s opinion is that there could be room for Google but people are just concerned about Google being the only option instead of one of many. That’s also my interpretation for GrapheneOS’s stance, they don’t intend on breaking compatibility with Google services but instead run them on your own terms, putting the user in control of how Google operates on their phone. Hence, I don’t see any contradiction in your two statements.

TheHobbyist, (edited )

I don’t know what GPU you have, whether it is AMD, Nvidia or Intel, but if you use Nvidia, the standard tool is nvidia-smi (if you install the proprietary drivers from here with the optional cuda package, you can access the command) . You will need to combine it with the watch tool for real time info.

Otherwise, the one I prefer is nvitop.

I’ve seen Intel specific tools too, but don’t recall them. And I’m not familiar with AMD tools.

TheHobbyist,

May I suggest the following shows:

Both of them are rather old, but quite great in their unique way. Perhaps you enjoy them too!

Mumin is even an anime, if you like that, but was widely distributed in Swedish as it was a collaboration work with a Japanese anime studio.

Edit: adding links

TheHobbyist,

I don’t know if you want to talk specifically about codec settings but it seems overall the compression quality goes AV1 > HEVC/h265 > h264. So if you have the time and/or hardware go for AV1, otherwise HEVC will be perfectly fine. I don’t know of any situations which calls for h264 nowadays.

TheHobbyist,

Sure there is a way. One option is Pegasus though it’s not particularly cheap, with previous price mentions in the millions of dollars. Perhaps within your budget?

TheHobbyist, (edited )

Setting up a server from scratch, exploring the possibility of making an LTE expansion card for the framework laptop (mostly an excuse to get some experience back in PCB design), and automating the creation of frequency lists for specific Japanese anime for language learning. Progress on all of them is slow :) Also trying to get into the habit of writing myself a privare blog to keep track on some of the progress and learnings in each of these.

TheHobbyist,

I’m not sure I understand your proposed solution, but the ferrite core guides the magnetic flux, there are very few losses, and both circuits are perfectly electrically insulated from each other. How would your solution be an improvement?

TheHobbyist,

2

TheHobbyist, (edited )

Thanks for the input. I had already looked up the EC25 which is based on the same Qualcomm chip and a tad larger than the L6, but am unfamiliar with Sixfab and the telit chip you mentioned. I’ll look them up more. Cheers.

Edit: the Telit chip is 28x28mm and is LTE UE Cat-M1 with 1Mbps up/down from what I understand. Does not seem to offer LTE cat 4 speeds.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • localhost
  • All magazines
    •
    Loading…
    Loading the web debug toolbar…
    Attempt #