mosiacmango

deleted by creator

Rose water is meh, but the worst part of turkish delights is the gelatin style chew. I also have a mild walnut allergy which makes them taste “scratchy” to me, so i doubt that helped when I tried it.

Yes to both, although i’ve only had the salted licorice a couple of times. I’m betting some brands would kick my ass, but so far so good.

Turkish delights tend to be terrible. Insanely chewy and sticky, floral and just unpleasant. I also tried some sweet “goat cheese and spice lollipop” candy from mexico i didn’t care for much.

Black licorice fucks though. I’ll stand with the swedes on this one.

Youve minimized login risk, but not any 0 days or newly discovered vulnerabilites in your ssh server software. Its still best to not directly expose any ports you dont need to regularly interact with to the internet.

Also, Look into crowdsec as a fail2ban replacement. Its uses automatically crowdsourced info to pre block IPs. A bit more proactive compared to abuseipdb manual reporting.

They have. Nebula is biggest im aware of, floatplane is another.

Summit is also considering supporting piefed.

Absolutely. I’ve been stalling to see if im going all in on piefed or not.

That would be a huge improvement. I’ve put off creating a piefed account mainly because of the lack of Summit support.

Peifed has a lot of features that the Lemmy devs don’t seem to want to support, and is rising fast in popularity because of it.

Yup, that’s the one. The sponsorblock app integrates directly with the youtube app. I don’t recall any config at all. Its made by siku2.

I’ve had the youtube app crash, but its infrequent and mainly seems to have something to do with it starting a new video without stopping a currently playing one.

It is cat and mouse, but it has an issue maybe every year or so, and they generally resolve it same day or next day with an app update. The hardest part is generating your own google api keys, but there are straightforward guides and you just need to do it once.

The upside? No ads at all, sponsorblock integration, and an easy 10ft interface that works great with a remote. No shitty web site to navigate with a mouse. Just a list of your videos each day. Click and play.

I cant speak to your overheating issues. I have kodi running on 4gb raspi 5s with the default case/fan. No problems. An external usb fan might help you out. I’ve used one of these with a raspi in the past. Its roughly the same size as the raspi, but dead silent. Moves insane air.

Have you tried placing a speaker next to the window and playing unmuffled engine noise at him whenever hes out there?

He probably wont enjoy it played back to him non stop, especially if it makes it harder to work on his actual engines.

You know what’s another great improvement? Toss the celery in the trash.

Lawerence systems has a recent video that is pretty indepth.

Belly fat is unfortunately just calorie reduction. You can’t target fat on specific parts of the body, just overall.

Improving core strength with no impact on the spine is tough, as most core exercises will likely engage the back and spine. I would try to implement standard, good form push ups if you can. Planking is also an option, but again form is king. Do them properly, for less time instead of “cheating” with loose form if you want to make improvements.

Another thing to add on to why weight isn’t a great measurement - You will put on muscle as you lose fat, especially if you shift your diet while you workout. It’s fully possible that you lose 5 lbs of fat in a month while putting on 5 lbs of muscle. Both are great strides forward, but the scale will show no change.

If you are going to use a scale, get one that will do body fat analysis as well. They aren’t 100% accurate, but its another measurement that gives a closer look at your actual fitness.

Basically 48 hrs between muscle groups. Mon - Wed - Friday is a reasonable pace if you do an “all in one” routine. If you want to do daily lifting workouts, you need to rotate arms/legs/core then repeat.

OP, it seems counter-intuitive to workout less to improve, but you have to realize that you gain muscle by straining and overworking current muscle. To build and repair, the muscles need rest, nutrients and time. If you don’t give it to them, they can’t build up.

Sleep, time and food are all as important as how much and how often you lift weights.

Perfect? Who said anything about perfect data? I said actually fact checked data. You keep movimg the bar on what possible as an excuse to not even try.

They could indeed build models that worked on actual data from expert sources, and then have their agents check those sources for more correct info when they create an answer. They don’t want to, for all the same reasons I’ve already stated.

It’s possible, it does not “doom” LLM, it just massively increases its accuracy and actual utility at the cost of money, effort and killing the VC hype cycle.

That’s how they work now, trained with bad data and designed to always answer with some kind of positive response.

They absolutely can be trained on actual data, trained to give less confident answers, and have an error checking process run on their output after they formulate an answer.

If “they have to use good data and actually fact check what they say to people” kills “all machine leaning models” then it’s a death they deserve.

The fact is that you can do the above, it’s just much, much harder (you have to work with data from trusted sources), much slower (you have to actually validate that data), and way less profitable (your AI will be able to reply to way less questions) then pretending to be the “answer to everything machine.”