So this sent me down an absolute rabbit hole. As a DM there's a few ways I'd consider to stop this being entirely game-breaking:

• You could argue that the only thing strength before death shows is that you can activate strength before death between hitting 0 and getting knocked out. A wizard is no samurai. Therefore concentration spells are not allowed.
• You could argue that life steal requires life to steal, and as such you can't life steal yourself.
• You could enforce the requirement of the figurine required for vampiric touch, then engineer a scenario to remove it at a critical moment and see if they realise.

Personally I would instead depart from RAW and point out a version of option 2, but a lenient one. Something like "you can do this but you are sapping your very essence to do it. Every time you do it, you permanently lose 10% of your HP" or "every time you do this you increase the number of death saving throws you must succeed before you die". Or my personal favourite: "every time you do this you perturb the very laws of nature. Nature is rather fond of its laws and so decides to perturb you right back. Roll on this table to see what happens." and make the table include the above alongside a few other things and maybe a roll on the wild magic table.

In the end I enjoy ingenuity but the role of DM gives you a lot of latitude to... handle... those who believe they found a loophole.

I mean aside from the obvious, they also try to show that insertion sort is better that bubble or selection sort by... Showing that their worst and average time complexities are the same? Just utter crap and anyone who should be writing things like this should have spotted that it's shit.

Edit: on closer inspection that entire comparison section is utterly dire. Completely nonsensical.

Oh man, as someone who worked directly with the finance side of TFS alongside many other car finance houses (and manufacturers and dealers) this amuses and traumatises me in equal measure. This ain't the letter you get from TFS, those do not gave grammar errors in them. They also would likely say the VIN or reg of the car that was coming off finance alongside likely your finance plan number.

Fucking sovcits.

That would be a 22° halo, a fairly uncommon atmospheric phenomenon where light refracts through hexagonal ice crystals in the atmosphere resulting in an average deviation from the angle it comes in at by around (funnily enough) 22°.

There are lots of other interesting atmospheric phenomena including sundogs, moonbows, and the much rarer 46° halo!

That reads to me as a F#m with an augmented 5th. The notes of a simple tonic triad of D would be D F# A. Meanwhile an F#m would be F# A C#. If you augment that C# to a D and take the second inversion of the chord then you again get D F# A.

The actual reason you would write it like this would really depend on what you are doing musically in the piece more widely. If you were going F#m -> Bm through D as a passing chord, you could consider it as an F#m aug5, however this kinda would make more sense if the other parts of the piece implied that chord to be an F# chord.

In general don't worry about it too much as often you don't really mean the alternative representations that it suggests, but there is some fun music theory underlying this.

Depending on where OP is, that's not strictly true. If you are in a situation such as this, at least within the UK, you are not strictly entitled to the rights of a tenant if you do not pay rent nor do anything in lieu of rent.

Basically in the UK if you do not have a tenancy agreement, cohabitation agreement, or license to occupy, then it can start getting very complicated. If they were named as a property owner, or had a common understanding of financial interest in the property, they might be able to fight for a stake of the house, but that isn't really the point here. In the end whether they can be kicked out legally is a complex issue (at least in the UK) and not really a question we could answer here.

Radical and altogether stupid idea (but a fun thought) is this:

Were lemmy to have a certain percentage of AI content seamlessly incorporated into its corpus of text, it would become useless for training LLMs on (see this paper for more technical details on the effects of training LLMs on their own outputs, a phenomenon called "model collapse").

In effect this would sort of "poison the well", though given that we all drink the water, the hope would be that our tolerance for a mild amount of AI corruption would be higher than an LLM creator's.

This poisoning approach amusingly benefits from being a thing that could be advertised heavily, basically saying "lemmy is useless for training LLMs, don't bother with it".

Now I must say personally I think that I don't really think this is a sensible or viable strategy, and that I think the well is already poisoned in this regard (as I think there is already a non-negligible amount of LLM-sourced content on lemmy). But yes, a fun approach to consider: trading integrity for privacy.

Honestly (and I see you do recognise this in your comment) but this really seems like a kinda crappy study that I'm surprised made it into plos.

For instance I couldn't find any evidence of them considering that the dietary choices of the guardian may affect the attitudes of the guardian to vetenarians (and thus the self-reported health of those animals). To take this further, in the scenario that a cat guardian believes their choices make their cat healthier, especially when going against vetinary orthodoxy, the guardian is probably less likely to take the cat to the vet for minor issues. This confounds the analysis of "healthiness" as performed by the authors.

Furthermore any cat that is not an indoor cat is likely also not fed a purely vegan diet (as they do hunt), so they should possibly account for that via a sort of bootstrapped approach. Generally the stats were okay though, and don't make super strong claims from some pretty weak data. Though GAMs were a pretty odd choice and I'd have preferred some sort of explicit model fit with Bayesian fitting or NLLS.

In the end all of this points to the sort of thing where they should really have been doing perturbational research. I.e. feeding cats different diets in a controlled lab space. This is not the sort of research that lends itself to surveys and that seriously impacts the actual practicality of its findings.

Also as an aside, I really cannot abide anyone who includes a questionably inspirational quote that they said themselves in the fucking French Alps on their own website. That's just pure wankery. The only people I usually see doing things like that are scientists like Trivers, which is not company one should wish to be in.

Yeah, that is exactly how sponsorblock works: crowd-sourced manual identification of sponsor segments!

So here's the rationale that is generally used: If you are in a country that utilises the death sentence then the only system that can decide that is the legal system. Vigilante justice, even when morally justified in the immediate, is not a rigorous or systematically moral justice system. Ergo if anyone is in danger of being killed then they must be protected, even if they are a terrible person, as they have not been sentenced to death (or even if they have, that sentence is not to be meted down by just some other random person).

If you are in a country with no death penalty, you as a society believe that no-one should ever be killed as retribution or as an example to others, thus the argument for protecting people from serious harm is obvious.

These same basic arguments apply for corporeal punishment.

Those who are believed to have committed horrific crimes such as those you mentioned will be in extreme danger because their crimes are fairly universally considered reprehensible (because they... You know... Are). The danger is that there is no perfect justice system. Miscarriages of justice do occur and whilst you may believe that actual perpetrators should be killed or maimed in prison, the risk is that innocent people may be subjected to a horrific and irreversible punishment for no crime at all. That is not acceptable to most people within most justice systems.

Best bit is with those colours you could create an infinite number of bro-bordered pool segments with each bro-bordered segment sharing a side with no other segment of the same colour.

Roland CS-10-EM are excellent binaural mics for a very low cost :)

My own classic was fiddling with the nvidia PRIME config to try and get rid of some very mildly irritating screen tearing. No graphics output at all. Now this is fixable of course, but it's a pig.

And I'd decided to do this 2 hours before an incredibly important progress review meeting for my PhD.

Got it back with about 10 mins to spare and decided just to leave the driver config alone after that.

Bonus round

Also a friend managed to bork his ubuntu 16 laptop by trying to switch from unity to gnome and ending up with sort of neither. That was reinstall territory right there.

Honestly the first set of students coming in at undergrad after covid that I had were simultaneously wonderful and also felt about 2 years behind where they usually are socially. It was a bit of a struggle getting them to properly sit down and think. They did absolutely thrive when you got them going though (with some kinda more experimental pedagogy) so I do still have hope.

I mean, just give them money?

Put it this way: getting a job is just one of many challenges facing homeless people.

For example, if you get a job but are already living absolutely hand-to-mouth, can you actually afford to have that first month of work with no money coming in on a day by day basis. If you cannot afford to even eat how will you make it to that first paycheck?

Even if you do, where will your job put that money? Many, many homeless people do not have a bank account, and what do you need to open a bank account? A home address and ID!

Were you fortunate enough to become homeless with a copy of your birth certificate or other form of ID? If not oh that's not a problem sir, it'll cost you £35, and then it'll arrive by recorded delivery to your home address. Where was that again?

Pretty much no person is homeless by choice. Most are there by a combination of bad luck, violence, a lack of a social security net, mental illness, and many many other factors. Very few people would choose a life of danger and unprovoked violence. You wouldn't want to be without a home, they don't want to be without a home for the exact same reasons.

So in conclusion, it is the very basics of human decency to feel bad for them. I would urge you to go further and try to help them, whether that be by direct contribution, by volunteering, by donating to a housing charity, or something else.

I don't really love this preprint from at least an academic perspective. They don't really talk at all about the specs of the monitor, cable, or target machine. When you're talking about emi interception then discussion of the test conditions is kinda important in a paper. What's the base emi leakage for the system? What's the range in commonly available cables. Is this affected by shielding?

Also I really don't see why they're using a hough line transform to detect the blanking interval. Those two things are not really related (in that probabilistically or exhaustively fitting lines to an image does not easily result in an estimate of blanking interval, and is horribly inefficient in realtime applications too.)

Basically in my opinion this is a cool idea with a pretty mediocre preprint attached, and one where a bunch of the sources are other preprints too. Not damnable but I'd expect more.

If you wanna see a much better paper containing more of what I'd expect from a physical attack vector paper, I'd look at the original Rowhammer paper from 2014

(also the use of the term AI in the actual article is irritating. It's a basic CNN, it's not incredibly complex stuff. Just call it ML guys...)

This is a confusing report and kind of one that misses some key points that I'd want to see investigated in more detail were I to be reviewing a paper like this.

For one, THC is not only taken by smoking it, but without seeing the conference posters this is based on I cannot see how they have controlled for that. I assume they haven't as they relied specifically on already present medical data which they themselves note is possibly miscoded between hospitals.

In all honesty smoking anything will likely increase these factors because, well, smoke is full of some fairly nasty stuff.

Additionally they specifically chose people for the study classified as having a cannabis misuse disorder. These will be approximately the worst case scenario.

I dunno, it just seems to miss out a lot of detail that I think is very important for understanding this question clearly and the articles about it are not querying that point at all.

I can say with relative certainty that at least some applicants went for this visa as a long shot because they had other reasons to move to the UK and this gave a more flexible visa than the Skilled Worker visa (like the ability to change jobs without getting re-sponsored).

It was never gonna be the main reason for anyone to move here and completely misunderstood how academics make decisions about their location.

Possibly try "Yet Another Call Blocker", though I believe I had to install it using fdroid.

I first found it when I had a day of 1 spam call with the first digits matching my own number every few minutes.

Even when reading the paper there was very very little meat. It's conjecture built upon conjecture but very little of it seems to stand on its own for me. It's another theoretical framework that is nice to write about but doesn't actually even try to explain much.

Their argument seems to be that there is selection working on everything to increase complexity. Even cursorily there seems to be major problems with such a conjecture. They feel to me like they confuse persistence with drive.

A thing that lasts longer is more likely to be observed by someone born at a random point in time. This is persistence. This doesn't mean that things try to get to a state where they last longer, particularly not chemical structures!

This reminds me a lot of that assembly theory paper that came out a week or so ago and was (in my opinion deservedly) battered by most reputable evolutionary biologists.