(my comment was mostly joke-y bit)

it’s very easy to have some ingrained racist views

I think part of the reason is that it is at least one part natural pattern recognition. I am from a “diverse” country and more than races, I mostly have region specific biases, of which, we have about 30 to pick from. So yeah, It is big pool to choose from. Naturally living in a diverse nation meant I never looked at people from races being inferior or superior. But at the same time, since I am from a particular region, and since your nature heavily depends on your surrounding, i have region specific things, and I am “region-ist”. Like I do go “people from this religion are very loud” or “timid” or “stupid”. But at the same time, my natural speech is not that harsh, and so even if i have regional preconcieved notions, I am not often hurtful to others.

thing is, models are too big now that vram iis not enough, so first ddr5 prices went to roof, and ddr4 production is simultaneously halting, and is still usable, so leftover ddr4 is also very expensive now

(I am a manga enjoyer)

If I have to choose b/w these 3, i think first is least problematic, so yeah, I must be racist

I genuinely can not even think how you mess this one up. highlighting a key being used is sign that click is registered correctly, and since highlighted key is what you expect, it seems the ui is supposed to pop that up. It also suggests that popping of keys happens independently of key input loop, and is likely that way for performance reasons. key entry should not wait for key pop up to finish, so they both must happen in parallel.

Now I guess the reason is maybe something like - hardcoding where a key would be in the display (or keyboard window coordinate) and if you register a click there, enter that key. Now that is roughly how a keyboard would work, but maybe they increased the hit target (you do not press at defined pixels on screen which lie exactly on middle of keys, so all keys have some “nearby fall zone” where if you click, the key is still registered. Now this should not happen normally, as pressing u should result in u, even if j’s fall zone is large enough to get to it (it should not be, at least not until you do something like swipe input and some prediction which sees what is more likely for the given coordinate and previous letters). Maybe some where the calculation for zones is borked. and it’s internal keyboard coordinate is not matching the keyboard.

Not saying errors in programs can not happen, but keyboard input seems one of things I assumed that you do once, and never touch again, as it is not really changing.

that is a good point. I partially touched on it in my comment, but one of the thing happenning is they are clearing forest in one of our tropical islands, and “reflacing equivalent land” in a temprate only partially forested land. think after giving a full stab wound and adding a temporary bandage on to it. EVen if I believe your planting efforts puts in equal trees, you uprooted a tropical forest with rich flora and fauna, with a place which has no wildlife other than a few monkeys and some other common animals

i paid a repair shop for the swapping. ($15 includes everything, battery itself is around $10 port is around $2). Whole swap took around 20 mins. though this is a third world country, so double or triple that.

but for my device, there was full repair guide available online (ifixit). the only reason I did not repair my self is because I am kinda scared doing anything battery (it was partially swollen, and i do not want lithium fire). Other major reason is I do not have screw driver bits for phone. I can do laptop repairs myself, but phone screws are even smaller, and use fancy heads, so I can not use a pointed knife’s edge to unscrew.

If your device has a ifixit guide, that would be the easiest way, they hand hold all the way through. If not, find replacement parts, see what it would cost, and maybe ask a repair shop for a price. if their quotation is within reason, then they can do a good job and you would not even need any other tools.

In fact, if I would have done it myself, I would have not repaired the port, as that required a bit more work than battery.

I recently got a replacement battery and also the usb c circuitry. Got it replaced for $15 (everything included). Phone is effectively new now.

that is indeed sad

(i am an arch user as well, but i mean this comment in good spirit, i am not tryinng to be sarcastic or downplaying)

have you tried bazzite/ublue or other “ready to use” immutable distros? In theory, 6900xt should work flawlessly with the open drivers that come with most distros. I do not think you have to buy anything. If you really want to buy, maybe 9070 (or 9060)(has that been released or not), but that would still be an upgrrade, and you seemingly want a downgrade.

file systems ideally should not matter at all for gpus. neither should encryption

ATI OS drivers for all GPUs, and drivers for AMD gpus

on arch, I do not think you have to do anything for arch drivers. you just need base linux package, and linux-firmware (default is all firmwares, but if you do not want want that, you just need linux-firmware-amdgpu). other than that, just vulkan-radeon for vulan, and that is about it. ypu may install something for monitoring power usage, maybe finnd something to under/overvolt, but that is it, there should not be anything required for setup.

distrobox is pretty cool. I do not use it right now, but i have used it and it worked fine.

Is there anything obviously wrong or bad about the idea to just use whatever distro you like on bare metal.

no. go with anything that gives you recent-ish updates plus security stuff (most good distros satisfy both).

I run Fedora right now but want to switch to something else.

any reason why do you want to do that. just trying out or any problems. fedora to me seems like a nice distro already. If there is any problem, maybe we can help.

Theoretically you could download the .rpm file which quite a few developers provide on and install it on Tumbleweed too?

It can work, but most likely it will not. to put it simply - most linux packages are effectively very fancy zip files. they use different containers and different compression algorithms. some (for example arch packages) are just zip files (or tarballs to be precise), and metadata is to be handled by separate files downloaded by pacman, hence you would not see people packaging anything for arch (you dont have to do anything). some others (for example debian deb packages) are zip file contaiing 3 other zip files. one of them is the package it self, other being metadata stuff (which have information of requirements, file lists, etc.). RPM packages are similar.

While suse also uses the same rpm packaging format, there is no guarantee that package requirements are packaged same way in suse as fedora. If a package does not have many dependencies, it will likely work, but I would still not recommend it.

But nearly all of them are only installable by executing a script and I don’t feel comfortable doing that.

there is a reason for that. they want something that works across distros and setups. also this way, they know how and where they installed theemselves, and after installation, they can manage themselves.

but for these programs, I would not worry much (the ones you listed are big projects, and trusted, and you just do this once). but more importantly - these install scripts are often very simple. effectively they just download some file from server (something like github release) and then extract to some desired location. other things they do include specific setup quirks management. What I mean is that these are simple enough, that I recommend just downloading the script and reading it. if it is not downloading anything unknown, it is fine. if it is not very readable, then that is a bad sign.

But the question developed if it would be wise to use distrobox to execute random internet scripts without altering your base OS/putting your data to risk.

no. things can escape containers. just try to not run scripts.

While I am very happy with this, I would like to add some context for India specifically.

To put it briefly, our current govt is not climate friendly, and I can say this without further reading, but our forest land has not increased, what has been done is clever redefinition of word forest itself.

Primarily, what would you consider to be true forest is given to big empires in name of development, and as a “eco tax” money is taken for planting elsewhere (which may have completely different bio-geography).

secondly, what would you define as sparse, or less dense forest, or just call them grasslands is also being labeled as forests now. this is done by lowering the required amount of trees per unit area in definition. essentially clever accounting.

Now, if the numbers are by some independent organisation, using satellite data or something, I would trust it and be happy, but if it is aggregating what the nations tell, then I do not trust it.

Nuclear plants receive subsidies in the US and most European countries.

but in most places, the official policy is not to expand their infra. it is either maintainance only, and/or decommision at end of life.

Subsidies for nuclear plants are usually payed out during construction and decommission of plants, but that’s still subsidies.

true.

but for other renewables, the actual material is subsidised. for example, in many places, solar panel installation recieve govt subsidies - 10-50%.

That does not compare in the least to the environmental damage and resource depletion that mining uranium causes.

please look up energy density of uranium. yes it still needs to be mined, but it is just so energy rich, that for equivalent energy production, it requires roughly 100x lower mining than coal. and just to say, we have a lot more uranium than coal (or any other fossil fuels, combined).

Unlike solar

just a side note, but solar panels have a life expectancy of 20-25 years. they also need replacement. and they can not be recycled well.

Uranium-235 is way scarcer than natural gas or oil

just the particular 235 isotope is rare, not all uranium. we do enrichment to concentrate the 235. and uranium is not the only element - there is plutonium, thorium. yes thorium reactors are always 5 years away, but that is partly because there is no interest for building more nuclear.

geological structures that are 100% known to remain stable into the far future.

this is kinda a solved problem. you essentially just drill 1 or so km deep, on lands which are far from tectonic boundaries. just put your waste, add cement/or rocks. then bury with dirt you mined. A great solution? no, but it works

In Germany, nuclear fission was successfully phased out for cleaner natural gas, without adverse effects on power grid stability, and with cost savings in the long run

which resulted in increased reliance over russia’s cheap oil, which after the ukraine conflict started, meant much increased costs. in same period, france, which has a strong nuclear network, did not have an increased demand.

… And now, “What if exposing yourself to radiation is actually good for you?”

you are using a guilt by association argument. yes the claim challenges what is currently percieved to be a appropriate model for nuclear damage (lnt).

I have no comments about your 2nd para. i partly agree. presenting your argument as it was done in video feels wrong.

but, what about the example studies. the town in iran, recieveing roughly 10-40 times radiation of currently considered safe limit.

Treating radioactive material and radiation produced by a reactor with extreme caution is the best practice regardless

I have written that very thing in other comments as well. idea is not to drop safety protocols. just change the fear of things by saying - you have not recieved a unsafe dosage.

here is an example of a very similar thing - consider vaccines which use weaker/incapable strains of virus. or consider the very first vaccine, where they used the “pus” from a cow, to effectively use the cow virus to develop immunity in humans. if you think about it, example kinda matches well - in very low amounts virus is not that deadly. in very large amounts, it caused a pandemic. does that mean that it also follows a linear model (no, afaik, it has more of a network effects thing, so it is sigmoidal).

I am repeating what i have said in other comments - “do not drop safety limits, spread awreness that it is not that bad”

Furthermore, your dismissal of other forms of green energy is outdated

I have completed my bachelors this year. I am by no means a expert, but i think i know enough to say that i am not 20-30 years behind. i am 5-10 behind at best (roughly the time studies take to actually be taught in courses). (yes i have studied energy).

I am very willing to actually listen things i said which are very out dated, but i would like to hear them, instead of a blanket - my information is outdated.

And reducing the cost of nuclear by reducing safety standards actually is unpopular

i have said the same in other comment, but we are not suggesting raise the limits, but make it to public that tiny amounts of radiation is not bad. so someone who protests building a nuclear power plant because they get an additional 1mSv of radiation (safe limitt currently is aroun 5mSv), it does not mean their risk of getting cancer has increased by 20% or something.

in case there is a small nuclear spill away, there is no need to a town/state wide lockdown, which completey brings all economic activity of that state to halt. plus the paranoia, and additional cost to handle increased medical vists. i am not trying to normalise spillaways, just that if it is contained, then there is no need to be paranoid.

that by itself was not a claim, i knew there would be comparisons with other forms of renewable energy, so i wrote why some of them may or may not work, so it is meant to be read like

“Are there not better means of renewable energy generation like solar?”

“Are there not better means of renewable energy generation like wind?”

where I live (cost seems to be rather location dependant) is creating cheap energy

mostly becuse in most places, nuclear does not recieve subsidies. most other forms of energy (renewable or not) are subsidised a lot. And most politician would not want to add subsidies because it hurts their popularity. it is always taboo to do anything nuclear. there are reasons why nmri became mri, nuclear fusion research project just goes by fusion research.

its not calculator, but often stuff which benefits from avx, so hashing, compression, etc.

what’s the hypothetical other option here?

this is not even about safety limits, the whole discussion is that currently the model says it’s never safe. what that means is that all people try to avoid it. all people get scared when it goes above some arbitrarily low limits.

if we raise the limits, we also tell people that yeah there are safe limits, and there is no need to be immediately paranoid. simultaneously, it allows us to adapt some existing coal power plants to be conveerted to nuclear. that can already be done, by currently the radiation limit is very low, and ironically, coal plants emit more radiation already. if you can convert existing infra, it reduces cost.

Even if you removed all safety requirements

nobody is even asking for that. we need limits. if not, people will be immediately get lazy (read industries not spending on safety for profit margins) and accidents would increase.

nuclear has a image problem. it is always presented as - better than coal, but not good for health. environmentalist dislike it for some “damages it cause to world” but the exposure is very low, as suggested in video. it needs a pr team essentially.

The reality is solar just wins in cost

it does not. a centralised nuclear power plant is a lot more energy dense. a small to medium scale nuclear plant will generate more power in some amount of time, as much as a few hectare of solar plant. it is simply because solar energy generation is inefficient (20-25%) and is expensive. it does not run day and night, and power generation is not constant thorough out the year.

solar is currently cheap mostly for the same reason as plastic are cheap - we get raw materials for free. you need high quality silicon, which requires finest of sands (average beach sand does not mean the criterion). you need silver, you need electrode material (for example, nickel or cobalt). for small scale, like housing, solar is fine. you can get one for your roof. but it is not going to keep getting cheaper. it is practically at minima already. battery tech is imroving, and will do for longer, but panels are likely not going to get any cheaper until perovskite happen.

cheaper superconducting links

one - that is not happening (before fusion). on a morre serious note - what does super conductivity solve? super conductivity is going to make only a few things better - whenever you want to do some action against some resistive force essentially. it does not help in any situation, where forces involved are conservative (non dissipiative).

they are federated just as piefed is, but the difference is that in lemmy/piefed, you do not want to follow particular users, you follow communities. on peertube, you follow users. so when you watch a video on one instance, you can watch a video from different instance too, its just that peertube does not have a great recommendation algorithm. in lemmy/piefed, when you go to home page, and just search something, content from all comunities is shown. in earlier versions of peertube, you could not search across instances. now you can. if you want a better search, try - https://sepiasearch.org/

Pardon us, but this post is not meant for this community. YSK is for “facts” and saying something is 1984 may fir “figuratively”, but this does not make it a “objective fact”