Spyke

Replies

Comment on

Water Containing Radioactive Materials Spills Over at Kashiwazaki-Kariwa Nuclear Power Plant Due to Earthquakes

The Japan Times reported that at Tokyo Electric Power Company Holdings' (Tepco) Kashiwazaki Kariwa nuclear power plant officials "confirmed Monday that water from a spent fuel pool spilled over due to the earthquake, but that no abnormalities in operation had been detected". In an update issued on Tuesday, Tepco said: "At the Kashiwazaki-Kariwa nuclear power plant, the readings on the stack monitors and monitoring posts installed at the power plant site boundaries are within normal fluctuation ranges, and there is no radioactivity impact on the outside world. The spent fuel pool cooling system is in operation at all units, and there are no abnormalities in fuel cooling. As of 12:25 pm on 2 January, all patrols had been completed and no abnormalities caused by this earthquake were confirmed."

https://www.world-nuclear-news.org/Articles/No-abnormalities-reported-at-Japanese-nuclear-plan

Comment on

advise needed

Oh my goodness. I have to comment here because some of the advice you're getting is a bit sus. Oh sugarcane farms, sugarcane farms. I've been building them for almost a decade. Optimizing them has been taking up room in my brain for such a long time.

First you need to know that sugarcane has 16 growth states. The block doesn't show any texture change but sugarcane will over time receive random ticks and count up in state. Once it has fully grown only then will it create a new sugarcane block above. Crucially once the sugarcane block reaches it's final growth state it doesn't reset. So if you have 1 high sugarcane and you harvest it immediately after it grows 2 high that only requires one random tick. However if you wait to harvest it until it's 3 high that would require 17 random ticks (which will take a LOT longer). This also means if you want to benchmark your farms you should wait until all the bottom sugarcane blocks are fully grown because freshly placed sugarcane will give you a very slow start.

Generally speaking there are two types of sugarcane farms that I see.

First, there are one-piston-per-sugarcane farms. Logically the two ways to optimize these farms are to break the sugarcane asap after it grows and reduce the resource cost. The farm you have right now is ok. You essentially have a clock that pushes the pistons periodically. Testing will tell you if you have the time dialed in, I'd suspect that a faster clock will be more efficient so that would mean moving the observer one block lower or even stripping out the current design for a different style of clock, but you would want to test to see how that turns out. The sugarcane can't grow on the tick where the pistons are extended so eventually that will interfere with the farm but it's not an issue until you have a fast clock going. If you want the farm to harvest immediately when the sugarcane grows, you either need an observer per sugarcane OR you need to use a budded piston setup. The budded piston setup the way I built it in the past would essentially mean that if you have 13 sugarcane you need 14 pistons + some other components like Redstone and 1 observer (IIRC). Sorry I don't have a tutorial, no one makes these farms anymore since they added observers. Don't worry about that because we can do more optimal than the one-piston-per-sugarcane farms using the second type.

Second there are slime-block/honey-block sugarcane farms. These farms push some kind of slime contraption across a field of sugarcane. They are a lot more resource efficient in terms of the redstone because the mechanical part of the farm can harvest a much larger area. Observers have made "sweeper" flying machines really cheap. 2 pistons, 2 observers, and you can build a sweeper made out of slime or honey blocks out to just short of the push limit in both directions. The key efficiency observation with this second type of farm is that sugarcane is cheap so don't worry about how fast you harvest an individual sugarcane, instead figure out how to pack a lot of sugarcane into the farm. For example this design is pretty close to optimal. The main problem with flying machines is that it's possible for flying machines to break on servers. This is a lot less common then it used to be as some bugs have been fixed. I think that this is much less of an issue today than it was in older versions so I think that if you want to optimize all of the best designs are going to involve slime/honey flying machines.

As a third consideration I think that your collection method can be optimized. Increase the random tick speed in your world and you will be able to test and iterate on your design much more quickly. What you will find is that sugarcane when broken likes to pop and glitch all over the place out of the farm. The pistons pushing sideways will from my testing cause more drops to escape than pistons pushing downward. This is another way in which flying machine farms are more efficient because the whole area needs to be covered in a collection system drops can only escape at the edges vs with one-piston-per-sugarcane you can usually only do two rows at a time.

In fact this brings us to the fundamental problem with optimizing sugarcane farms. As a certain point it's just a matter of scaling up whatever design you find interesting at the moment. If I'm going to build a sugarcane farm these days the important thing to me is that the design is not boring. I've build too many sugarcane farms and so it's nice to pick something that's a bit different than what I've tried before. Scaling up will hide a lot of inefficiencies. My advice would be to take your current design even if it's not perfect and figure out how to scale it up. For example instead of one line of sugarcane mirror it so that there's two rows of sugarcane that way sugarcane that might escape will instead just land in the other row.

If this wasn't enough information feel free to reach out and have a good day.

Comment on

Apple wasn’t storing deleted iOS photos in iCloud after all

This has nothing to do with the Files app, nor does it have anything to do with re-indexing of the Photos library. This has to do with fighting CSAM. Apple has started (in this or a previous update), to scan your device (including deleted files) for anything containing nudity (search for "brasserie") and adding it to your photos library in a way that it is hidden. That way, anything that the models detect as nudity is stored in your iCloud database permanently. Apple is doing this because it allows them to screen for unknown CSAM material. Currently it can only recognize known fingerprints, but doing this allows them (and the other parties that have access to your iCloud data) to analyze unknown media.

The bug mentioned here accidentally made those visible to the user. The change visible updates the assets in the library in a way that removes the invisibility flag, hence people noticing that there are old nudes in their library that they cannot delete.

...

And speaking of deleting things, things are never really deleted. The iPhone keeps a record of messages you delete and media, inside the KnowledgeC database. This is often used for forensic purposes. Apple is migrating this to the Biome database, which has the benefit of being synchronized to iCloud. It is used to feed Siri with information, among other things. Anything you type into your devices, or fingerprints of anything you view are sent to Apple's servers and saved. Spooky, if you ask me. But the only way we can have useful digital assistants is when they have access to everything, that's just how it works.

Nudes are meant to persist on iPhone. You're just not meant to notice.

linux

Comment on

7 Common Linux Myths You Should Stop Believing

Reply in thread

After several years of using Linux for work and school, I made the leap to daily driving linux on my personal computer. I stuck with it for two years. Hundreds of hours I sunk into an endless stream of inane troubleshooting. Linux preys on my desire to fix stuff and my insane belief that just one more change, suggested by just one more obscure forum post will fix the issue.

Comment on

Outcry from big AI firms over California AI “kill switch” bill

California has pushed out badly worded laws in the past. Here's a definition from the bill.

“Artificial intelligence model” means an engineered or machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs that can influence physical or virtual environments and that may operate with varying levels of autonomy.

Tell me that wouldn't also apply to a microwave oven.

Comment on

*Permanently Deleted*

Reply in thread

You're basically relying on the security of minecraft, and your ability to quickly patch. The Log4j exploit is one good example of the kind of threats you might face.

Another is just that revealing your ip can open an opportunity for various forms of harassment. Lots of us skate by on obscurity and luck without to many issues, but that's not a very robust solution.

Comment on

If you had to teach a casual beginner class on FOSS, digital privacy, and Linux what would you include?

Something I often see missing from discussion on privacy is that it's not always about you, the listener. Sometimes it's about protecting the most vulnerable people around you. For example, someone escaping from domestic violence might have a different view on how their information is protected. People struggle to see the value in privacy because it's not been a big problem for them personally or because they think it's hopeless. An introduction to privacy in my view is all about teaching empathy, hope, and advocating for others.

Once they have that goal in mind, you can tie in how open source helps empower people to take back their privacy

Comment on

Generative AI’s Energy Problem Today is Foundational

Reply in thread

The actual paper (emphasis mine):

In 2021, Google’s total electricity consumption was 18.3 TWh, with AI accounting for 10%–15% of this total.2 The worst-case scenario suggests Google’s AI alone could consume as much electricity as a country such as Ireland (29.3 TWh per year), which is a significant increase compared to its historical AI-related energy consumption. However, this scenario assumes full-scale AI adoption utilizing current hardware and software, which is unlikely to happen rapidly. ... A more pragmatic projection of worldwide AI-related electricity consumption could be derived from NVIDIA’s sales in this segment. Given its estimated 95% market share in 2023, NVIDIA leads the AI servers market. The company is expected to deliver 100,000 of its AI servers in 2023.10 If operating at full capacity (i.e., 6.5 kW for NVIDIA's DGX A100 servers and 10.2 kW for DGX H100 servers), these servers would have a combined power demand of 650–1,020 MW. On an annual basis, these servers could consume up to 5.7–8.9 TWh of electricity. Compared to the historical estimated annual electricity consumption of data centers, which was 205 TWh,2 this is almost negligible.

The article:

de Vries has analyzed trends in AI energy use and predicted that current AI technology could be on track to annually consume as much electricity as the country of Ireland (29.3 terawatt-hours per year.)