Posts by brainwrong 344 publicly visible posts • joined 7 Aug 2007
Surely the input requires a temperature difference to extract the energy to power the heat pump part of the device. This implies a cold sink, which isn't mentioned.
To keep pumping heat is going to require some sort of repeating cycle of operation, which also isn't mentioned.
My low end Casio watch doesn't display the year or month. It's not necessary, because it's quite difficult to lose track of those.
I have never woken up in the morning and suddenly thought to myself "Oh fuck! What month is it, I've got an appointment in March", and rushed to look at my watch.
I figured out where the craters were, but am confused by the dark area only being visible in the upper 2 pictures. I don't buy the explanation in the article because of the well-defined boundary. Were the pictures taken on different dates? Could it be anything to do with observing and lighting angles and retro-reflection?
That's mostly a description of the gig economy. Because there will be fewer normal jobs. More people will have to become self employed, and compete directly with 5 billion others for less work, because automation. No job security. Just even more plates that people need to keep spinning, no time to rest.
"The alleged perpetrator used household spotlights with strategically placed colored bulbs to simulate the flashes and security features found on physical identity documents under real light,"
Clearly they weren't giving the hologram much scrutiny, it really should be obvious that it reflects light differently to boring flat things, especially when moved.
I'm sure someone will be along soon with an AI hologram detector.
“I could never remember ffmpeg's flags,” “I'd google the same conversion command for the tenth time and …"
Cheat sheets are much simpler. I have some on paper that I wrote several years ago, when I was new to linux, and now maintain a .txt file with a bunch of useful command templates for ffmgeg and other useful tools. New information is found via web searches (which may be old and out of date), cross-referenced with man pages (which may be incomplete, or ambiguous for non-experts), and tested until I figure it out or give up. That way, I understand what's going on.
It isn't possible to build a machine that can discriminate human from LLM written text, because LLM's are specifically designed to mimic human written text. If such a machine were to be built, then surely it would be adapted and used to create a better automatic text generator to replace or augment the current LLM's, thereby rendering such a machine useless for it's original task!
To quote Mike 125 above, "The question is: Why are 'intelligent' people not able to grasp this?"
"It would take around 33,000 years to count up to a trillion."
The current issue of Private Eye features a piece about a nigerian named Favour Ogechi Ani, who has set a new record for counting aloud, up to 1,070,000. He says he did 2 sessions a day of 7 hours each, and it took him 70 days. Scaling that up linearly (ignoring the ever increasing pronounciation time) gives over 179,000 years, or over 104,000 years if you cut out sleeping and eating and life.
I don't know who was checking he didn't miss any.
"such as blacklisting VPN ingress or egress IP ranges, don't work for long. VPN operators are adept at moving these"
Although netflix and suchlike seem unable to keep up with VPN operators swapping addresses (1), in my limited experience most piracy sites have no problem detecting if a VPN is in use or not.
(1) - I've never understood why netflix & co don't block content they're not contractually permitted to show you based on the account you're logged in on, rather than the supposed location of your IP address. Looks to me to be a deliberate weakness.
It's worse than that, if you want 100W of electricity then you'll have to dissapate substantially more than that in heat, the thermocouples that convert heat energy to electrical power aren't very efficient (<10%).
New technology may improve things, I recently read about a company called Fourth Power who had recently demonstrated a thermophotovoltaic cell with over 40% efficiency, although they may be using temperatures somewhat higher (up to 2400 C) than you could use with radioisotopes. Would anyone fancy a white-hot can of liquid Pu-238 inside their nuclear battery?!!
None of this shit is ever going to power your everlasting gobstopperlaptop.
Icon because fire!
Trivy - KICK - liteLLM - CI/CD pipelines - API keys - GitHub tokens - GitHub Action component - trivy-action - npm ecosystem - Docker Hub - workflow files
What's all this drivel for? So people can claim to be at the forefront of IT? All I see is complexity and lots of space for holes.
I also saw this, "Trivy version 0.69.4". 'nuff said.
"*An* answer. The point being it may not be true."
It appears that there isn't a single answer, because of how it works. Data and metadata typically have different write profiles (default is 1 copy of data, 2 of metadata). It'd need to know the future mix of those 2 to give a single accurate answer. It could probably do better though.
"No, that is ZFS."
I said they claim that for btrfs, I don't know that it's actually true, but I've not seen any reason to doubt it. But if writes don't happen in the correct order and are interrupted midway then there may be a problem. I've never had a problem here, but I am quite selective in which features I use. Some parts of it appear robust, other parts are by their own admission not production ready. FaceFuck apparently use btrfs extensively, and contribute code to the project, but are themselves selective in which features they use.
"> I'm slightly confused (or better yet -uninformed) - what are the biggest benefits oh using bcachefs over btrfs?"
This is complex and depends on what features you want to use. The only real answer to this is to spend time reading up on them, but much of the info found by search engines is out of date. Bcachefs promises to do everything btrfs was supposed to do, and more, but I think it's still early days.
"1. It cannot reliably report free space. In other words, the `df -h` command lies to you. This means software can't check if it can safely do something without risking filling the volume."
This is annoying. btrfs own tools can tell you the answer, but may require knowledge of the filesystem architecture to interperet the results.
"2. Any attempt to write to a full Btrfs volume _will_ corrupt the volume."
I'm sure I read somewhere that's no longer the case, that it will reserve some space for key filesystem operations. Copy-on-write means that any operation (even file delete) needs free space for the writes. It is worth noting that btrfs allows you to use a greater percentage of the available space than ext4, because inodes are not allocated in advance. I have a btrfs filesystem that's currently 99.9% full, it's got a bit slow. Any filesystem will be troublesome when near full.
"3. There is no working `fsck`. The repair tools do not work. SUSE relies on Btrfs and SUSE's docs say, with a bright red WARNING heading, "do not attempt to use `btrfs-repair`.""
I think the claim is that the nature of the fs is such that it cannot become inconsistent, and therefore fsck is not needed. This does depends on writes being made in the correct order, so beware of writeback caching. I have never had a problem.
There is another nasty gotcha I've read about. If you have a 2 disk raid1, and one disk fails, you only get one opportunity to mount the degraded filesystem as read/write, so you must replace the failed device during this time because that requires writes. Future mounts of the degraded filesystem will be read only, because it is unable to satisfy the raid1 write profile with a single disk. So no using of degraded fs whilst you await delivery of a replacement disk, unlike mdadm raid.
"So you can have some zram as well as a swap file, and it will move the least recently used stuff from zram into disk swap as needed."
I was looking at these recently, and have been unable to find out if data remains compressed when zram moves data to the disk swap. I feel that if it did, then this would be a feature worth telling people about. My understanding of the architecture of the swap system is that if the zram swap device fills up, then the existing data will remain there, and new data swapped out of memory will go (uncompressed) to the disk swap. If data were to be moved out of zram swap to disk swap, it would be de-compressed as it leaves zram.
Zswap does write compressed data to the backing swap disk, so reducing the amount of writes, which is good for ssd's.
It appears to me that zram is only good if you have no disk swap, and zswap is the better option if you do.
I always encrypt swap partition, because raw memory contents may be sat there in the open.
But can we be sufficiently certain of the trajectory of the asteroid (before and after an impactor) far enough ahead of time to know we've done the right thing?
An asteroid will need to be diverted years before its potential impact, it will take years to get a mission to it, and we'll probably need several years of observations to pin down its trajectory. The principle of a small change increasing in time may allow to amplify our small shunt to a larger deviation, but this also gives rise to a larger uncertainty in its future trajectory.
We're gonna need more, larger impactors and much better observations to be able to make use of this.
"Nvidia CEO Jensen Huang has often said Washington must allow export of his company’s products, to ensure US tech dominates the global AI industry."
I would expect china would use the american designed chippery to try to get ahead in AI, rather than wait for their own chippery to catch up first before being able to get ahead in AI. Then once the chinese chippery does catch up, they'll shift their AI systems over to it. I'm sure they won't allow their AI models to be stuck on foreign hardware.
Old methods of plate glass manufacture couldn't get the thickness very even. It makes sense to put the heavier bit at the bottom.
Glass is solid. An experiment you could try is to smash some glass, find some sharp bits, store them for a long time, taking care not to damage the edges, and see it they're still sharp years later. If it were a liquid then you would expect surface tension to blunt the edges. I haven't done this, but maybe your descendants could report back in the future.
I recall it being pronounced "you-rain-us" until the flyby, and television started fretting about the mis-pronounciation and gave us "yur-a-nus", this being the trigger for the spitting image sketch.
I always thought "yur-a-nus" was stupid, and anyone saying "your-anus" was an idiot. I've never understood why anus's are funny.
