- Posts
- 2
- Comments
- 302
- Joined
- 1 yr. ago
- Posts
- 2
- Comments
- 302
- Joined
- 1 yr. ago
Yeah Microsoft is already enshittifying Github Copilot at impressive rates. They have figured out there's only money to be made with AI in the enterprise space. Giving it to the plebs for $10 per month was a losing proposition, so they are back peddling real hard.
The VC money is going down fast and they need to show something, so they need to stop hemorrhaging money ASAP.
The blotches come from biasing in the sensor, some of these patterns were caused by the manufacturing process, others come from the mounting inside the camera and others come from environmental factors. When you get near the noise floor of the sensor, these kinds of defects can become visible.
In order to fix this, take a whole bunch of flat files and dark files. I like to have a set I prepared one time, but also create a set each time I start a session. It doesn't take long and helps with environmental noise. Then stack away at your data in order to boost the signal to noise ratio. These kinds of things should disappear if you do it right.
I use Arch BTW.
Like just huge arches instead of windows or even doors, Arch is all you need.
I’ll be honest, I don’t actually know what all the physics in those papers means. That’s not my job and not my expertise.
Well maybe not the right dude to ask then?
My 2 cents: we'll never get to any sort of practical quantum computer size. As size increases, decoherence becomes a bigger problem. This is currently fixed by having more qubits to compensate. But as the size grows, the amount of qubits needed just to compensate for decoherence grows faster. So there's a practical limit on how large of a machine is possible. And it isn't like a smaller machine is just slower, it actually simply can't do any of the cryptography breaking stuff.
From my understanding decoherence is a fundamental part of reality, which can't be helped. But who knows, there might be some breakthrough that allows for it to work. It might also be impossible given the laws of nature. And what I gather it's also impossible to prove it can't be done.
So that's why quantum computers have been in this limbo state for years now. They might be just around the corner or they might never exist.
In the security world people are worried stuff is stored today, for it to be decrypted in 20 years time. So there is a push to think about this and take precautions. This seems smart, not because they think quantum computers will exist, but just as a precaution in case it turns out they do.
Very good! Your spidey senses are working perfectly. Hey I want to comment this calculation, why don't I move it into a function so the name can explain what it does. Good call!
Sometimes the algorithm is inlined for performance, sometimes it's a class with a bunch of functions that as a whole is primarily based on an algorithm, so comments might make sense in those cases. Most of the times it's a library, so the name of the library kinda gives it away and hopefully has good documentation as well.
Asking an LLM to add comments is actually pretty much the worst thing you can do. Comments aren't meant to be documentation and LLMs have a habit of writing documentation in the comments. Documentation is supposed to be in the documentation, not in the code. LLMs are often trained on things like tutorials, where super obvious statements are commented to allow people to learn and follow along. In actual code you absolutely do not do this, obvious statements should be obvious by themselves. At best it's extra work to read and maintain the comments for obvious statements, at worst they are incorrect and misleading. I've worked on systems where the comments and the code weren't in line with each other and it was a continual guess if the comment is the way it was supposed to work, or if the code is correct and the comment wrong.
So when do you actually add comments? That's actually very hard, something people argue about all the time and a bit of an art form to get right. For example if I have some sort of complex calculation, but it's based on a well known algorithm, I might comment the name of that algorithm. That way I can recognize it myself right away and someone that doesn't know it can look it up right away. Another good indicator for comments are magic numbers. It's often smart to put these in constants, so you can at least name them, but a small little comment to indicate why it's there and the source can be nice. Or when there is a calculation and there's a +1 for example in there somewhere, one might ask why the +1, then a little comment is nice to explain why.
Comments should also serve like a spidey sense for developers. Whenever you are writing comments or have the urge to add some comments somewhere, it might be an indicator the code is messy and needs to be refactored. Comments should be short and to the point, whenever you start writing sentences, either start writing documentation or look at the code why it's required to explain so much and how to fix that.
Another good use for comments is to warn away instincts for future devs. For example in a system I worked on there is a large amount of code that seems like it's duplicate. So a new dev might look at it and see a good place to start refactoring and remove the duplicated code. However the duplication was intentional for performance reasons, so a little comment saying the dupe is intentional is a good idea.
I've also seen comments used to describe function signatures, although most modern languages have official ways of doing that these days. These also might border on documentation, so I'd be careful with that.
LLMs also have a habit of writing down responses to prompts in the comments. For example the LLM might have written some code, you say: Hey that's wrong, we shouldn't set x to y, we should set it to z. And the LLM writes a comment like // X now set to Z as requested. These kinds of comments make no sense to people reading the code in the future.
Keep in mind comments are there to make it easier for the next guy to work on the code, and often that next guy is you. So getting it right is important and hard, but very much worth while. What I like to do is write code one day and then go back and read it the next day or a few days later. And not the commit, with the diff and the description, the actual files beginning to end. When I think something is weird or stands out, I'll go back and edit the code and perhaps add comments.
IMHO LLMs are terrible at writing code, it's often full of mistakes and oversights, but one of the worst parts is the comments. I can tell code was AI generated right away by the comments and those comments being present are a good indicator the "dev" didn't bother to actually read and correct the code.
- JumpDeleted
Permanently Deleted
To estimate when to blast out the CME to wipe us out as revenge
The thing about a games store like Steam or Epic isn't the software itself. Steam has shown having a good client is a large part of it, but it isn't the most important part. The most important part is the negotiations with the publishers and game developers to have them publish the games on that store. There is a whole lot of legal and pricing stuff involved. Another important part is a large CDN around the world to deliver the data to customers at speed.
Many large companies have tried pouring millions into this and haven't had a lot of success. There is so much involved and large market forces to contend with.
As for just having something to manage the games on your system there is Lutris. It allows you to easily manage different game libraries and individual games. Plus tools like emulators and such to run older games for example. It's fully open source and an initiative well worth sponsering.
- JumpDeleted
Permanently Deleted
For those of you driving stick: Don't rest your hand on the stick, it wears out the synchro rings. Also when standing still, don't keep it in gear with the clutch pushed down, that wears out parts of the clutch assembly. Just put it in neutral and release the clutch.
I don't know if this is actually true, but my mechanic friends keep banging on about it. In my experience other parts of the car wear out much faster. And most people don't keep cars around more than 5 years. My own car is 11 years old and we have done SO many repairs on that thing. My girlfriend's car is over 30 years old tho, we've kept that thing in tip top condition. It's much simpler than my car so repairs are fast and cheap. It's a Toyota so it should be good for 20 years to come.
- JumpDeleted
Permanently Deleted
The interesting thing about this is that it could be a double whammy. The collision that formed the Moon not only made Earth smaller, it also ejected a lot of material away from the orbit. This made Earth even smaller than it would otherwise have been, had the two bodies merged. And the Moon also formed in the process. The Moon causes the tides which are theorized to have a significant beneficial effect on evolving more complex forms of life.
So just being small might not be enough and having a big moon might also not be enough, but Earth was lucky enough to have both. And that's just some of the things in a long list of things that have to go right to get complex life on a planet.
My feeling is that life is pretty rare, but given there are so many star systems in our galaxy there might be a lot of it still. But most of it is probably very simple stuff. Getting to where Earth is, might be a once every couple of millions of years event within our entire galaxy. So there really might be nothing intelligent out there at this moment in time, there might have been earlier and there might be in the future, but for right now we are it.
It's funny they still refer to this kind of shit as a bombshell. Must be a real small bomb for the zero reaction or consequences we've seen thus far.
On the other hand, you're doing the first ever Moon landing, trying to manually find a good landing spot, running out of fuel and trying not to die. And all of a sudden your navigation computer starts throwing 1202 errors. That has to be one of the most butt clenched moments of all time.
Computer issues are basically tradition at this point.
Because these tech bros like to put forward a fairy tale where they themselves are geniuses and basically carry the company on their shoulders. They maintain this carefully crafted image of being super duper smart and all of the setbacks are just because humanity is catching up with their vision. They use this image to dupe investors into investing huge amounts of money based on ideas that are in principle good, but in reality very much not feasible. A lot of the money is invested purely on the story and image of the one person, where they build each funding round on the previous one, snowballing into something huge.
In reality these people are scam artists, they know how to project their image, how to sweet talk investors and how to lie and commit fraud on a running basis. Sometimes they are caught, like with Elizabeth Holmes, most of the times they are not.
People who actually know their stuff clearly see those tech bros like Altman and Musk are full of shit talking out of their arse. But the general public doesn't know that and believe the lies, but more importantly the investors believe the lies as well.
Yeah those refurb drives from eBay were the last good source. I got a bunch of them last year, 2 of them had issues but were replaced under warranty by the manufacturer. All of those seem to be either gone or not priced very well.
Doing anything PC related these days is very rough with prices being sky high. And even if you are willing to pay, there isn't a lot of good stuff to get. It sucks ass.
But how could this happen? The DOW was above 50.000?
You are repeating marketing. First of all Polymarket isn't against this, they have openly welcomed insider trading. They claim insiders have they best information and thus make the prediction better. Second, when we actually look at the data, it turns out sites like Polymarket are only marginally better than actual guessing. It has nothing to do with predicting anything, it's just straight up betting. There have been a couple of outliers, but those turned out to be because of insiders heavily betting. They skewed the results a lot and that's not prediction, those people knew the outcome. Third, these kinds of companies heavily influence the betting themselves, often "buying the other side of the contract". This raises the stakes and increases engagement. Having it appear one side is favored, means people tend to pile on.
There's plenty of good stories and videos about how fucked up these "prediction markets" actually are. John Oliver just did an episode on it. I would personally recommend this Patrick Boyle video for a good overview: https://youtu.be/e0nsou-1Q2k
Tldr: It's all a huge scam and rigged betting.