dan_mctree

For a lot of companies, switching databases is effectively impossible. Perhaps not quite as impossible as switching game engines, but certainly impossible enough that it would kill the business. A lot of companies use database specific functionality that's far from trivial to replicate in another database. A lot of that database specific functionality can also be legacy that no one really understands anymore. Migrating without an option to keep these poorly understood but critical systems will set you up for unexpected data loss, corruption or availability issues. And that's after spending a year on your migration. If Oracle pulls anything like Unity here, this will kill off a lot of companies

I don't know why but no one hires or listens to the architect who recommends sticking to safe monolithic-esque systems. If you're not talking cloud and microservices and the newest unproven frameworks, you're an old fogey who needs to get with the times. Even though those types of systems are rarely the most efficient, powerful or safe systems

I work for a company that builds software to be used internally by other business. We have like 200 people tops using it simultaneously with no usage spikes, a perfect environment for regular web servers as we'll never have unexpected scaling issues, yet everyone is dying to go to the cloud. Why? I think it's just because our management, programmers and even customers are convinced by cloud provider marketing that the cloud is cool

There's not nearly enough transformers in reserve to properly rebuild the grid and you can't always just drop in and replace either. New transformers are also not currently able to be produced fast, and having society collapse around you probably wouldn't help while trying to increase their production rate. If we'd lose most transformers, it could be many months or years before large regions could see their power back, enough time to cause vast damage directly and indirectly.

Interesting idea, of course it seems very unlikely, but running with it, what could potential tells be? I imagine there might be tells something weird is off if you got really close to them, so they may stay on the periphery of most humans lives and largely stick to their own (definitely at least for mating).

I'd bet you could tell with DNA analysis, or through other medical means such as scans since they wouldn't have had time to adjust for that. If even a single mimic was found the secret would get out, so they'd have to be very good at avoiding the medical world.

I guess you'd probably be looking at either small groups of particularly successful people who have the resources to keep their mimicry hidden and keep to themselves (nobility?), or at some societal outcasts that roam around, never getting close to anyone.

They'd somehow have to avoid their nature being discovered even after death which may be one of the harder things to believe, since presumably they could also die suddenly from accidents etc. which would often prompt careful medical examination which would then reveal the truth. Of course it's possible some individuals did discover the truth, but no one believe them, but it's hard to imagine that at scale. So if these mimics exist, we're probably talking a few hundred or thousand of them at most

Hey this fits my entirely unsubstantiated pet theory: evolution is not just partially Lamarckist, but there's also a purposeful mental component to it. Animals can -presumably subconsciously- partially pass on traits that they themselves think are advantageous. Eg, in this case, the beetle wants to look more like termites and so its offspring becomes more like termites as seen by the beetle even in ways that don't actually matter

I have no idea how such a mechanic may work, but there'd be a great advantage in success for animals that could develop such a style of evolution

Cooling my forehead by applying water has helped me fall asleep, especially on warm days. I suspect an inability to fall asleep plus chaotic mind wandering is often related to an overheated brain and that this is a good artificial way to help the body cool your head. Brain cooling is supposedly one of the purposes behind yawning, but at least for me, yawning can't seem to keep up

Note: if you try this, don't dry the forehead after applying water. The main longterm cooling mechanism is through slow evaporation just like with sweat

Can we all just be honest and just agree that prohibitions do in fact make access to the prohibited thing harder and reduce the prevalence of it. This is true whether your topic of choice is drugs, guns, abortions, alcohol, some form of sex, vpn access or whatever you want to talk about.

Sometimes the prohibition can make obtaining illegal things more difficult and risky, and many of us are too lazy or risk avoidant to push through that. Sometimes accessing a prohibited things requires social contacts not everyone has access to. Sometimes people will just straight up respect the law and not obtain the illegal thing even if it is easy or avoid providing it. Prohibitions can be very effective in reducing the incidence of some thing, especially if enforcement is draconian

Prohibitions can have negative effects, obviously especially for those who like the thing that is being prohibited, but it just seems dishonest to try to pretend like prohibitions don't change behavior whenever that happens to suit some political preference

If you only offer creditcard payments and want international customers, you will missing out on a lot of customers. For example in the Netherlands, iDeal is the standard of payment and many Dutch customers do not have other payment options easily available.

Maybe they could attach a cable and drag it up to a ship? I'm sure there's ways to help these people with joined effort

And I'm done with developers who proudly boast of 95% test coverage using mocked tests because "we don't want to test the database" and then have database changes cause production crashes. "But my tests were all green!"

Databases, external APIs, dependency version updates, etc are among the most likely places where unintended changes get introduced. Why would you not test that when catching unintended breaking changes is one of the primary purposes of tests? Your simple business logic layer function converting one object into another is not going to be the problem, the database is

We do know that AI's have at least some advantages. For example they're more accurate and orders of magnitude faster at floating point calculations. They'll also have communication interfaces with other (sub)systems that are way more precise and fast than anything we could manage. And then there's the perfect memory.

I imagine such advantages also come with at least a lower time bound on solving many classes of problems and presumably that would be experienced as an incomprehensibly smart intelligence. I imagine it'd feel like chess computers in most areas of life in that the AIs actions would feel impossibly perfect at all turns leaving us far behind in attempts to compete

Not as extreme, but garden path sentences sometimes give me a glitchy feeling in how they lure you into an parsing error that takes difficulty to overcome. For example:

The complex houses married and single soldiers and their families.

It's possible that the crafts are not designed for the earth environment and that it is different enough from expected environments that there's some ways for these to crash. For example, they may not be able to handle earth storms, longer term exposure to microbe-containing rain or random bird collisions.

Imo the most likely scenario if it's true, is that these are automated von Neumann probes that traveled here without faster than light propulsion and that they're here merely to explore, observe or possibly self-replicate. I believe this is the most likely option because it takes just one successful civilization in our galaxy, or even a close one, that employs this strategy.

Crafts made deliberately to meet us would have to originate from much closer civilizations unless the civilization has both FTL travel and much more sensitive and high resolution observation technology than we currently believe possible.

Supposing it is a von Neumann probe. There may be a planet or nearby system where these crafts are being built, but we're unlikely to be near the main star/region of the civilization. We might be stuck with just handful of these crafts for probably many (hundreds?) of thousands of years before any real first contact happens.

They could be von Neumann probes having travelled for a long time. Even without FTL travel, von Neumann probes could colonize an entire galaxy in reasonable time (a few million years). Even just a single successful civilization somewhere in the milky way or even a neighboring galaxy could've accomplished that

Imo its this type of self replicating AI that is likely to cause the first global AI catastrophe. A worm-like AI, even if not close to human level intelligence, could if autonomous and sufficiently good enough at hacking, infiltrate a huge amount of vulnerable devices, and paralyze all of our computer infrastructure in the process. The fallout could be massive, especially if's disruptive enough to take out our power infrastructure by say overburdening our power control systems.

Superintelligence is not necessary for rogue AIs to cause big trouble

What exactly is the ideal sort of hardware to be able to run and train large models? Do you basically just need a high end version of basically everything?

Someone ought to start a charity to hire personal assistants for people like Terence Tao. Or get some government tech grant working on that. That would be probably one of the most efficient investments in progress you can get

The bad guy with an AI may very well build such a competent and fast acting AI that there's no defense possible. To strain the analogy, if the good guy with the gun already has a bullet in him, he ain't stopping much

This is how a lot of the younger adults are keeping their head above water. Not everyone has this luxury though, it's not like everyone's parents made bank

>I ask, how.

A superintelligent AGI could easily follow this three step plan:

1. Optional: Overtake and spread computation to security vulnerable computers (presumably, basically every computer)

2. Gain a physical presence by convincing humans to build critical physical components. For example by sending them emails and paying them for it.

3. Use that presence to start a grey-goo like world takeover through replicating assemblers (they don't have to be tiny)

Now I'm not a superintelligent AGI, so there may be even simpler methods, but this already seems quite achievable and nearly unstoppable.

Never quite got this argument. You can frequently do 1000 fold improvements or do even better by optimizing within the same algorithmic complexity. That frequently flips the bit of "is this software usable".

Sometimes you can even make real world performance significantly better by ignoring the technically better algorithmic complexity, especially when you almost always with low n's

>It's likely the planet will become hostile to (human) life long before AI will be able to do us any harm.

Sure, because the planet is already hostile. However, AI could potentially do much more harm much more quickly if we're not careful. In the pessimistic scenario we could all be dead in 10 years from now due to AI or experience a modern society ending event literally tomorrow. Climate change never posed such an immediate and complete risk

At this point ChatGPT is blowing our human ability to use language so far out of the water on so many levels, I'd argue we should start putting quotes around our human ability to "understand" language. GPT-4 has already eclipsed us when it comes to language

How about just picking a non-cyclical time? Just count the amount of seconds/milliseconds since 1 Jan 1970 where the duration of the second is synced to those on earth. Just have no official days/hours/years whatever, no leap seconds, no adjustments, no nothing. Those details can be filled in at will by whoever needs lunar time for daily use.

As someone who mostly just runs into SO through searches and rarely asks: I want to see these duplicates thrive, I want to see the ideas. It's so common that the first SO thread is useless but that a near duplicate does actually provide me with the answer I need. Unfortunately, near half the questions I land on which I feel might actually help me are shut down, which is not just a complete waste of an opportunity but also a waste of my time.

It's easy to underestimate just how difficult it is to get your context indexed and ranked well. Personally I was under the delusion that if you make a solid website, you'll get some organic visitors through search engines from people searching related terms. Turns out that might well be wrong. In the first few months of launching my fairly serious effort website I received a total of 3 organic visitors. While proper SEO and link spamming would surely get me some more, I found it disappointing enough I just gave up

For me I don't believe there's a way to make the activation small enough of a thing that I would be able to consistently do it. If my task were to consistently lift a finger every Monday I don't think I'd be able to do it.

ChatGPT can provide correct citations because somewhere deep in its weights it does lossily encode real texts and citations to real texts. That makes real citations in some cases be its most confident guess for what is supposed to come next in the sentence. But when there isn't a real text it is confident about giving a citation about, but it still feels like a citation should be next in the output, it will happily invent realistic looking citations to texts that have never existed and it has never seen in any sources. On the outside, as readers, it's hard to tell when this occurs without getting an outside confirmation. I'm assuming though that to some degree it is itself aware that a linked citation doesn't refer to anything

This is absolutely stunning. I'm beyond curious what is possible in this space. Could we potentially start regenerating and even redesigning lifeforms, including humans? How about inducing growth of a second brain? Or perhaps we could learn how to merge multiple already existing brains together and see if we can create one more effective brain