Technically it's any medium. The lower the refractive index, the closer the particle needs to travel to the speed of light in vacuum. But you can for example measure Cherenkov Radiation in the air (where n~=1.0003 or 99.97% of c) from highly energetic cosmic rays.
Especially with the current rate of progress, you asking ChatGPT today probably means you'll get a way better answer than whatever the author got using a model from last year.
This, and the fact that GPT is writing for you personally. A text written to persuade a publisher, or to manipulate Google results, has the deck stacked hard against it vs. asking even a cheap AI that can size you up personally.
That's more or less what they did according to their announcement. They fired it at a whole bunch of high end math problems and merely concentrated all efforts on one after it made some promising progress.
It seems like that’s the opposite of what happened. They started attacking the problem when they got a wind of a possible solution from certain individuals.
But they didn't know which problem had a possible solution. So they fired it at a huge amount of problems and then merely focused on the one that turned out to be promising. The model found the promising path itself, they just re-allocated the available resources once it became apparent.
I actually had so many troubles with vmware on windows over the years that once I tried out hyper-v for the first time last year, I somehow have never touched vmware again. Sure vmware has better configs, cleaner interface and some very subtle design advantages. But damn, hyper-v just fucking works and it keeps working across updates. I have a VM that has been running for a year and it is still perfectly doing exactly what it should. I never experienced that with vmware.
> I actually had so many troubles with vmware on windows...
I assume this means that you were using Vmware Workstation? The VDDK is for Vmware vSphere, which -IIRC- runs on Linux and coordinates computers running ESXi, which are where vSphere VMs actually run.
Not that many people are going to be using Vmware Workstation to run a large number of VMs.
I think you're conflating different requirements/usecases.
Sure, a type-1 hypervisor like vSphere is technically better since there's no host OS creating unmanageable contention and if you have guests with higher performance requirements then yeah it's settling for something half-assed.
But in reality, we have some similar shit to a "fax server" that has manually set up weird manufacturer printer settings (local admins tried to be automated/distributed) and old third-party software libraries that "infects" a system globally (2 versions of that library cannot co-exist on a computer).
If that piece of crap can be contained on a VM, as long as it's contained and keeps trucking after a restart/powercycle everyone is happy. What kind of VM container hosts it is really secondary.
(What the software in question is can probably be inferred from the comment, the software is liked by users at the client so it's not going away)
> I think you're conflating different requirements/usecases.
I disagree. protoculture's core statement seems to be:
A lot of people think VMWare Workstation is the pinnacle product of the company and I have NFI how this came to be.
A company's pinnacle product can be completely useless to me, but that doesn't change the fact that the products of theirs that I do use are very simple and unimpressive when compared to it. [0]
Having said that, I expect that a lot of people think Vmware Workstation is Vmware's pinnacle product because they've never heard of vSphere, ESXi, or anything else the company has made. "People have no knowledge of like 95% of what most companies do." is always a solid first guess when trying to work out confusion like protoculture's.
[0] Please spare me the "But how can it be a pinnacle product if it's not useful to you?" retort and ones like it. I've deliberately made a distinction between a product's usefulness to me personally and the sophistication of that product.
I don't think we're particularly in disagreement here, I think pinnacle product in manys eyes can be both reputational(prob workstation) or economic in addition to technical, so my reply was mostly in terms of replying to the NFI part of that statement and not really in terms of any personal judgement.
I think waaay more people struggle with this than HN would have you believe. In the real world, not everyone is a software dev with a developer mindset to using these tools. Normal people essentially type the equivalent of "Make me X!"
and complain when the model assumes anything in their underspecified mess of a prompt. There are skills like grill-me that can potentially help these people a lot, but in the end I believe models will just be smart enough to understand your level of knowledge and intent to do this stuff on their own. They are getting much better on pushing back on poor user input already. The problem is that when they double down on hallucinations (very rare nowadays but I still see it happen in enterprise projects with the latest models). So you kind of need to know when to push back on the model as well. But for that you have to be really good at the subject.
I tend to agree. Skill files become less useful as developer skill increases.
As a skilled developer my repetitive instructions are mostly one or two sentence phrases for staring something like a highly-interactive planning session, or a self-supervised implementation session with my preferred setup of implementation and review subagents. I can specify those out by hand, or save a couple keystrokes with a tiny skill file.
But if you are not a software dev you might lack the vocabulary to tell the agent what you want. If you don't know what tenant isolation is, chances are your app will have a broken security model because you can't ask for it, and probably won't think to ask the agent for a security review either. Skills can mitigate a lot here
I still find the superpowers to be very useful, just as a way to manage the feature process, getting you a spec to go with the feature, quizzing you about edge case handling.
It's interesting as grill-me and the other Matt skills are very much positioned to people who would consider themselves as developers. In fact, I'm not sure if people who were completely new to development would have heard of him at all.
>Agent First’s supercluster of real, well-behaved browsers bypasses or solves reCAPTCHA, Cloudflare Turnstile, hCaptcha, & other captchas
Can you share how you prevent spammers from simply using this to do what they usually do, or, if not, how you will stop Cloudflare et al. from marking your browsers as bots and hammering them with captchas? Easy access to the plain web for bots is a pretty tough cat and mouse game and I have been playing it with very short term success at best myself.
The API is default rate-limited, so spamming page requests shouldn’t be possible; not behaving like a bot is the relatively easy part – doing so cost-effectively is harder and what I spend more time optimizing for.
If you have followed news reporting, you probably heard that SamA was touring D.C. to make sure this release went without any regulation hiccups. If anything, they learned how to play the whole politics game - especially after the Anthropic fiasco. And even though all parties involved are terrible choices, more eyes on a potentially civilisation altering product does make me feel minimally better.
More like not ignoring the company owned by one of the president's biggest donors warning said president's administration about your model. Anthropic basically tried to close their eyes to make this go away. Which of course backfired spectacularly and led to one of the biggest release fuck-ups of all time. OpenAI saw that and simply decided to not do it that way. Any sane business would have done the same.
I think you just have to be smart enough that when the administration calls up and says "amazon, the nsa, and half a dozen other companies say we have a problem" your response isn't "well, actually we don't."
Mathematics in the sense of a complete set of axioms can't, but human research into mathematics apparently just needed enough compute to achieve the same output as a high-tier faculty.
Do you never order stuff online from outside of scandinavia? I've used it extensively for over a decade with zero issues and still find it the best payment method for online shopping if you are a consumer. You only need an email to send money everywhere in the world and almost everyone supports it. No need to hand out your actual credit card or banking info to random websites with questionable security. And if you do get scammed, it's trivial to get your money back. So I totally understand some small merchants who don't want to receive from papal, but I don't understand why a consumer would not want it.
There are entire countries where it's not a thing to use PayPal. Way too many people were scammed out of their money by them. In general you don't give your credit card number to random sites - it mostly goes to Stripe and similar services.
These days stripe is perhaps getting dominant, but there was an age before stripe and many small companies still have custom solutions. And how would Paypal scam you out of your money? I've only ever seen this happen to people who did illegal shit and even then you won't lose much unless you also use paypal to store money like a bank for some weird reason. But then you're definitely more of a merchant than a consumer, so it already sucks because of other reasons as mentioned above. If you use it only for sending payments, getting locked out is a non-issue, because there is no money stored there and I can always revoke their access to my bank account.
Well, baltics have access to the EU; Norway only in a limited manner. I noticed this when someone wanted to purchase magic cards and when I looked at the extra bureaucracy I decided to do something else with my time. (Note: this may not have been mega-fair to the customer in Norway, but at that time I did not yet know how insane the EU rules are, they are just a mega-hassle. Since then I simply give notice that I avoid shipping into Norway, Switzerland etc...; and the fault for this really lies in the EU, they make so many trivial things needlessly complicated.)
Other than that, often you will then rely on in-between traders.
USB was developed by a big alliance of major software and hardware vendors. This stuff on the other hand always feels like Anthropic is just scared of dropping out of the news cycle because other companies are constantly releasing models or making them cheaper, while they currently have nothing real to offer themselves. MCP was a complete clusterfuck an announced at a time when OpenAI absolutely dominated with reasoning models while Anthropic's first reasoning model was still months off. The only reason it took off is because noone else bothered to make something better and managers loved it so much that everyone else got dragged on board because of demand.
> The only reason it took off is because noone else bothered to make something better
Yes, and that alone is sufficient reason for them to do it. They did it with MCP. They did it with Skills. OpenAI did it with tools. All of these had many months, sometimes even years, of existing prior art. All of these could've been done much better, too. But "worse is better", first one to get a properly packaged MVP out the door and into spotlight wins, and then everyone is using that.
If OpenAI has set the API standard more or less without even trying just by offering one of the most straightforward ways of interacting with LLMs via http. To the point that they got stuck with this legacy thing now that they and many others keeps reasoning traces private on their servers, which actually hinders future adoption. If they had forced their function calling and structured output API standards just a bit more in response to MCP, they probably could have smothered it in the crib. If this part was standardised more, MCP would offer hardly any extra benefit.
https://en.wikipedia.org/wiki/Imaging_atmospheric_Cherenkov_...
reply