Cort3z

Don’t sell you company to Microsoft then

"This item is not available for purchase in your region"

This is such a bummer. I live in the European Economic Area (EEA)

If this all works, and becomes widely available, I wonder if people would get a lot of health anxiety. Now they can see stuff that is normal, but strange or unexpected.

I’m no doctor by any means, but what if, as an example, an organ fluctuates in size, or composition, naturally. A medical professional would know these things, but a random person off the street might get stressed out and start to panic, or perhaps overcompensate with their diet or something.

I think more data is generally good, but data without context or insight can be problematic.

I really like it. Here is my 2 cent:

- the gravity is weird in my opinion. There is basically a gravity going down the screen. I would have expected there to be some "pull" towards the planets. I get why though, you try to prevent mega-long straight shots upward. Perhaps experiment with some drag, where the ball slows down over time. Or perhaps have the walls be gravity points, pulling the ball towards them.

- i would add a long- press to restart immediately, so restarting is faster.

I would add long-press to restart

I can see Apple doing something similar in the future. Just like how they are pivoting away from flat design in their ui, perhaps the time is ready for a more "organic" design. Wonder what marketing term they would use

I believe this stuff is straight up illegal in the EU. And also: "A computer can never be held accountable, therefore a computer must never make a management decision.” - IBM Training Manual, 1979[1]

1: https://www.ibm.com/think/insights/ai-decision-making-where-...

So, are we way into diminishing returns for these models at this point? If so, I think we can calculate when it will be available at home. Given this requires a GB200 NVL72 which has about 1,440 PFLOPS, the current 5090 chip has about 1,676 TFLOPS, so about a 1000x scale-up to the GB200. If we can assume Moores law, which might be broken, but still. We are looking at log2(1000) = 9.96, or about 10 years.

I agree, but if there were a common place where everyone did the registration once, it might be less of a problem. Something similar to the «verified» accounts on twitter etc.

I can imagine GitHub becoming this filter somehow

I wonder if there is a way to blanket prevent these types of problems.

Possible solutions I can think of:

- Require an account with a paid service. Fix = require money - Require an account verified with real ID/passport etc. Fix = link to real person - Automated reply system to "waste tokens" if it is an AI that is responding. Fix is increased cost of spammer. - Have some kind of "vetting system" where you get on an allowed list to report these types of things. Seems not good to me, but perhaps there is something in it.

I wonder how much open source code is lost because maintainers must deal with this type of thing versus the "good" that AI can bring in productivity.

Are there a total ad time percentage metric in this law too, or will they simply be watching many more smaller ads?

I fairly recently tried to ask a question on SO because the LLMs did not work for that domain. I’m no beginner to SO, having some 13k points from many questions and answers. I made, in my opinion, a good question, referenced my previous attempts, clearly stating my problem and what I tried to do. Almost immediately after posting I got downvoted, no comments, a close- suggestions etc. A similar thing happened the last two times I tried this too. I’m not sure what is going on over there now, but whatever that site was many years ago, it isn’t any more. It’s s shame, because it was such a great thing, but now I am disincentivized to use it because I lose points each time I tip my toes back in.

Most problems in most distros are solvable with enough knowhow and enough research time. That time investment is not always worth it in a commercial context. It is a fairly known amount of time to reinstall a distro and get back up and running, and an unknown amount of time to fix an exotic Linux problem.

Many seem to be interested in knowing my distro. I’m not interested in throwing shade on a distro in particular, but it is one of the bigger and well known ones.

I’m not trying to cast shade on a distribution in particular. I’m using one of the big ones. If you have not had any of these problems, great!

Shameless self promotion, but I 80% vibe- coded a pip package for interfacing with LLMs right from the terminal. ‘pip install lask’. It has helped me a lot since it works from the terminal regardless of what the graphics drivers are doing.

I have been working professionally on Linux for many years. But about once a year I have to reinstall the os because it craps out for various reasons. The same story goes for most of my team, but for some reason they seem ok with this. My issue with Linux is this: I don’t feel like a consumer, but a janitor. I don’t want this. Yes you can do whatever you want, but I don’t want to do those things. I want to write code and play games, not maintain the intricacies of a running computer.

For a server there is no better choice than Linux, but for my desktop/laptop, I find other alternatives better. Perhaps I haven’t found «the right distro», if so let me know, but until Linux is as low maintenance as windows or macos, it will be for those with an interest in doing that maintenance.

I realize I have a love-hate relationship with Linux. It is perfect, but flawed.

I love semantic html. It feels so much more modern and efficient than all of these fashion-driven front end tools.

Just implement exponential backoff for slow requests logging, or some other heuristic, to control it. I definitely agree it is a concern though.

Does anyone have a link to the source of the statement without a paywall in front? I could not see any reference to this 150M$ anywhere.

Makes some sense that the owner can't just eavesdrop on every conversation on the platform. That is very illegal many places in the world.

It is free, as I understand it, not 200k per year.

Wonder how the ROI on this is going to be for salesforce.

They keep shrinking in the wrong dimentions.

Their selling point is to be a more open version of what cursor has. So the alternative is to use a full llm. So it is 10s+ 10s vs 10s+ 300ms

As far as i understand, this is not +-300ms. It is 300ms vs. 10 sec or something. That is a huge difference. I personally find the time to wait for these larger models a limiting factor. It’s also probably a resource waste for fairly simple task like this. (Compared to the general function approximation of the llms)

But I honestly feel like the task of smartly applying edits falls somewhat within traditional coding tasks. What about it is so difficult it could not be done with a smart diffing algorithm?

What are they protecting against? Honestly. LLMs should probably have an age limit, and then, if you are above, you should be adult enough to understand what this is and how it can be used.

To me, it seems like they only protect against bad press

Kind of serious question. Do we have any alternatives to html? If not, why? It’s essentially all html. Yes, browser will render svg/pdf/md and so on, but as far as I can tell, it’s not what I consider "real web" (links to other documents, support for styling, shared resources, scripting, and so on ).

I would have loved for there to be a json based format, or perhaps yaml, as an alternative to the xml- based stuff we have today.

I’m a noob to this subject. How can a build be non-reproducible? By that, I mean, what part of the build process could return non-deterministic output? Are people putting timestamps into the build and stuff like that?

I assume it would be possible to make yourself with the OpenAI api together with a locally run voice model to only detect the activation word. There might be of the shelf solutions for this, but I am not aware of any.

I usually find Claude to be my favourite flavor of LLMs, but I still pay for ChatGPT because their voice offering is so great! I regularly use it as an "expert on the side" when I do other things, like doing bike repairs. I ask it things like "how do I find the min/max adjustments on my particular flavor of front derailleur", or when cooking, and my hands are dirty, I can ask stuff like "how much X do I usually need for Y people", and so on. The hands-off feature is so great when my hands are literally busy doing some other thing.

I really wish Claude had something similar.