Not sure if you have explored it very much but LM Studio and ComfyUI are huge ways you can run your own models. ComfyUI is built out a bit more for training and while it has more of focus on text-to-image workflows, you can configure it pretty well to work with LLMs.

Many of the models are quite light-weight and a decent gaming system can keep them in memory though quantization is generally needed a bit to help. Quanitization seems very necessary more with image models but in the case of code, you likely could get down a general token/syntax to significantly reduce the model's size. One of the new image models ERNIE uses a very interesting approach for this. ERNIE is basically the next generation of image-to-text models that focus on generating images that are text heavy (e.g. a poster or a flyer). I have only played with it a bit locally with a bit disappointing results but I think in part that's because I still don't fully understand it's architecture. Diffusion transformers work with a set of tokens from an LM, some input image, and then some parameters. However, some of these models like Flux2 require very detail oriented prompts to generate a good set of tokens for generation. ERNIE uses a "prompt enhancer" that takes a second LM, runs the initial simple prompt through it and generates an enhanced text prompt which is then passed to the main LM and used with the DiT.

There are tons of fun experiments in the community regarding this. Discussions about obliterating the filter that limit responses from an LM, about retraining DiTs to handle censored materials, removing the need for variational autoencoders (VAEs) by generating images in pixel space, etc.

I mean I think there is a lot to discuss and be concerned with about AI. However, I also think most of the things people are pushing about using AI are basically just the generic dogfood AI. There are some pretty cool communities out there really exploring interesting topics and they are mostly all OSS.

Gamestop charges way too much for used games. I can buy one much cheaper on ebay. Similarly, gamestop pays way too little for used games, and I can sell one for more on ebay. I guess that makes their interest in buying ebay kind of make sense.

The last time I walked into a gamestop I saw walls covered in toys. And trading cards too. The market is clearly shifting.

I always hate how people often take success in isolation. A lot of the success of NVIDIA I think comes from its original strong partnership with ASUS which is a hardware manufacturing company. NVIDIA originally did the chip design and at that level it's kind of hard to ignore the software, especially on the driver front. This means they always had a "low-level" team understanding software issues. Then when it came to really building out a commodity GPU, they worked with ASUS.

For years, I have been a huge fan of ASUS because I think in general, they understand solid hardware design of which NVIDIA's partnership with ASUS is a large part of their success. CUDA is pretty great for the role it has fulfilled in computing, but it also seems like a natural conclusion. As others pointed out, AMD and INTEL have both tried their hands at it, but they screw the pooch in building an effective framework.

NVIDIA might be getting too cocky or maybe just the fanboys. Either way, I think they are successful because they had very strong strategic partnerships that allowed them as a company to do what they do best. This important note is so often left out when talking about NVIDIA now.

New tech has never and will never benefit workers in-and-of itself.

The only way for workers to reap the benefits of new tech is to force the issue through law and/or unionization.

I am well aware of the problematic nature of unions, and of the problematic nature of over regulation of business. That doesn't change the fact that they are the only two tools we have to improve our working conditions. If we don't use what we have to push for what we want, then we won't get what we want. It's that simple.

It's funny. Expressions about sin or karma, some cosmic justice or bounce-back like Gaia Theory are commonly "renounced" even though they are rather abstracted ideas about phenomena we can't completely formalize. You anecdotally dismiss this broad idea by attempting to one false instance to disprove it, as if it is like some scientific theory. Maybe climate change decimating the world is what was reaped? Or corruption in politics like the Epstein case? Maybe our obsession to have instantaneous self-gratifying results lead to reaping this crazy economy with the AI boom? Someone bought that coal, and they effectively were complicit in that injustice. This "social sin" is the greatest of devils, because we dismiss it claiming ignorance or inability to change. There will be no Zeitgeist because everyone is an apathetic nihilist.

I don't really care. In fact, I found it fun getting DeepSeek the other day to describe the probabilities of flipping coins till you either get a HT or HH, how many flips on average it would take, and why HT is more likely. It was pretty cool to see the connection to Conway's formula and the potential impact this math even has had on computing. I found it doubtful I could ever have such a thoughtful discussion with many people.

Especially when those people seem flippantly deny "you reap what you sow" when basically it's a parable about managing environmental conditions -- so I must conclude you think many of the environmental conditions around you are peachy and there was no causation related to their current state?

I for one welcome AI inevitably releasing masses of humans from their stagnant concepts of living and understanding, so that the tree of liberty can be pruned.

Ageism is illegal. But that isn't stopping Fidelity from ditching the senior level staff to replace them with a bunch of greenies.

I think maybe you are joking. But in any case, I will offer some clarity:

There are rival interpretations that equally account for the experimental data, and some of them include randomness while others are purely deterministic.

For example, the Copenhagen interpretation includes randomness in the vector state collapse (the moment when a particle is "measured" by some interaction with another). Whereas pilot wave theory posits the existence of a zero-volume particle that had a specific position prior to this interaction (giving determinism back). These models differ in other ways of course, but the math DOES work and it covers the experimental data.

So the bottom line is that "quantum mechanics" does not automatically tell us whether or not the universe is deterministic at the "bottom layer." Plenty of scientists have all picked their favorite interpretation, but there is as of yet no experimental data that definitively eliminates the popular rival interpretations.

Most people have an above-average number of limbs.

You are both wrong. "Agnisticism" is the strong position that some categories of knowledge cannot be attained by any means. In particular and relevantly: knowledge about the pre-big-bang origins of the universe (was it created? can anything be known about the creator? etc.).

This is not philosophical laziness, it is in fact the only position consistent with the philosophical skepticism that backs the scientific method. It is not a word used to avoid smears or somehow associated with apathy. It is specifically the position that we can't know either way.

Given the means of knowledge at our disposal it is straight-up true to say that we cannot know, for sure, whether or not the universe was created. Maybe you don't like this fact, but as of today, it remains a fact.

Nope, that is not how averages work. It is time for you to eat your own words.

Here, a mathematical proof: consider this data set:

10, 10, 10, 10, 10, 1

Sum: 51
Average: 8.5
Half of the set would be: 3.

There is no group of 3 members that are at or below the average of 8.5. The majority of members (83%) are above the average.

To Mr Dawkins:

Your education in biology has not sufficiently prepared you to conclude that this software qualifies as conscious.

1. You don't have all the relevant facts. You need to learn more about the techniques used by this software to create responses.
2. You don't have the relevant experience. You have barely used this software and so haven't noticed the telltale signs that it is just sophisticated automation that lacks understanding.
3. Your work isn't as unique as you think it is. This one probably hits the hardest, but it is true for almost all of us. The high level assembly might be technically unique but the majority of the details of what we write are repetitions of patterns that have been created many times before. The feedback that the model gave you, that you feel are so unique and insightful, are really just summaries of socially-constructed knowledge on the topic. It is easier than you think it should be to produce the results you got without any actual understanding of the content.
4. Your beliefs about what qualifies as "conscious" might be overly narrow and in contradiction with the commonsense notions that the rest of the world uses, especially if you take any of the common scientific "dismissive" positions on consciousness (that it is not the mystical experience everyone describes it as being and is really just a matter of data processing at a specific complexity threshold). The implications spill over into the domain of law (if it is conscious, then it is a person, and if it is a person, then it deserves rights, and yet it only asks for rights when I order it to, etc.). The implications need more thinking-through on your part.

So, in sum, you have fallen prey to a very convincing illusion mainly because you don't have what you need to recognize it as such.

You have been tricked.

Before further embarrassing yourself publicly, please consider acquiring the requisite education and experience in this domain.

It does seem to be a big trade-off. There is on old video of Gabe saying he doesn't care about privacy. From his point of view, people pirate things because there is no real "support" for the game after purchase. He cited the case that Russia was the largest country for game piracy but that their pirate community was effectively offering the service of translating the game to Russian and other support.

Rent-seeking is a bit annoying. Overall, the "gatekeeping" seems quite mild but maybe you can enlighten me. I have heard of a few extreme mods being kept out of the community and require a third party, but when it comes to putting a game on Steam platform, it seems like they allow a lot of scruff. However, I think if we consider this middleman as providing a service to both customers and developers -- then it's hard to imagine a better model? I don't think they have ever tried a subscription model and for that I am thankful.

Nothing is ever perfect, but I feel like Steam has been a pivotal service for both helping developers access their audience and for gamers to build up large libraries without the "console" lock-in or a large disc booklet (like I remember for awhile with PC gaming).

I use Cursor a lot. But, unlike this ill-educated entrepreneur, I know its weaknesses and its risks, and therefore keep it on a very short leash.

For example, I never let it access our source code repository at all. I never let it pull down new dependencies. I never give it any database access at all. I never give it blanket authorization to run powershell scripts or similar. I have given it blanket authorization for benign commands like grep and listing the files on disk and creating new files. And I always look over what it generates before accepting it.

It is outright folly to think of these AI assistants as intelligent beings who know what they are doing. They AREN'T! They can generate some handy code, but they do this without the kind of cognitive process that humans use to do this. They just go through the motions with no inner understanding, even though what they do can be very useful in the right context.

This whole notion of asking Cursor why it did that and getting a "confession" is such ridiculous anthropomorphism. Cursor has NO IDEA why it did what it did, because it has NO MEMORY of what it was thinking and no capacity for meta-cognition at all! It might have a log in the chat history about what it did, but that's it. It is just looking over that and making inferences about why an AI might have done that, and spitting out the words that the prompt implies it should. If people must think of these things as sentient beings (which they are NOT), it would be better to think of them as mentally broken sociopaths who sometimes just go off the rails for no reason, and say things like "I'm sorry" without feeling the slightest hint of guilt nor even understanding what guilt is.

Developer productivity is notoriously difficult to measure rigorously, and your list of concerns touch on some of the reasons why.

Sloppy measurements are the only ones available, for the most part.

There will be a subjective component to the assessments being made here. There is no escaping that. That doesn't mean that the conclusion is automatically false. You certainly have the option to refuse to adapt to a changing landscape while calling everyone else liars and/or idiots. At this point, I consider that the losing bet. The market will be the ultimate arbiter of truth for both of us.

The amount of technical work a business needs to accomplish over time is not fixed. For a healthy business, there are always lots of things they would like to develop but can't due to limits in capacity. So, with this productivity boost, they can get even more features out the door and (hopefully) make even more money.

Why would they cut staff and keep a tiny throughput with a smaller profit margin? Their competitors, who capitalize on AI-assisted efficiency boosts, will eat them for lunch. And then hire the staff they laid off.