Report spam to spamcop.net

This is the most likely outcome indeed.

Not so, although such claims would benefit from verification, as those are big juicy claims. I repeat not surprising, plausible.

I'd like to read more of it.

Lets pick some popcorn and watch.

I think this is true. The future is with human curator of AI content. Because non-curated AI content will always be slop.

Aggressive moderation was what made it worth compared to the web forum posts that contained very low quality unmoderated, not-curated, and too often broken/deleted and scattered content.

In the late 80's early 90's, the main source of knowledge was Usenet NNTP forums.
In the late 90's to early 2000's it started to shift to forums, problem forums were low quality, did not last

When StackOverflow arrived, it was refreshing, curated content, limited duplicates, made it a quality source. And yes I was also answered "Why would I want to do it this way?", or even "What was I trying to do?". Then I reformulated my question, and admitted it was not clear or not focused. And this alone helped me find a solution. Because what StackOverflow did really well, is force people to write quality questions, questions that would match many other people questions and needs.

As for AI LLMs, I am constantly horrified by their hallucinations, standing up on lies, wrong code, stupid recommendations, proposal of decades obsolete syntax, inability to check/verify/spot their own in/accuracy, invent API methods, loosing criterion during conversation.

The biggest issue with LLMs is that they output what can fit best statistically, regardless of pertinence or accuracy and it has no verification pass, cannot run code portion, cannot debug code portions even from own output and it is wrapped with meaningless excuses, chronic liar promises.

I am very sad to see StackOverflow"s decline.
I am even more worried to be redirected to LLMs as a "replacement".

Sadly so true. At this point the scammed students might seek competent and qualified legal advice on their best options to deal with this situation.

I agree 100% with Everything you said. When you are your own employer, you cannot escape the realty, that you can only get your customers pay for the value you created, and if you fail a something, the consequences are entirely on you.

Working by the clock, means you have a very constant, steady workload. So the amount of accomplished tasks equals the time spent working.

Although:
- Employers will tend to increase the workload within the same time span.
- Employees will push the other way and slow down their work until end of work time.

This method is praised because it simplifies management. It is just laying time tables and checking employees are on their assigned time.

While paying by the accomplished tasks within a defined time frame is more appropriate and fair for both employers and employees.

It has virtues:
- Employers pays proportionally to the value created by the work actually done.
- Employees are motivated to optimize their tasks, do time managements, adjust their work load to fit their own constraints, health, children education, income needs.

The drawback of this, is that it needs more qualified, more capable managers with a good understanding of tasks, capable of managing projects, dispatching tasks and adjusting the workload per employee, perform periodic team synchronizations, adjustments.

To summarize, the clocked work time need time managers, whereas the tasks oriented work requires projects managers.

Sure it'd be safer, unless the blind replacement does not fly off mid-air like that door plug.

As de Gaulle once said: Les cimetières sont remplis de gens irremplaçables cemeteries are full of indispensable people.

You mean I can shelf my Cap’n Crunch whistle and still screw Ma Bell for free?

To address the question: "Where can I find open-source, local-only AI solutions?"

There is a vibrant online community called localllama dedicated to this very topic. Localllama is a great resource for individuals interested in running AI models locally without relying on cloud services. You can explore a variety of models and tools within this community.

One popular option is 'llama.cpp', a high-performance library for running large language models locally. 'llama.cpp' is designed to be efficient and can run on both CPUs and GPUs. For those who prefer a more comprehensive framework, the Hugging Face Transformers library is another excellent choice. It supports a wide range of models and provides extensive documentation and community support.

While a GPU is recommended for optimal performance and faster inference times, it is possible to run these models on a CPU. However, be prepared for significantly longer processing times if you choose to use a CPU.