I believe the Haiku people open sourced after their company was down, I saw it as an act of good faith towards the people who trusted them, I say try open sourcing it, who knows what the future holds...
I wonder who else has to deal with ed also... recently I had to connect to an ancient system where vi was not available, I had to write my own editor, so whoever needs an editor for an ancient system, ping me (it is not too fancy).
amazing work by the creators of this software and by the researchers, you have my full respect guys. those are the real engineers!
Two years ago I worked for a sw company where we had to post RUST tutorials on the Internet.
for me, one of the main concerns was that any rust code I post on the Internet, has to be validated by the rust compiler before I publish it,
I was using artificial intelligence at the time I think it must have been early February 2023 or maybe March 2023, at the time, the large Language models were not as good as they are today, so we actually spent the time with my boss to actually sit down and write an automatic-validator rust tutorial programmer.
It takes as an input an idea for a rust tutorial, it writes the computer program, and then it automatically validates the computer program against the compiler.
and if there is a ERROR it will automatically try and fix the Rust Error, and then (and only then) when everything is validated and executed successfully then we can say we have a finished Rust tutorial.
so, what this means is that the artificial intelligence will not kill the internet, because there are (probably) many people like me (actually plenty of people like me) that actually care about the readers and will validate everything they publish.
I also understand that there is a lot of people that will not validate and I think this is sad and there is not a lot we can do about it right now.
the only thing that comes to my mind is to write a program that will automatically try to validate tutorial so that you can save time.
But the main point i'm trying to make is not that "I made a thing", not the main point at all, the trying to make is that "people care about other people", people care about their readers, and people will continue to make efforts to validate what they publish on the internet.
so dear reader from the internet please do not worry, instead I will suggest you focus your efforts on the automation of the validation of information generated by the artificial intelligence, and the creation of oracles that validate the results of the AI.
I hope this idea will be met with optimism rather than pessimism.
I hope I sincerely hope that this idea will be met with a wider adoption, if anybody is interested in discussing this for their feel free to message me at [email protected]
Perhaps offering people education, is a way to allow them to access those jobs you describe, perhaps education in robotics? idk, I think this take is kinda gloomy.
for me rescuetime.com and hubstaff have helped in terms of productivity, I also have written a couple tools myself to make myself more productive over the years, vim also comes to mind, there are several ways to be more productive, one just has to care enough.
perhaps, with the advent of AI, one will be able to convert a current book into a more detailed book, and also a current book into a smaller book, so maybe this idea is even easier to implement than 4 years or so ago, before chatgpt (but after summarizers, which prompted the idea in my head).
I would humbly appreciate any feedback on the concept of parametric books, for now, it's just an idea, but it's a free one, anyone is free to implement it.
if you go look in the source code of the html you can see some <ASP> tags probably not rendered by the browser so in order to READ this document you have to view-source (it doesn't look that different, but it makes more sense).
“This program, print the numbers from 1 to 10 in python”
In order to run my simple program just copy paste it into your LLM, of choice and execute it
Notice that the program does what it supposed to do it is deterministic unless you use something like gpt2 and it solves the problem that I aim to solve