This sort of system is only useful because LLMs are limited. If they can be told to farm certain categories of step to middleware, then when they encounter such a step, they should farm out the request. I've found, with trying engineering problems, that LLMs consume a lot of steps finding out what to collect, with a risk of hallucination. That's exactly the sort of thing that can be farmed out.

According to both Claude and ChatGPT, that sort of process is the focus of a lot of research, right now, although apparently it's not actually been tried with reasoners.

Yeah, that would count.

Specifically, with something like Protege, I can define how things relate, so I could set up an ontology of cameras, with digital and film as subtypes, where lenses are a component of cameras, films are a component of film cameras, and pixel count is a property of digital cameras.

The reasoner could then tell you about how the bits relate, but a SPARQL search could also search for all records in a database pertaining specifically to any of these parameters.

At least some search engines let you do this kind of search. On those, you can specifically say "I want to find all pages referencing a book where this is specifically identified as the title, it isn't referencing something else, and that specifically references an email address".

So, in principle, nothing would stop Claude or ChatGPT saying "I need to find out about relationships involving pizza and cheese", and the reasoner could tell it.

That means if you're designing a new project, you don't use any tokens for stuff that's project-related but not important right now, and you use one step no matter how indirect the relationship.

This would seem to greatly reduce hallucination risks and keeps stuff focussed.

What you're suggesting can bolt directly onto this, so Protege acts as a relationship manager and your add-on improves memory.

This triplet would seem to turn AI from a fun toy into a very powerful system.

AI makes up things 100% of the time.

That's just a feature of Large Language Models.

Game-playing AI instead tries to figure out the "best" solution based on what inputs it has. If it tries making something up (e.g. attempts an illegal move in chess), the game server will catch it and simply reject the move, while other AIs will have already looked at more valid options.

A hallucination is when you don't like the answer.

No, a hallucination is the AI generating something completely false, as demonstrated by yet another lawyer using an AI text generator to invent fake cases and passing them off as real.

Why are they monitoring syscalls?

The correct solution is surely to use the Linux Kernel Security Module mechanism, as you can then monitor system functions, regardless of how they are accessed. All system functions, not just the ones that have provision for tracepoints.

For something like security software, you want the greatest flexibility for the least effort, and Linux allows you to do just that.

Because it's fine-grained, security companies can then pick and choose what to regard or disregard, giving them plenty of scope for varying the level of detail. And because the LSM allows services yo be denied, there's a easy way for the software to stop certain attacks.

But I guess that the obvious and most functional approach would mean that the vendors would have to write a worthwhile product.

The issue with DOS is that the development tools basically ceased development around 1995.

DJGPP lasted until 2000, which is about one year after the source code of Quake was released. IIRC, it could compile the game without difficulty. And... it was an upgrade compared to pirated versions of Borland C++, Qbasic, etc.

but software written for newer tools is usually not compatible with tools from 1995.

Software written for newer tools won't be comparable with MS-DOS, considering that they usually rely on graphics, network, and many other features that expect a modern operating system.

If you have MS-DOS, you need specific libraries, or you do stuff manually. That is, manually call the interrupt table, make hooks, and so on. Also, hope that you used the right functions to unhook the interrupt table in case the game crashes, otherwise you also crash the system and need to reboot.

After my first job, no one has ever asked about my degree. They ask me about my work.

We have a society where everyone wants to be right; we need a society where everyone wants to be correct.

and plenty that don't.

I think you will find that's a non issue. Since is solvable.

I"m in a union, and I am a white collar employees. COMputer programmer and data analysis. Our organization also has a union for business analyst.
There have been teacher unions for decades.

A union is jsut a group of people with a contract around working right, policy, and wages. Can be any group.

There is no "liberal indoctrination", it jsut appears that way because conservative consider anything counter to their myopic view liberal.

In order to have a meritocracy, all people must have the same start. SOme oney, same education opportunities.
Of course, there is no real definition to meritocracy either.

Remember, the term 'meritocracy' was created as a pejoratively.

IT also relies on whos merit? Bezos is a billionaire, but it was from him, it was his workers. Should the people who created his site, the engineers the built it for him have the real merit?

https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2F...

Some people hear a word and then jsut assume it's good without actually reading up, and that needs to stop.

" Fewer people die of heat than die of cold. "
More people will die from the heat. See: dew Point.

"Longer growing season."
lol, excessive heat is bad for crops. And some areas of the earth are already losing farming capacity, not higher yields. major commodity crops like corn, rice, and oats are starting to experience reduced yields due to heat stress and changes in water availability.

" Bring on the CO2."
We produce more CO2 then the plants can handle. BTW, too much CO2 is bad for plants. Just like to much O2 is bad for people.

"We can handle the heat. We will be fine."
Why can't you people grasp the simple fact that as we keep producing more greenhouse gasses, the heat will keep going up? WHy are you so ignorant you think civilization can stand that level of heat growth?

You don't understand what accuracy mean from a science standpoint. But keep smugly posting ignorant post, since that seems the only way you can pretend you have any meaning and matter.