Wow. Does the US have this type of agreement with any other country?
2,406 karma · joined February 24, 2012
This is what I think is the solution - I've tried it a few times with good results.
1. We need our own project or task plan. A simple forward-thinking TODO list is good enough. This needs to be prepared before the day begins. Daily.
2. Focus and limit scope. Focusing on doing ONLY what's on the list is the discipline I (we) lack and which we need badly.
3. Side quests are killing us. Allowing ourselves to veer off the planned daily tasks should require a complete away-from-the-keyboard planning session with pen and paper. Print it out but plan it away from the LLM. Use the LLM to summarise the problem, by all means, but do the thinking away from the LLM. The current TODO list needs to be the main input of such a deliberation. The output is the new TODO list. The objective of this away-from-keyboard exercise is simply to maintain control of the next steps.
On days where I worked with my TODO lists, I clocked out and joined the family for activities, slept sooner and better. On days I lost control of that list, the exhaustion mounted almost on the same day.
Make a finite list for the day, tick it off and go spend time with the missus, the kids or the dog. 2c.
This is, as you say, exactly the paradox of having ourselves assisted by LLMs. We are no different from our cats. Except imagine the cats start making the laser lights for themselves.
That means folks using LLMs starting off with 300 bits KNOW that they lack the full payload of information to transfer to you. IOW, they know they need to transfer much more than 300, so they use LLM to fill those gaps. That's the crux of the slop universe out there. Folks are using LLMs for the 700 bits on top of their 300 bits and passing off the full 1000 bits as their own.
I echo the writer's sentiment. "I don't want to read the clanker's 700 bits. I want only your synthesis." (I can get the clanker to generate those 700 myself. Unless ... unless this whole LLM slop market is all about saving you the time to get an LLM to generate those 700!)
I believe the universal answer is: incompetent malicious actors are now capable too. Which implies the pool from which to draw the intersection between capable and malicious has grown. (That's my reading)
In the “enterprise scene”, this sort of purge strategy is likely linked to a big percentage of money wasted and sometimes total failure.
Why do managers keep making this mistake? “Legacy” is only a bad word in IT. :-(
(I think HN should consider creating TLDR type of comments for upvoting separately from commentary. @Dan incentivise with different karma credits?)
Was the code for the threading model written by hand or was it translated from the WIP threading model the human PG team is busy with as part of the 2028 roadmap?
If they needed permissions for conversion, they'd have made a specific mention of that and thereafter confirmed legal ownership.
Is if a one thread per request model?
Does it support async await type of architecture?
Impressive.