Most British people don’t respond well to attempts to cow them. They refuse to be cowed! To allow oneself to be cowed would be cowardly. To do so would have one accused of cowardice, of being a coward.
I meant the social media platforms. If HN charged a fee for posting and commenting it doesn’t matter how cheap the AI models are, it’d be punishingly expensive to sling slop.
Invert the economics. Right now there is value in posting LLM generated content that is more than the cost of using the model.
If platforms had a subscription model that you had to pay for in order to do more than just read comments, there’d be a lot less LLM content. There would also be a lot less of all content. But maybe that’s the price you pay (literally) to get rid of AI slop.
If we just put enough effort in and write the right spec/prompt/design then the programmers/llms/plug compatible coding units will produce the correct output first time!
Closing feedback loops. That’s the whole thing. WE Deming would have recognised agile (little a) as a PDCA system and approved.
The standard pattern to avoid select for update (which can cause poor performance under load) is to use optimistic concurrency control.
Add a numeric version column to the table being updated, read and increment it in the application layer and use the value you saw as part of the where clause in the update statement. If you see ‘0 rows updated’ it means you were beaten in a race and should replay the operation.
But you’re not getting slower and slower for every car. Lets say 100 cars pull in front of you, and let’s be say each car adds 5 metres of space, so you have 500 metres of ‘lost’ space to regain.
You do indeed use JOINS. The goal is to retrieve exactly the data you require in a single query. Then you get the DB to `EXPLAIN VERBOSE` or similar and ensure that full table scans aren't happening and that you have indexed the columns the query is being filtered on.
You had me until the sweeping generalisation. I can picture things clearly but only deep inside my head. I never (awake) see anything in front of my eyes that isn’t physically there.
Thread.sleep on a platform thread takes that thread out of action until the sleep ends. If an executor with 10 threads got 10 tasks that all called Thread.sleep(1000000) then it can run no more tasks until one of the threads wakes up.
When a virtual thread sleeps or blocks on IO it is unhooked from the underlying platform thread so another virtual thread can run. You can have an almost unlimited number of virtual threads multiplexed over a small number of carrier (platform) threads. Hence M:N