Maybe I interpreted it differently, but playing an RPG where every NPC is essentially its own agent/AI with its own context would be very interesting. In most RPGs, NPCs are very static.
I think they would just pay the taxes but increase their prices so their bottom-line looks about the same. In the end it would be the consumers in Germany that paid the price.
What LLM is it using? Is it something local? Or does it call out? It wasn't obvious from the docs, and I didn't want to dig through all of the code to figure it out. Should probably be clearly stated on the front page.
But the project looks interesting, I have been looking for something similar.
The requests involve relatively long context windows, which require high-quality reasoning.
Only recent high-performance models like GPT-4o or Claude Sonnet are capable of reducing the manual workload for this kind of task.
You posted your opinion without any kind of accompanying argument, and it was also quite unclear what you meant. Whining about being a target and being downvoted is not really going to help your case.
I initially understood your first post as: "Let's not try to make the internet faster"
With this reply, you are clarifying your initial post that was very unclear.
Now I understand it as:
"Let's not try to make existing protocols faster, let's make new protocols instead"
More that if a protocol has met it's limit and you are at a dead end it is better to build a new one from the ground up. Making the internet faster is great but you eventually hit a wall. You need to be creative and come up with better solutions.
In fact our modern network infrastructure returns on designs intended for limited network performance. Our networks are fiber and 5g which are roughly 170,000 times faster and wider since the initial inception of the internet.
Agree, and he even mentions in the article "If you process less than 10k requests per second, you’re not Google nor are you Facebook."
There is a huuuuuge gap between services handling 10k req/s and Google/Facebook.
I think one big upside with JWT that he doesn't mention is that if you have some services geographically distributed, then having decentralized auth with JWTs is quite nice without having to geographically distribute you auth backend system.
So, yes, if you have a monolith or services colocated, or have some kind of monolothic API layer, then no, perhaps JWT does not make sense. But for a lot of distributed services, having JWTs makes perfect sense.
And you don't have to introduce JWT revocation for logout, if you have short token expirations, you can accept the risk of token leakage. If the token is valid for like 30 seconds or 1 minute, you would probably never be able to notice that a token has been leaked anyway.
Did you actually read the conclusion at the end of the article?
"Of course, we can't conclude that you should never export to CSV. If your users are just going to try to find the quickest way to turn your data into CSV anyway, there's no reason why you shouldn't deliver that. But it's a super fragile file format to use for anything serious like data integration between systems, so stick with something that at the very least has a schema and is more efficient to work with."
I very strongly disagree. Having "integration" tests are super powerful, you should be able to test against your interfaces/contracts and not have to spin up an entire environment with all of your hundreds of microservces, building "integrated" tests that are then dependant on the current correctness of the other microservices.
I advocate for not having any integrated environment for automated testing at all. The aim should be to be able to run all tests locally and get a quicker feedback loop.
We have been using all those GCP products and more without any significant problems.
But I do agree, there are some gotchas. PubSub examples: Duplicated messages, shitty DLQ implementation (in my opinion), some developers had improper error handling which lead to to infinite resends of messages (because they nacked message on error), etc..
However, I think the scaling and setup weighs up for all of that. You just need to specify a topic and subscription, and then you don't really have to care about resources or scaling at all, and that is SUPER nice. Also, PubSub is stupidly cheap in comparison to any other similar product, at least that I know of.
