[wkoszek]

His project is cool. Students could use it as an example of porting code. Companies could switch to it, if it works. There are hundreds of reasons why people may want it, so it's awesome he's publishing one. Something needs to be done to get interest and then adoption.

[fg137]

Let me just point out:

The project is not cool. This is not a new idea, and there is nothing special.

Students won't use it as an example of porting code. I am not aware "porting code" is part of standard software engineering or computer science curriculum. That's not the kind of thing being taught in schools.

Companies won't switch to it unless their CTOs are either insane or incompetent.

[wkoszek]

If someone did this before with PG, provide links. Practical pragmatic example where I think this project is cool: process vs thread. There were members of the PG core team that wanted to explore it, and members who said it'd tank a project. pgrust is an amazing experimentation ground for it.

I published similar project here: www.emuko.dev - emulator for RISC-V. This one turned out to be 3x as slow as QEMU for example.

re: CTOs - if the improvement is 3% nobody of course will look at it. If the improvement is 30%, it'll be too big for big players to ignore, so as a CTO you'll be tasked with trying it out. It's really a matter of whether this thing is safe and secure, losing data or has a trojan. If the authors can prove it's all valid working code etc., it'll be a viable project.

[fg137]

It's a matter of whether this project will provide bug fixes, will continue to exist next month and whether the company will end up scrambling to replace it. No CTO cares about (very) questionable "improvements". They care about if it works and will continue to work.

Sorry, as someone who has been involved in many of these decisions, I don't think you understand how any of this actually works in the real world.

[wkoszek]

I mean.. they made it 2 weeks ago. Clearly the time isn't now. Why be so negative immediately.

[quantummagic]

There are people who see red whenever AI is involved. We've been automating away jobs for years, but now that we're starting to automate our own jobs, all of a sudden it's a moral outrage. People are just being emotional and not thinking straight. It will take a generation or two for it all to be accepted without drama.

[wkoszek]

There's so much to do in databases, and it's so hard to get right and test is all, I think we'll all have 10x the jobs. The easy stuff will go away, but to e.g.: test the new DB engine and ensure there aren't any exploits -> let's just work on it. But it's not that the introduction of compilers made engineers obsolete.