I am working on Grog, the “grug-brained” alternative to Bazel. Bazel has a very steep learning curve and is pretty much overkill for most medium-sized teams. Grog already powers all of our internal mono-repo CI and is a lot more fun to work with.
I think that there is a huge leap between using the Kurds to do your fighting (like in Syria) and what he claims if you look at the map that he pulls up in that section.
He makes quite a few quite controversial assertions without evidence or even bothering to explain them. At the top of my head and paraphrasing:
- "The conflict is a game of chicken." Only in the vaguest sense in which any war or confrontation could be called that.
- "The USA is not equipped to handle a war against drones and fanatics". Idk they seem more capable than any other nation save Ukraine given that those two things have been a major feature of their recent wars.
- "Countries that are poor have more energy and are more cohesive". This is just demonstrably false, but he does not bother to explain why he thinks this.
- "The US wants to break Iran into ethno-states that compete for water until they are all dead." He even admits that this is pure conjecture and hand-waves this as the game-theoretic "optimal" strategy which is completely bonkers.
I do share his negative sentiment and outlook about the war, but there are way better critics that don't resort to this type of intellectual laziness.
I think we should hold people to a higher standard especially when they talk with this much confidence and present themselves as a professor. The issue is that "Game Theory" is a technical term so there really only is an "Academic" sense otherwise it's just a marketing term for whatever ideas you want to push.
While I agree with some of his sentiments the entire video reeks of half-baked conspiratorial thinking and shallow engagement with the facts.
A quick tell is that the video's title includes "Game Theory", while only referencing game-theoretical concepts twice in an off-hand comment. In both instances the usage is plainly wrong.
In general, he loves making big assertions without backing them up with evidence or explanations that go beyond hand-wavy examples.
Sops can natively handle .env files. All you need to apply them to your process is a small wrapper script that sources the decrypted file before invoking your command.
We just recently adopted this and it's crazy to me how I spent years just copying around gitignored .env files and sharing 1password links. Highly underrated tool.
I think you can come to a balanced view here where you acknowledge that Feynman was overhyped posthumously while maintaining that he was an exceptional physicist with some personal flaws. That's precisely the point of the video.
It's less axe grinding and more counter-acting an inaccurate narrative.
Even worse so: Why does he not simply ask these people? What is it with this trend of sneering at expert decisions without even doing the bare minimum of engaging with them?
I am working on Grog, the “grug-brained” alternative to Bazel. A mono-repo build tool where all you do is provide your build commands and interdependencies and the Grog will run everything in parallel with aggressive caching.
I am working on Grog, the “grug-brained” alternative to Bazel. A mono-repo build tool where all you do is provide your build commands and interdependencies and the Grog will run everything in parallel while caching as much as possible.
I fully agree. I was also thinking about docker or symlinks, but it seemed hard to design the API without any actual user feedback. Pants environments [0] look interesting.
Very cool that you are also recognizing this issue and working on it. I sent you an email in case you want to exchange further.
There is a space between the two types of repositories you are describing. One where you just have enough tools/langs that a single-language setup does not cut it for you anymore, but investing all that effort into Bazel does not seem worth it yet. That is the gap that Grog is meant to fill.
https://grog.build/why-grog/