I am just using a more restricted definition of distributed that you are. In a more generic sense git is also distributed. I was requiring online synchronization in my definition.

Git is not a decentralized database. It is completely independent copies of the database which can communicate to each other. That is not the same thing. A sharded database is decentralized. Git is replicated.

Differentiate decentralized development from git itself being decentralized.

This centralized vs decentralized make no sense. Run multiple copies of Scion and it will behave just like multiple copies of git. Data in git is not spread out in as decentralized way, everyone has a complete clone if they asked for it; just like running multiple copies of Scion. The way you control this is with who has push permissions. Git is only decentralized in the sense that it allows everyone to share in a controlled way.

The kernel development flow under Linus is different. It does not use 'git push'. Instead Linus pulls from his lieutenants, locally tests and inspects. And then he finally pushes to master himself. AFAIK only Linus and GregKH can push to the master Linux repo. Allowing AI to do uncontrolled pushes into a master repository is asking for trouble. There are likely stages in their development process which should be converted from push to pull steps.

My interpretation of their problem is that their CI system was doing git clone and cloning the entire repository, then checking out the head. I agree, the performance of doing that sucks. But they don't need to do that, instead clone with --depth == 1. Doing that copies over the compressed file and then the local check out decompresses them. This is a common problem in CI systems done by people unfamiliar with git details.

And good luck beating git's performance.

They don't know how to use git because they are letting claude do everything. Just use '--depth 1' in the CI clones and now git does exactly what they built.

The publishing industry is demanding this via lawsuits. Their position: you can not make copies. So if the book is turned into electronic form, then the physical copy must be destroyed. Only one instance of the text can exist. This has been a fight now in multiple lawsuits where AI companies were forced into paying for a physical copy of every book they scanned.

AI can easily write the lean for any proofs they do. The problem is that lean is missing large parts of established mathematics.
My current lean project has to cite all of these externally due to missing lean support.

    Functional equations / means
    - lit_aczel_1948 — Aczél: symmetric + homogeneous means power/quasi-arithmetic mean (the CES forcing)
    - lit_aczel_1966_weighted — weighted Aczél characterization (weighted means)

    Fixed-point / topology
    - lit_brouwer_1911 — Brouwer fixed point
    - lit_cellina_approximate_selection_1969 — approximate selection (closed-graph correspondences); with Brouwer
      Kakutani
    - lit_glicksberg_1952 — Glicksberg fixed point (infinite/Bayesian games)
    - lit_berge_maximum_theorem_1959 — Berge maximum theorem (upper-hemicontinuity)

    Probability / large deviations
    - lit_sanov_1957 — Sanov / method of types (large-deviation rate = KL)
    - lit_fisher_tippett_gnedenko_1928 — extreme-value theorem (GEV limit laws)
    - lit_kolmogorov_1931_fokker_planck — Fokker–Planck diffusion equation

    Optimal transport / matching
    - lit_sinkhorn_1967 — Sinkhorn matrix scaling (entropic OT)
    - lit_lp_strong_duality_1951 — LP / transportation strong duality (Gale–Kuhn–Tucker)
    - lit_entropic_penalty_cominetti_sanmartin_1994 — entropic-penalty -convergence (T0)
    - lit_gale_shapley_1962 — deferred acceptance produces a stable matching
    - lit_gale_shapley_proposer_optimal_1962 — proposer-optimality of deferred acceptance

    Stochastic calculus / PDE
    - lit_ito_1944 — Itô's lemma
    - lit_black_scholes_pde_solution_1973 — closed-form solution of the Black–Scholes PDE
    - lit_liouville_dirichlet — Liouville/Dirichlet (harmonic-function / PDE result)

    Dynamical systems
    - lit_saddlenode_passage_time — saddle-node "bottleneck" passage time / (Strogatz/Fenichel)

Don't ban from arvix for not proofreading something like citations on a hot result. Where do you draw the line? Drawing the line is hard to do so just push them into the secondary queue. Then they have to redeem themselves to get back into the main one.

This is so obvious, if they start banning people a second competing service will emerge.

A better idea is to simply have two submission queues. If you get banned from the high quality one you have to use the low quality queue. If they don't do that someone else is going to do it. It is a fact of life that papers are going to have AI generated writing in them going forward. Just accept this fact and make two queues.

How is this profitable to them? They sell the same TV all over the world. AFAIK the US is the only company with a monopolist controlling TV listing guides. This monopolist changes them a monthly fee for every active TV, Sony getting no further revenues. I haven't looked in a while but you as a consumer, can also subscribe to this service. I think they want $3/mth from retail customers. I haven't dealt with this in over ten years so I am forgetting the details.

It is because the directory info comes from a US company with a monopoly and they charge exorbitant amounts of money for it and then use copyright to shutout competitors. Tribune Corp used to own it (Chicago Tribune the Wriggly family, those monopolists).

Pretty sure that all taxes are paid by humans in one form or another, I don't think robots are paying taxes yet. Sure you can tax a corporation, but you are just indirectly taxing the shareholders, and they are people too.

You vastly overestimate the taxation capacity of the economy. Go read the fed's papers on optimal taxation and you will see that the US is close to the optimal rate right now. The theory of optimal taxation is to maximize tax collections over time. Sure you can raise taxes a lot right now, but if you do that people stop investing and you lose future taxes. Optimal taxation theory shows how to maximize the combination of current and future taxes.

Since we are sitting pretty close to the optimal number currently, we need to reprioritize spending, not just spend more. Spending more simply increases the debt. Increasing the debt increases interest expense. And the more the government spends on interest, the less it can spend on other priorities.

You need to consider the currencies. The Romans had metals based hard currencies. Try doing this with a fiat currency like the USD and you'll be toting wheelbarrow loads of cash to the grocery store. Look at what is happening in Iran right now with loss of confidence in the currency, 56% inflation in a month.

Total government spending federal, state, local in total is around $10T with $7T of that being federal spending. There are 133 million households in the USA. Now divide. That works out to $75,000 per household per year in government spending. The US median household income is $83,000.

Does anyone really think we need more government spending? Maybe the problem is that the existing spending is not being effectively utilized.

What happens when interest expense exceeds 50% of tax collections? The U.S. government's interest expense consumes a significant chunk of its revenue, recently reaching around 19% of all federal revenue in FY 2025, with projections suggesting it could rise to 27% by 2035. Maybe we can support all of that debt, but we won't have any more government spending programs.

Read the click-bait article carefully. They are not saying real ID is unreliable, they are saying a real ID card can be forged. When there is suspicion of a forged document it can take longer to verify its authenticity.