I get your point but I'm not sure it matters all that much.
Did harbor / tb2.1 cap the swap available to docker runs?
There used to be a bug that would allow dockerized instance runs to use more memory than the specs allowed. Some of the original tasks weren't really possible to complete without exploiting swap. Even the oracle solutions didn't pass if you stopped docker from having access to swap.
I think crack-7z-hash and filter-js-from-html had that problem off the top of my head, but i haven't looked at this in months, so i'm not sure
Thank you. This is very interesting. I'm excited to see an open source project adopt a zanzibar-oriented approach to resource management. This is exciting!
Just before I hurl myself at this for several days/weeks -- where were the pain points? I'm usually wary of new projects in this space but OpenFGA looks pretty mature already
Claiming the mantle of "small government" was simply an exercise in marketing to relax regulation meant to prevent bribery and corruption. In practice, the current slate of government officials believes in absolute control of whatever they want whenever they want.
It's a mirror case of the supposed "free speech absolutists" who immediately turned around and silenced, sued, fired or jailed once granted the power to do so.
This argument strikes me as massively disingenuous. The central problem of the US tax system is caused by a combination of:
- high net wealth individuals essentially being indifferent to income tax.
- income tax and short term capital gains are taxed at much higher rates to long term capital gains.
- lower net wealth folks (ie. the general public) receiving most of their income as income.
- high and ultra high net wealth individuals now making most of their money through dynastic trusts and inheritance.
This combination ends up making it so that, as Warren Buffet would put it, he ends up paying a lower effective tax rate than his secretary.
I effectively don't really care if it's a wealth tax or some other more targeted technical fix, but it's not sustainable to have the very wealthiest individuals taxed at a lower effective tax rate than everyone else and also able to pass on their wealth directly to heirs without significant estate taxes.
I will never buy another google hardware product again after my most recent pixel experience. I was sent a phone with a defective modem that they refused to replace. This is despite having bought 5 other pixels and also using google fi and a bunch of other google products.
I will never trust them with a hardware purchase ever again.
It's already the case that you get much better results out of LLMs by forcing agents using them to go through additional layers of planning, design & review.
The future is going to dynamically budget and route different parts of the SLDC through different models and subagents running on the cloud. Over time, more and more of that process will be owned by robots and a level of economic thinking will be incorporated into what is thought of today as "software engineering." At some point vibe coding _is_ coding and we're maybe closer to that point than popularly believed.
I work at runloop and I've spent a considerable amount of time getting various benchmarks to run with very high concurrency (thousands at once). My experience is similar to your own: it takes a ton of time and effort setting up benchmarks to run at scale with protection against reward hacks.
Keeping a benchmark test harness secure and fast is non-trivial. You need to keep the grading script and the solution off the box, use network controls, deal with external resource usage, etc. It's a lot of work. I don't think it's realistic to expect benchmark authors to bullet proof their benchmark runners. Most benchmarks are written to be run conveniently on a single machine (ie. in docker), not to run in parallel across tends of thousands of secure, isolated machines.
I realize there's only so much you can fit into an article, but this article glosses over a monumental shift in welfare spending in the US: the transition from defined benefit retirement plans to defined contribution. It's not so simple as a split between private and public directed asset allocation: it affects the growth of companies that offer these plans and the wealth of participants in the plans.
The US has pushed the burden of retirement onto individuals, hoping that the private sector will offer incentives like 401k matching and generous health care plan subsidies, but this is a fundamental difference in who qualifies, what they receive, and how it's funded. These effects compound wealth and income inequality. If, for whatever reason, you're locked out from a job that would help pay for these programs, there's no coming back. You are dependent on the government at the same time as the government is underfunding the program you rely on. It's not a great situation.
Did harbor / tb2.1 cap the swap available to docker runs?
There used to be a bug that would allow dockerized instance runs to use more memory than the specs allowed. Some of the original tasks weren't really possible to complete without exploiting swap. Even the oracle solutions didn't pass if you stopped docker from having access to swap.
I think crack-7z-hash and filter-js-from-html had that problem off the top of my head, but i haven't looked at this in months, so i'm not sure