tsets are described in more detail here: https://buck2.build/docs/rule_authors/transitive_sets/. Bazel's depsets were one of the influences on their design. To users, they will seem fairly similar and would be used for solving similar problems, there's some differences in the details of the APIs.
I'm not well-versed on the internal implementation details of bazel's depsets, but one interesting thing about tsets that may further differentiate them is how they are integrated into the core, specifically that we try hard to never flatten them there. The main two places this comes up are: (1) when an action consumes a tset projection, the edges on the DICE graph (our incremental computation edges) only represent the direct tset roots that the action depends on, not the flattened full list of artifacts it represents and (2) when we compute the input digest merkle tree for uploading an actions inputs to RE, that generally doesn't require flattening the tset as we cache the merkle trees for each tset projection node and can efficiently merge them.
But you don't understand. Car analysts since the early 90s have been saying that electric cars would become profitable and become a serious chunk of the market by 2020 because they all knew that musk was going to come along and show us how great electric cars are and save the world.
That's nonsense. Take a moment to breathe. Think about just typical https, when you request some nytimes.com content that is actually being delivered to you througha bunch of intermediaries. So, by your logic any of those could just change that content. But it turns out, this was a problem solved like 50 years ago. With https, the content is signed by some key that you trust to represent nytimes.com.
Sidenote (this is really not targeted at the specific comment): Is there some site where people actually have a clue? On any topic that I have any passing knowledge about, HN is just completely spewing nonsense and the worst part about it is that they think they have a clue. I have to assume that HN's comments are just as useless on things that I don't know about.
This is actually pretty much the dumbest argument if you take just a moment to think about it.
Imagine some drug that costs $100 million to develop and that the most it can earn in the market is $70 million at a price of $1000. Okay, but then the government thinks it's really important to provide, so it's going to subsidize the development to the tune of $40 million. The drug company could then spend $60 million for development and earn $70 million. Over the typical drug development and sales lifetime, that's actually a quite mediocre return, but let's not worry about that.
But wait, the government paid 40% of the development cost. Why the fuck should that company be able to charge $1000, they should only be able to charge 60% of that, they should charge $600. Well, then they are fucked because they are right back at the same equation for whether it is profitable to work on that drug.
In general, with a moment's thought, any intelligent person will realize that expecting discounts on products based on how much the development was subsidized by the government completely obviates the entire point of subsidies.
That's bullshit. Companies have "whole letter"s for pretty much everything they do. Gilead had a "whole letter" when they said they'd provide 1.5 million doses for free. Was that also loaded with crap and the "$0" price excessive?
Haven't looked at their build, but it's almost certainly the case that they are generating some sort of rust binding from the typescript code and the compilation times are actually from needing to recompile rust.
Sure and an argument could be made that the specific things happening in this situation are wrong. That's not the argument that I replied to. I replied to a flippant argument that it is always wrong to charge different people different amounts.
I just went to what I considered a clear counterexample, though it does have problems that it is rare (though local coffee shops and bakeries, for example, fairly regularly will provide things for free to those in need within the store exactly like they would for regular customers).
Is it wrong for universities to provide financial aid to those in need?
Would it be wrong for a city to provide subsidized public transportation? subsidized school lunches?
You can come up with other harder to decide examples. A really good one would be whether it's wrong to charge tourists and locals different amounts. On the one side, this feels unfair to the tourists, why should they have to pay more? Generally, I lean to unfair things being wrong. But, the consequences of charging them the same are, to me, much worse, especially as the means of the two become further and further apart.
From the article: Steven Brust: "Ask what's more important, human rights, or property rights. If they say property rights are a human right, they're on the right."
That's only the case as long as complying with California's emissions laws is not too much harder than what they would otherwise have to comply with. That's a complicated analysis, it's kinda got a bit of prisoner's dilemma to it. As long as companies are following CA's restrictive emissions laws, there's little incentive for other areas to make their own laws more restrictive. If companies started to ignore CA's laws (and not sold there or sold different versions) you should expect other states to start to pass more and more restrictive laws.