That's quite the leap though, and is just confusing correlation and causation. Maybe the previous leadership was simply getting in the way of the engineers and managers that had the good ideas. And the new leadership was more hands-off, or focused in other areas like marketing. Or those cases are just flukes. For every case like the ones you cite, I could find two where the exact opposite happened.
If you're downing a shot of vodka every morning, and suddenly stop, then yeah, your health is going to improve.
In my opinion, many (if not most) of these CEOs are business-focused people with no technical (or even non-technical) knowledge of anything they purport to manage. And on the whole, they really don't affect the value of the company one way or the other.
I'm not saying any schmuck could have done that. I'm saying that the engineers and managers at Apple (to use your example) are just as (if not more) responsible for the success than Jobs. Those lower-level engineers and managers also explain the repeated successes. And that, I would say, is the case in most market successes. The CEO is not remotely deserving of all the credit, or even most of the credit, in most situations. They don't really deserve to be paid what they're paid.
There's this certain anti-historical proclivity to create heroes for worship. Because it's a simple story to tell and it gives you the opportunity to put yourself in the hero's shoes. But the simple story is almost always wrong.
> You don’t have to randomize the first part of your object keys to ensure they get spread around and avoid hotspots.
From my understanding, I don't think this is completely accurate. But, to be fair, AWS doesn't really document this very well.
From my (informal) conversations with AWS engineers a few months ago, it works approximately like this (modulo some details I'm sure the engineers didn't really want to share):
S3 requests scale based on something called a 'partition'. Partitions form automatically based on the smallest common prefixes among objects in your bucket, and how many requests objects with that prefix receive. And the bucket starts out with a single partition.
So as an example, if you have a bucket with objects "2025-08-20/foo.txt" and "2025-08-19/foo.txt", the smallest common prefix is "2" (or maybe it considers the root as the generator partition, I don't actually know). (As a reminder, a / in an object key has no special significance in S3 -- it's just another character. There are no "sub-directories"). Therefore a partition forms based on that prefix. You start with a single partition.
Now if the object "2025-08-20/foo.txt" suddenly receives a ton of requests, what you'll see happen is S3 throttle those requests for approximately 30-60 minutes. That's the amount of time it takes for a new partition to form. In this case, the smallest common prefix for "2025-08-20/foo.txt" is "2025-08-2". So a 2nd partition forms for that prefix. (Again, the details here may not be fully accurate, but this is the example conveyed to me). Once the partition forms, you're good to go.
But the key issue here with the above situation is you have to wait for that warm up time. So if you have some workload generating or reading a ton of small objects, that workload may get throttled for a non-trivial amount of time until partitions can form. If the workload is sensitive to multi-minute latency, then that's basically an outage condition.
The way around this is that you can submit an AWS support ticket and have them pre-generate partitions for you before your workload actually goes live. Or you could simulate load to generate the partitions. But obviously, neither of these is ideal. Ideally, you should just really not try and store billions of tiny objects and expect unlimited scalability and no latency. For example, you could use some kind of caching layer in front of S3.
Yep, this is still a thing. In the past year I’ve been throttled due to hot partitions. They’ve improved the partitioning so you hit it less, but if you scale too fast you will get limited.
Hit it when building an iceberg Lakehouse using pre existing data. Using object prefixes fixed the issue.
This is my understanding too, and this is particularly problematic for workloads that are read/write heavy on very recent data. When partitioning by a date or by an auto-incrementing id, you still run into the same issue.
Ex: your prefix is /id=12345. S3, under the hood, generates partitions named `/id=` and `/id=1`. Now, your id rolls over to `/id=20000`. All read/write activity on `/id=2xxxx` falls back to the original partition. Now, on rollover, you end up with read contention.
For any high-throughput workloads with unevenly distributed reads, you are best off using some element of randomness, or some evenly distributed partition key, at the root of your path.
Presumably the superior solution is the product that bears the same name as this blog post. Which I take it is in the process of being released since I can't find many technical details about it.
That builders can't construct housing in whatever form-factor they want and whatever part of a municipality they want might be a problem. But the analysis of why these kinds of restrictions exist, to my mind, is not correct.
Do niche interest groups have influence on housing policy? Sure. But these niche interest groups don't usually have a monetary interest in the outcomes they're promoting. But individual homeowners (who often band into interest groups of their own) and large real estate conglomerates do have such a monetary interest. They consider certain types of housing built in a certain way to have *more value* and be worth more. So they promote politicians who introduce zoning and other rules to protect that value.
For someone looking to make a profit off of housing (or even to invest in housing), what is more appealing? A traditional U.S. suburb? Or a Kowloon walled city? One is denser, cheaper per capita, and (if you're not careful) unappealing to look at. In other words, it's worth less. So there is a great monetary pressure from people who already own homes to prevent "mixing" this type of housing into existing planned communities. People who need homes, on the other hand, are a little bit less discerning (to say the least). They don't have a monetary interest necessarily. They're primarily looking for a permanent residence.
So I just don't buy this "abundance" stuff in general. If you remove all of these restrictions, will some companies start building housing? Some will. But my guess is most will say the juice is not worth the squeeze -- the profit margins and the long-term values of these properties will make it unappealing. Just like it's unappealing for grocery stores to set up in big urban areas. Or for hospital providers to set up in rural areas. Food deserts don't exist because of too much government intervention. A lack of rural hospitals is not a problem because of too much government intervention. It's because those things are not profitable.
So in my opinion, if you want to reform zoning rules or things of that nature, it's only really going to be effective if you *force* (or if you want to be politically correct, "incentivize") housing companies to build in these areas too.
I don't really consider this an "anti-trust" argument. It can be equally true if there's a lot of competition in the housing market and if there's next to no competition. It's more of an incentives argument. This is an argument that, like with medical care, we're treating something that is a fundamental need of every living human to have a stable and peaceful and fruitful life as if it were a standard market commodity. And when you do that, you get poor outcomes. We need to support the building of housing *even if* it's unaffordable or has low (or even non-existent) profit margins
I'll just note that this seems entirely predictable. So much so that I can't help but see it as purposeful. The federal court system itself only has about 25,000 employees. SCOTUS has 9 judges plus a couple dozen clerks and other assistants. Lower courts already do not have enough employees to contend with an executive branch made up of millions of individuals, especially when that executive is ordering its employees (seemingly) to just ignore or purposefully misinterpret laws, leading to an ever-increasing number of lawsuits. To further reduce the power of lower courts at this time (which this SCOTUS seems to do in almost every decision involving the executive) means even more cases for SCOTUS and even less time for arguments.
Conveniently, we have the shadow docket. A way to issue diktats without any arguments before the court and in many cases without any reasoning whatsoever.
And conveniently, lower courts can then interpret a lack of details from a shadow docket decision however they want. So that the executive can appeal yet again, to get another thumbs down from SCOTUS (without any explanation), and round and round we go. The executive gets to keep the plates spinning while it essentially does whatever it wants.
With this situation, why shouldn't we simply pack the courts? If SCOTUS is going to take more cases than it can handle and not provide any real guidance to lower courts, then clearly they need more employees.
The rise of the “shadow docket” is driven by changing behavior of litigants and the district courts. The main reason cases get on the shadow docket is litigants seeking, and district courts granting, TROs and preliminary injunctions against major executive or legislative actions without trials or often even full briefing. It can’t possibly be true that a district court can block a major action within days but the Supreme Court can’t correct it until years later through the regular appeal and certiorari process.
I'm not seeing how that applies. There's a clear asymmetry between lower courts issuing temporary restraining orders and preliminary injunctions on the basis of well-established precedent, vs. the Supreme Court overturning those with little or no explanation or justification.
When the executive or legislative action is "major", that would seem to make it more reasonable that the lower courts put the changes on hold pending a trial. Drastic changes should be implemented only with strong justification, and when a drastic change seems to be very clearly in violation of existing law, it is in dire need of checks and balances with teeth.
It certainly isn't the Supreme Court's job to help the executive pull off major changes more quickly.
The complaint isn’t about the “asymmetry,” it’s about the symmetry. A hastily issued district court order should be (and is) subject to being overturned by the higher courts just as quickly as it was issued.
The precedent may be binding, but injunctions—unlike final judgments—are discretionary relief. The district court doesn’t have to grant discretionary injunctive relief based on precedents that are in doubt, and if they do, it’s perfectly fine for the Supreme Court to overrule that.
> Drastic changes should be implemented only with strong justification
That’s exactly backwards. The elected branches don’t have a “mother may I” relationship with the courts. That’s not the design, and that wasn’t the practice for most of the history of the republic. Injunctions are by definition “extraordinary relief” reserved for the most unusual cases.
Go read Marbury v. Madison again. The Supreme Court bent over backwards to avoid having to enjoin the Secretary of State to perform what it concluded was a “ministerial,” non-discretionary action (delivering a letter that had already been signed by the outgoing President). Most injunctions against executive action in the recent decades would have been considered unfathomable overreach by the founding generation.
> The elected branches don’t have a “mother may I” relationship with the courts.
That "mother may I" phrase implies asking for pre-approval for routine stuff, which obviously isn't what's at issue here.
> Injunctions are by definition “extraordinary relief” reserved for the most unusual cases.
The unusual and extraordinary may have become quite common of late, but that doesn't mean it starts being wrong for a lower court to intervene when the president tries to fire an FTC Commissioner in clear violation of precedent about firing FTC Commissioners. A president ignoring the law is definitely sufficiently "extraordinary" to justify temporary injunctions, and the fact that he's doing it a lot only reinforces the need for the checks and balances.
> A hastily issued district court order should be (and is) subject to being overturned by the higher courts just as quickly as it was issued.
Don't be ridiculous. The entire concept of temporary injunctions is built on the fact that many situations need to be slowed down and the potential harm prevented until serious legal questions can be properly considered. The lower courts aren't being "hasty" in the sense that they're issuing these orders on a whim with too little justification, they're being hasty in the sense of acting with commensurate speed to the executive actions of questionable legality that the courts need to keep in check. But the Supreme Court shadow docket decisions do come across as insufficiently thought out.
> The district court doesn’t have to grant discretionary injunctive relief based on precedents that are in doubt,
A lot of these precedents really aren't in doubt, except to the extent that it's clear the current Supreme Court doesn't like them. But until such time as the Supreme Court furnishes the proper decision overturning existing case law (not just hinting at a willingness to do so in the future), it's reasonable for lower courts to continue applying that case law when it is extremely obviously applicable.
> That "mother may I" phrase implies asking for pre-approval for routine stuff, which obviously isn't what's at issue here.
There is no relevant “routine” versus “non-routine” distinction in either the constitution or the law. Obviously, the folks who had just overthrown their government and created a new one contemplated elected branches that could make dramatic changes!
> A president ignoring the law is definitely sufficiently "extraordinary" to justify temporary injunctions, and the fact that he's doing it a lot only reinforces the need for the checks and balances.
It’s not, actually. The primary check on the elected branches is elections, not lawsuits. The courts exist primarily to vindicate personal rights, not to manage national policy. The framers never even envisioned that courts could enjoin the President for discretionary acts. Even Marbury rejects that notion.
Here, the personal right—someone’s employment as an FTC Commissioner—is the tail that’s wagging the dog. It’s not “extraordinary” because losing your job is such a grievous injury. It’s just a proxy for the government policies the FTC Commissioner has the power to execute. So the courts are being invoked in a battle over policy, which is exactly where the power of the courts is the weakest (by design).
I’m personally a fan of choosing by lot, from the appellate bench, a random slate of justices for each case. (That court of rotating judges would be the one in which “the judicial Power of the United States, shall be vested” [1].) You could do this entirely through legislation—nothing in the Constitution requires lifetime appointments to a permanent bench.
The democrats are just not interested in being as opposed to this as they morally should because a large portion are bought biy the same capitalist forces that buy republicans.
Whats more likely are states to shift policies to ignoring the federal courts and localize.
The essential problem is the federal government is attacking the most fortified jurisdictions while choking their own support structures.
That doesnt mean theyll fail, the same way shitty policies dont prevent Taliban rule.
I've never understood this argument. When I drive, I'm relying on more than just vision. I'm relying on sound, tactile feedback, an understanding of traffic law, an understanding of human behavior on the road (some of which is specific to my locality, the time of day, etc.), an understanding of weather conditions and geography, past problems I've run into on the road and heuristics I've learned over time to improve safety, weighing the risks vs benefits of certain maneuvers (especially considering how specific cars around me are behaving), and so forth. Some of this a purely vision focused FSD could simulate. But much of it it cannot. That's why alternate sensing systems really are needed, whether it's LIDAR or something else.
Every thing is a sensor. It provides input. Tesla decided to remove more and more input while claiming it’s better. Even Tesla knows because the robo taxis need more sensors than they sell in stores. So latest HW4 you buy is already obsolete
In the sense that LIDAR can detect some things that visible light sensors cannot, it can make up (in some respects) for the lack of those other capabilities. It's not meant to replace them per se, but to provide another dimension of detection.
This is decidedly not equivalent to those things, even roughly. Interpreting a statute (which by the way this Supreme Court has expressly limited, which is ironic given your comment) is quite different from interpreting an amendment to the Constitution. What you're describing is like saying a president can re-interpret the meaning of 'militia' in the 2nd amendment and start confiscating everyone's firearms because they're not members of a militia. Even despite rulings like Heller. This is exactly the argument that the dissent in this case makes.
With all due respect, what about my comment makes that ironic? I’m generally quite careful not to post if the only thing I have to offer is an opinion, and try to keep opinion out of it entirely.
> What you're describing is like saying a president can re-interpret the meaning of 'militia' in the 2nd amendment and start confiscating everyone's firearms because they're not members of a militia. Even despite rulings like Heller.
That’s basically what was done prior to Heller. Not for “everyone”, but functionally for several minority groups. Since Heller those efforts have primarily been at the state level (which is somewhat different, even with incorporation).
> This is exactly the argument that the dissent in this case makes.
Prior to today, I wasn’t super familiar with the case law here. It does look like SCOTUS has ruled in the past: US v. Wong Kim Ark (1898).
It’s after 4am here, and Wong Kim Ark is a long ruling. I simply don’t have the attention remaining to read the whole thing right now, but I skimmed it and ran it through my a couple of LLMs. In skimming it, it appears to be applicable here both broadly and specifically. All the LLMs I’ve tried agree with that statement, including when phrased in the negative.
In short: I was wrong, I believe you’re correct, and this EO should be overturned citing this precedent.
——
As we’ve spoken of opinion, I’ll break with my normal course and state plainly: I believe birthright citizenship should be ended. However, given what I know now, the only acceptable way to do so would be a Constitutional amendment.
I’m opposed to anything that oversteps governmental authority, regardless of whether or not it happens to serve my own political interests at the time.
> This is roughly equivalent to the EPA or BATFE changing the way it interprets an longstanding statute in a new way — which, by the way, they do very frequently.
The recent Supreme Court ruling that overturned Chevron deference limited the ability of such organizations to interpret the law. If the law does not make it clear then it (now) must be interpreted by a judge. I assume the irony they are referring to is in thinking this Supreme Court would have ruled in a seeming contrary way in this case. (It is a bit ironic but is also beside the point, which I guess explains why they put it in a parenthetical.)
----
> I believe birthright citizenship should be ended.
I am curious to understand why you take this position. Would you mind explaining your perspective?
What you say may be true with respect to breaking laws. But illegal immigration is one of those relatively small infractions, and only now is there some sudden insistence to prosecute all of them and deport them. So this is a self-made problem.
All of the evidence available to us shows us that migrants, on average, commit less crime than U.S. citizens. The evidence shows us that they pay into social programs without reaping almost any benefit. The evidence shows us that they take jobs that the average American isn't interested in. An evidence-based political program would not target migrants as a first priority, except to provide some more straightforward way to become documented and legal.
The other issue is -- the U.S. has 300 million+ citizens. This argument that migrants will "clog up" the courts seems ridiculous if you also believe U.S. citizens deserve due process. If your court system can't handle a relatively small percentage of your residents committing the crimes you have on the books, then maybe those crimes aren't really serious crimes are they? Or else not funding the courts appropriately to satisfy the political program is purposeful. The goal is to avoid due process and accountability, for citizens and non-citizens alike.
They are bothered by it. But the anti-liberal, extra-judicial, law-ignoring method this administration is levying against Harvard is also being levied against many, many other progressive priorities and interests worth even more than taxes on a $52B trust fund.
That's fine, but allegations in a law suit aren't prima facie evidence of anything. Especially when the text of that law suit is filled with political invective (calling protesting students and faculty "uncontrolled antisemitic mobs" and so forth).
There's a very easy determination to be made here about which students are or are not being victimized. If my knowledge of current events is still accurate, not a single pro-Israel student has been extra-judicially kidnapped and imprisoned. Pro-Israel Jewish students very well may feel victimized or scared. But put into perspective, I can imagine that pro-Palestinian students feel it much more so.
That's quite the leap though, and is just confusing correlation and causation. Maybe the previous leadership was simply getting in the way of the engineers and managers that had the good ideas. And the new leadership was more hands-off, or focused in other areas like marketing. Or those cases are just flukes. For every case like the ones you cite, I could find two where the exact opposite happened.
If you're downing a shot of vodka every morning, and suddenly stop, then yeah, your health is going to improve.
In my opinion, many (if not most) of these CEOs are business-focused people with no technical (or even non-technical) knowledge of anything they purport to manage. And on the whole, they really don't affect the value of the company one way or the other.
reply