cpleppert

A bunch of reasons.

1. Napoleon's goal was to pursue and defeat the Russian army in the field, not necessarily capture cities. Going north would have meant releasing the pressure on the main russian field armies and let them engage his main force at their discretion while exposing his flank.

2. St. Petersburg remained the political and economic capital of Russia; St. Petersburg never displaced Moscow in real world importance.

3. St. Petersburg was shielded from land and sea with prepared fortifications on both and Napoleon lacked a fleet to effectively blockade it.

4. Its in the middle of a very dense forest and swamp, not the the best logistics and ability to maintain a siege.

The Nazis made the same choice in WWII and even though they were able to control the Baltics and had Finland as an ally never seriously threatened to take the city.

To be clear on https://www.mozilla.org/en-US/privacy/faq/ they changed this >>It seems like every company on the web is buying and selling my data. You’re probably no different. Mozilla doesn’t sell data about you, and we don’t buy data about you.

to this: >>It seems like every company on the web is buying and selling my data. You’re probably no different. Mozilla doesn’t sell data about you (in the way that most people think about “selling data“), and we don’t buy data about you. Since we strive for transparency, and the LEGAL definition of “sale of data“ is extremely broad in some places, we’ve had to step back from making the definitive statements you know and love. We still put a lot of work into making sure that the data that we share with our partners (which we need to do to make Firefox commercially viable) is stripped of any identifying information, or shared only in the aggregate, or is put through our privacy preserving technologies (like OHTTP).

Of course, saying "(in the way that most people think about “selling data“)" makes the guarantee completely meaningless. The rest of the paragraph is just marketing puffery. Its meaningless bromides about how much they value your privacy. Notice they only say they put "lots of work" into stripping identifying information provided to commercial partners (which is just another way of saying selling). Again, this is meaningless. They went from a very strong guarantee to no guarantee at all. Any company that sells your data that makes any effort at all to strip identifying information can make this claim regardless of whether personally identifying information can be recovered with a modicum of effort.

Short answer: yes. The AGPL should be avoided at all costs because it has never been robustly tested in court and its unclear what the licensing implications actually are.

There are various explanations in plain english sometimes offered about how the AGPL will apply. None of these are true.

Companies that have a made a business decision to provide AGPL licensed code do so with the understanding that no serious business will ever consider using such a product in their software stack. If you choose an AGPL licensed product it will (rightly) become a gigantic headache at some point. It will certainly become a problem if anyone does due diligence.

There isn't any evidence that models are doing any kind of "system 2 thinking" here. The model's response is guided by both the prompt and its current output so when you tell it to reason step by step the final answer is guided by its current output text. The second best answer is just something it came up with because you asked, the model has no second best answer to give. The second best answers always seem strange because the model doesn't know what it means to come up with a second best answer; it 'believes' the output it gave is the correct answer and helpfully tries to fulfill your request. Sometimes the second best answer is right but most of the time its completely nonsensical and there is no way to distinguish between the two. If you ask to choose it will be strongly influenced by the framing of its prior response and won't be able to spot logical errors.

Asking it to do lateral thinking and provide examples isn't really helpful because its final output is mostly driven by the step by step reasoning text, not by examples it has generated. At best, the examples are all wrong but it ignores that and spits out the right answer. At worst, it can become confused and give the wrong answer.

I've seen gpt-4 make all kinds of errors with prompts like this. Sometimes, all the reasoning is wrong but the answer is right and vice versa.

This isn't a simple composition; entire details were ai created. You can see sections around her hair that look like stable diffusion inpainting. The texture will immediately be cut off and a very softer blurred ai generated part will continue. There are details all around the image that look like that. The jacket and both sweaters aren't original for instance. Large sections of the background on the left were clearly added in as well.

Virtually the only original parts of the image that are unmodified are the faces and hair from the neck up and fingers on the two older children. The person who made this clearly had nothing to go on.

Fantastically chronicled in Atoms and Ashes. [^1]

The explosion was so powerful that the personnel in the monitoring station who were not expected to receive any radiation at all had to hide in a back room of their concrete bunker and hope an evacuation was possible. The radiation & fallout was so intense that all the monitoring ships had to leave their assigned stations. The bomb was a full 2.5 times what the designers expected: 15 megatons as opposed to 6. Just a complete unnecessary tragedy for the islanders.

[^1]: https://www.amazon.com/dp/B09VCPQZWP

The Mayan and Aztec conquests were completely different. The Mayan polities were situated in terrain that was much harsher for an invader, military capability doesn't mean much when a military force can't be effectively maintained in the field against. There also wasn't much incentive for the spanish to extend their domains into the mayan polities at that time. The spanish were far more focused on populous areas where they could extract wealth which was the only way for spanish conquistadors to gather volunteers. Even under these conditions the actual military engagements were very one-sided with most of the Yucatan conquered within thirty years of the Aztec conquest.

>but it might be the second foremost reason why Cortez was able to conquer them - the first being Montezuma’s reticence to slaughter them in the jungle when he first heard of them and his subsequent religious obsession with the white conquistadors, until it was too late and they were well established.

Montezuma was never in a position to slaughter the conquistadors in the jungle. The Tlaxcalans (a rival state to the mexica) engaged the conquistadors in a series of brutal skirmishes (most with the advantage of surprise) on their home turf but failed to make any real progress. Its very clear that the attacking Tlaxcalan warriors suffered horrendous casualties when they engaged the castillans in close quarters combat due to the superior steel weapons(mostly swords) of the latter. Since the Tlaxcalans vigorously defended their independence its hard to argue the Mexica could have done any better fighting at a distance that would have stretched their supply lines.

As for his supposed religious obsession, its unclear and isn't necessary to explain his actions. There is certainly no evidence that Montezuma was confused or acting irrationally; he repeatedly tried to deter the conquistadors from coming and made attempts to play them off against the Tlaxcalans.

Their most recent scientific paper they published 3 months ago only says that scientific breakeven (Qsci > 1) is possible. Which has been claimed many times before pesky engineering challenges and unmodeled behavior starts to interfere. The paper actually mostly just rehashes their marketing so its impossible to judge how realistic their claims are. Regardless, they don't seem like they are close to any kind of actual prototype let alone one that can demonstrate scientific breakeven or the order of magnitude harder engineering breakeven.

Their claims seem very hard to believe to put it mildy.

Helion has never demonstrated that their technology actually works on any scale. Even if they magically made breakthroughs and were able to hypothetically build a prototype plant it would be almost impossible to solve the operational challenges to build a commercial scale plant in seven years. It would be far more challenging than building a nuclear plant for example. That ignores all the complexity of their D-He3 aneutronic fusion cycle; they probably cannot operate a commercial scale power plant (that would presumably use D-He3) without having first constructed D-D reactors.

When economists talk about production in the context of an economy they always mean the value of that production. Economic growth never implies the raw quantity of economic production increased regardless of its value. That's a contradiction in terms. If an economy increases production of some finished good by 50% yet the value of that finished good declined that was not economic growth it was a contraction.

Economic growth literally means the value of goods and services produced by the economy rose. It does not mean "monetary value" in one place or "number of widgets" in another place. Robin Hanson is not confused here by what he means by economic growth and his argument does not rely on some ambiguity in terms.

>You can produce the exact same amount of potatoes, if the value of a kg of potatoes increases, the GDP rises.

Yes, of course.

>If you look at the value of oil to gauge its production, there's one day in 2020 when we sent oil back to earth[1].

I'm not sure what this is supposed to illustrate. If a finished good or service does not have value then it cannot contribute to economic growth. The entire point of an economy is to produce goods that have value.

> Also, as usual, what's the definition of economy being used there? Money? Production?

Production. The first result on google for "Economic Growth" defines it.

Its 'dropping' fertility and population into the mix because his thesis is that declining population will lead to less innovation thus retarding economic growth. Innovation happening in a recession is not in any way a counterexample whatsoever.

GDP is literally a measure of aggregate production. The entire point of GDP is to measure production in an economy. GDP does not measure money any more than mass measures kilograms. It measures the value of all goods and services produced in the economy.

> It is an alternative to the built in macOS automatic mode which only switches when the user steps away from the computer.

Huh? Setting a schedule/location for nightshift and setting the dark mode setting to auto will always change instantly. If you use a launcher or spotlight then a simple one line applescript can change the setting as well. (tell application "System Events" to tell appearance preferences to set dark mode to not dark mode).

I looked at the highest quality and the artifacting is just crazy here. There are no artifacts around the instrument at all, just on the supposedly levitating rock. The quality on this and the other video from the user looks like its very good so I don't really buy this.

The context here is rather strange as well, the same user uploaded another video that no one would believe actually demonstrated the meissner effect. Its a very small magnet on just a piece of paper. Did the user try again with this video?

They aren't reacting because this isn't really proof of anything. The test apparatus they used can't actually measure 0 resistance. Superconductivity is implied (thats the most straightforward explanation) but not actually demonstrated here. Its very hard to interpret from the limited data presented. There could be a whole host of contaminants, issue with the experimental setup etc etc. Its hard to go from look at the interesting material properties in this graph to room temperature superconductor. This is also the only confirmatory information we have from the team; their sample didn't show diamagnetism.

The reasons to be skeptical are: 1) lot of videos floating around that are just at the threshold of convincing 2) poor results from other teams 3) there is a history of superconductivity discoveries like this that never pan out 4) skepticism of the original paper

If this pans out fusion energy would be completely viable with enough investment. Right now tokamaks don't really have a pathway to being commercially viable and are basically 40 years away. For starters, If you can cut the cost by 2-4x you are right in the ballpark of what you would need to build a working tokamak fusion power plant. You would need to do better than that for fusion to be a viable power source but tokamak fusion was always only a magnitude away unlike other hypothetical fusion energy alternatives.

There are hundreds of nuclear power plants that have been in operation for decades. Most operating nuclear power plants run with dramatically higher uptimes than what they were originally conservatively designed to meet. Nuclear power plants are operating in an active war zone right now! Six reactors were the site of a military battle and ended up sustaining damage beyond anything foreseeable by the designers. Every one of them shut down successfully and at no point was there any evidence that there was ever a safety risk to the public unless a malevolent actor wanted to cause such an accident.

This body of evidence strongly supports the view that nuclear is not only safe but incredibly safe. We have incredible knowledge about how to run nuclear plants safely.

Lazard's LCOE metric is absolute nonsense. Besides ignoring intermittency and the variability of renewable generation it uses an incredibly unrealistic cost of capital comparison that favors capital intensive renewables. Rooftop solar would look better if the cost of building a massive utility scale solar farm had to funded at realistic corporate discount rates.

The significant development after the end of apartheid was more due to a very favorable international economic environment and the elimination of sanctions that restricted south africa's ability to export resources. Additionally, the end of apartheid was in itself a massive economic boon as it created new sources of labor.

The performance of the ANC has always been poor in South Africa. The party engaged in widespread AIDS denialism and dramatically cut investment as it attempted to help poor black africans through a variety of mostly poorly run schemes. Corruption was always present and connections with the ANC became essential in south africa. Today's troubles are the result of decisions made years ago.

You can also use a duckduckgo query like ! site:site.com query. That is more reliable when using something like wikipedia when you aren't sure of the exact title and don't have to type it all out.

This bookmarklet will work:

  (function () { let sel = window.getSelection(); let Qr; if (sel && sel.toString().length > 0) {Qr=sel.toString()} else {Qr=prompt('Search Site for','');} let hna=window.location.hostname; if(Qr) {  location.href='http://www.google.com/search?&q=site:'+encodeURIComponent(hna)+'+'+escape(Qr)  }})();

The mac doesn't use ctrl modifiers to use copy paste. If you remapped Alt/option and windows/command keys on a windows keyboard alt-c & alt-v (i.e. the keys to the right/left of the spacebar) would be recognized as cmd-c and cmd-v on the mac. PyCharm supports a windows and linux keybinding that lets you use ctrl-c and ctrl-v to copy paste but the mac generally uses the cmd key as opposed to the ctrl key. If you want to use ctrl-c and ctrl-v you can remap both of those pretty easily using karabiner elements to cmd-c and cmd-v.

Carbon fiber is a lot easier to work with than metal in this case because welding metal creates failure points. I don't think they could afford building a titanium pressure vessel. Weight is a factor too, carbon fiber is lighter so you can make the pressure vessel thicker and still have enough buoyancy to reach the surface(probably a good thing).

From their perspective it kinda made sense even though its not safe at all.

Immediate implosion if a far more likely explanation because give that we knew it imploded the most likely time would be immediately after the pressure vessel reached maximum or near maximum stress and it failed. The time actually is not the biggest factor; it is the number of cycles the pressure vessel has endured. An implosion after a relatively low number of cycles is consistent with past incidents with pressurization failures.

As was pointed out, the submersible is pretty large and its a essentially a large cylinder which induces stress in the middle. Usually you make deep sea submersibles out of a sphere or spheroid because you don't have this failure modes.

It isn't surprising it failed. They were strong claims that the company did do enough test, rejected safety advice and even had the CEO claim that at one point the titan wouldn't be able to reach the titanic because it (a model maybe?) had experience a pressurization failure. To say nothing of the materials they were using.

Just comically bizarre that anyone involved in the project thought it would somehow survive let alone be safe.

They don't advertise any of their "new submersibles" on the website and the specifications they claim for Titan don't match any of their new submersibles; Titan is rated to go to the Titanic and not any lower. Titan isn't a previous version either. Unless they shelved the first Titan they probably just added some strengthening and called it a day.

I'm blown away that for a sub of that size and shape they used carbon fiber. The sub is positively gigantic and a cylinder is very susceptible to failure after repeated cycles. I'm pretty sure the claims they made about working with NASA and having a system to monitor the structural integrity of the sub are just BS.

The Xbox controllers are used to control the periscope which is not a safety critical device. Regardless, the navy uses wired controllers and did extensive testing and verification. This outfit didn't do anything like that; in one video with a journalist the bluetooth controller was a 'feature' because they could pass it around the sub.

None of those sources has any relevance to the bottom line accounting cost of large scale utility storage for intermittent renewable energy. You presumably have to over provision both storage and baseline power generation to assure the grid of enough reliability. The assumptions you make in doing that drive the cost of the entire solution.

Lazard has mostly ignored those costs when it tries to come up with a LCOE for intermittent renewables. A real world grid operator can't ignore those costs. To take a real world example, Germany requires dramatic subsidies to build out its renewable energy portfolio. These subsidies occur at every level of electricity grid. There is no simple way to untangle the costs and come up with a single number that can be compared to non-intermittent sources of energy. The share of biomass in the German grid is strongly suggestive of the fact that renewable energy isn't cost competitive without significant subsidies.

There is no free lunch, decarbonizing energy generation requires more expensive sources of energy.

My understanding is that Apple can still contractually require developers to hand over the 30% even if developers use a third party service. The judge's ruling was pretty narrow; it essentially allows nothing more than a link to an alternative payment platform. I haven't seen any in-depth analysis of the case to suggest that this is any kind of 'loophole' that could be used to avoid paying apple.

> It looks like the stages failed to separate and it started to spin.

Most complicated part of building any rocket by far and Starship's is a lot more complex than any previous one by far. It was always the most failure prone part but everything else looks great.