HNHacker News
TopNewBestAskShowJobs

e63f67dd-065b

1,891 karma · joined September 14, 2019

submissionscomments
e63f67dd-065b··on Costco gold bars are selling out within hours
Ah yes, my gun and my gold will protect me against the organised army out to raid for food.

That's what I meant -- people get organised very quickly. In Rwanda, people quickly formed armies and militias, the same in Ethiopia, Columbia, and literally every other time this has happened before. One guy is not gonna accomplish much.

Civil war is also, by most measures, the least likely outcome. Usually civil war causes the breakdown, not the other way around. Usually it's really a deterioration of state capacity that leads to a powerless central government (see Venezuela, Yemen, Syria), or repeated coups that lead to unstable governance (all of west africa basically). Things are mostly stable, by and large.

e63f67dd-065b··on Costco gold bars are selling out within hours
I don't understand why people keep thinking that a financial meltdown means societal meltdown. There's plenty of places where the economy is or has already melted down even right now -- Sri Lanka, Argentina, Venezuela, etc. We can observe what happens when political institutions completely and utterly collapse, because it's happened multiple times already. What happens is not immediate apocalypse where people riot and society breaks down; it's mass protests, shows of solidarity, people help each other out, etc. The prepper scenario of "I've gotta defend myself against mobs out to raid for food" really doesn't happen.

If we stretch it back a few decades we see things like the Wiemar republic, complete anarchy during the Russian civil wars when blacks, whites, reds, etc all marched across the countryside, Chinese civil war, or even just recently we have Yemen, ISIS, Ethiopia, and in none of these cases have a zombie apocalypse-style "have to defend myself with my gun and ammo" societal breakdown occured and people bartered with gold. It turns out that people organise very quickly, and a every-man-for-himself-style situation like in stories just doesn't happen.

Maybe a civil war breaks out, maybe there are food shortages, but for the most part if violence breaks out, it's organised, mainly in the form of civil wars or local power struggles. One guy with a gun is not gonna help. Food protests are common, but rarely end violently. People keep thinking that "this time is different", and they're wrong every time.

e63f67dd-065b··on Costco gold bars are selling out within hours
You can see it happening right now in Argentina: the local currency is worthless, and people are flocking to ... US Dollars. Not gold, not bitcoins, but good old benjamins.

So gold isn't even a good inflation hedge: if you walked the streets of a failing economy (pick one: Argentina, Venezuela, Sri Lanka, there's plenty more) and tried to exchange gold for goods, you're not gonna get far (and will probably get mugged). You have to figure out how to convert that gold into something somebody will actually take, and that seems like the hard part.

e63f67dd-065b··on The Internals of Deno
You're onto something here, I feel; it's not that it seems LLM-ish, as LLMs are certainly capable of generating very high quality writing with good prompting. It's more a mixture of low-quality/effort writing and a lack of style that makes it seem like a writing factory churned it out.

Clicking on a few random paragraphs:

> A significant portion of Deno's codebase is crafted using Rust, a highly popular and secure programming language. Rust's robustness adds an extra layer of reliability to Deno's architecture. Various essential elements of Deno, such as the CLI (Command Line Interface), module graph management, runtime execution, operational functionalities, and core mechanisms, heavily utilize Rust.

> The subsequent phase in executing code involves establishing permissions. Permissions stand out as a distinctive feature within the Deno runtime system. Deno provides an exceptional and safeguarded environment for running programs. This exclusive safeguarding is achieved through the mechanism of permissions

> We have repeatedly traversed these steps multiple times within the recent preceding sections. We begin by considering the root module and initiating its loading. Following this, we load all associated dependencies. Afterward, both the root module and its dependencies are brought into existence. This phase involves the instantiation of the root module along with its dependent modules

It's -- lacking soul, is the best way I can put it. The writing feels very mechanical, with one sentence at a time, slowly trodding along, like the many high school English essays that I've read too many of. There's no variation in sentence length, no voice, no soul. It's the writing equivalent of a presentation that's just reading individual bullet points from Powerpoint.

e63f67dd-065b··on The Internals of Deno
I find the format of a book a bit puzzling for what seems to be a mixture of technical documentation, introductory material, and educational material, all in one place. What exactly is the audience of this book? Why is it not just part of Deno docs? Why does it seems torn between multiple objectives? Surely this will just become outdated as Deno changes over time, which it will as a fairly new project.
e63f67dd-065b··on The$109T global stock market in one chart
> By "inflation" here you mean something like "average price level", which of course is also skewed by changes in the money supply, according to standard microeconomic supply and demand.

That's the whole point -- CPI doesn't cares only about price levels over time, with as consistent of a basket of goods as we can get. It doesn't matter if price levels increase because of increased money supply, or supply chain shocks, or trade policy, or a whole host of other factors. Those are all incidental to the thing we're actually trying to measure.

To put it another way, money supply is one amongst many other factors that we effects price levels, which we must adjust for when trying to measure economic growth over time. The 70s oil shock, for example, dramatically decreased economic activity, with almost no effect on money supply at that time.

> So it does not strike me as a good way to measure "real economic growth"; the fact that it is the usual method in all developed countries is a bug, not a feature ... A properly functioning economy does not need a single number for "growth". The fact that government micromanagement requires such a number is, again, a bug, not a feature ... But central government micromanagement doesn't work. The Soviet Union spent most of the 20th century proving that.

There are many quibbles amongst economists with GDP as a measure of the economy, and this is not one of them. The problems are manyfold -- it doesn't capture informal transactions, which is a big chunk of many developing economies.

I think, however, you're labouring under a misconception here. There are very few areas of the economy in which the US government tries to micromanage, and even then, uses GDP as the key metric to do so. There are only a few places where the government does central planning, and in almost none of these cases are they even close to Soviet-style central production quotas. The few that come to mind are: military production (for obvious reasons), strategic stockpiles (eggs for vaccines, petroleum reserve, the infamous government cheese saga, etc).

GDP played no part in deciding on any of the above. I don't think you quite understand how Soviet central planning worked. There would literally be a central planner who decided that "we will produce x nails this year", and that was that. The factory got their quota, and would produce the nails, with no regard if anybody's gonna use the nails on the other side, or what kind of nails they would want, or of what quality. The only part of central planning that was well-run was military procurement, since the military would regularly reject goods for being of substandard quality.

The US got the memo that this was a terrible idea -- that was one of the objectives of the Cold War, in fact. I've heard many criticisms that the US is too capitalistic, but the "too much central planning" is certainly a new one.

> I understand that this is a common claim among mainstream economists, but first, that doesn't make it right (monetary expansion is a common feature of failed civilizations in human history, and to just blithely adopt it as policy is not wise), and second, "grow with economic activity" is not how the money supply in any developed country is actually determined.

Those failed civilisations in human history operated under completely different economies than modern ones, and comparisons are practically useless.

There is good reason that no mainstream economist looks to return to the gold standard. Economists like to debate about monetary policy, but there is still some measure of consensus on how to manage it. The decision process is not "oh look GDP went up, we need to expand supply"; the supply of money is a means, not an ends. The goal here is to maintain stable monetary policy -- to maintain low inflation, high employment, and sustainable economic growth -- and money supply is merely one of many tools in the toolbox to achieve that goal.

The easiest example to see is that money supply is currently contracting as the fed restricts credit conditions to fight inflation. Once again, money supply is a tool, one of many central bankers use to achieve objectives.

> (There is also the fact that monetary expansion is equivalent to a wealth transfer to those who get the newly printed money from those who don't, which generally means enriching the already rich at the expense of everyone else.)

That's ... not how this works? 2020 saw the largest expansion of money supply in recent history, and guess where all that money went? Right into the hands of ordinary americans. The fed doesn't print money and hand it out to rich people, you should read up more on how monetary policy functions.

Very broadly speaking, monetary policy acts on broad factors -- mainly the target fed funds rate. Old econ textbooks will have discussions about reserve requirements and what not that modern economists have found to be not the main factors anymore, as far as I can tell modern money supply is controlled by open market operations and the myriad of central bank facilities. They effect demand for cash, which then in turn feeds into cost of capital decisions, which then flows into the rest of the economy.

Wealth allocation mainly happens in the realm of fiscal policy, which is very much not controlled by central banks. That's the politicians' job, which is a whole different story.

I've rambled long enough now, but the main idea is that you need to read more about how monetary policy actually works. You keep talking about government micromanagement, when the Chips act is the first time in a very long time we've tried anything close to such closely targeted industrial policy.

e63f67dd-065b··on The$109T global stock market in one chart
Huh you're right, I read the numbers on my portfolio calculator wrong. Inflation-adjusted CAGR from 1993 to 2023 is 7.09% (as calculated using Vanguard's total stock market fund, which closely tracks a broad market index).
e63f67dd-065b··on Apple Should End Their Google Search Partnership
If I'm reading this right:

- Apple should light $15b/yr on fire, so that they can:

- Build their own search engine, which will somehow be better than Google, and

- Monetise it even better than Google did, somehow

Wow I haven't read such a bad take in quite some time.

e63f67dd-065b··on The$109T global stock market in one chart
This is mostly correct, but in an extremely unhelpful way.

Inflation-adjusted CAGR over the past 30 years is ~~10~~7%. During that time, the average P/E ratio of the S&P 500 (that's what I could easily find) has hovered pretty consistently around 15-20. So the rise in equity prices is, as far as I can tell, directly attributable to increases in earnings (as it should be; after all, theoretically, stock prices are discounted future cash flows).

Directly adjust for money supply is not a particularly good way of measuring, well, anything. There's a reason talk about M1/M2/etc have mostly fallen out of favour: we realised that they don't really mean anything, and the numbers are mostly useless for crafting policy or analysing things we care about.

Adjusting for inflation is the standard metric for good reason. That's a much better measurement of real economic growth (real in this context usually means adjusting for inflation, instead of money supply. Almost nobody does the latter). Money supply should grow with economic activity; it would be strange for a much larger economy to be measured relative to a much smaller relative money supply.

e63f67dd-065b··on The$109T global stock market in one chart
> He reads headlines of UAW workers going on strike to ask a company making $10b/yr in profit to give them a bit more and he sides on the side of the company, you know what I mean?

He makes a good point though. Ford's current return on equity is a bit north of 9%, there's really not much more squeeze to give. People like to think of companies as giant monoliths of money that can be freely diverted, but that's really not the case. The other big automakers are roughly around the same place I believe, so it's really not that far off. 10% is roughly the RoE you're looking at for the auto sector, going higher means they're somehow getting capital for cheap, too low and they're not efficiently deploying capital and should just return it to shareholders for better use elsewhere.

The current two-tier system doesn't make much sense and is mostly a relic of the 08 era, but the current union demands are just ludicrous. The union protected old members at the expense of new ones, and a re-negotiation of that makes sense, but most of the other demands are nonsensical.

> He thinks that what we classify as a $20/hr job in America is $2/hr elsewhere.

Seeing as low-skill service workers are making $20/hr in major cities, and $2/hr is literally what the exact same job gets you in developing countries, he's not wrong? The guy manning the Walmart checkout counter probably makes close to 10x what the guy manning the same counter in, say, Bangalore.

> If that's the case culturally and the Chinese value working hard (where as in America if you scroll on Instagram you'll see 5-10 memes recommended to you about 'you are not your job, fck your job, fck your meeting'), why is our economy more successful than theirs?

There are other factors of play, of course. Labour productivity is only one part of the economy, arguably not even the most important. If "working hard" is all that's required for a strong economy, then yes the Chinese with their 996 schedules would have the strongest in the world. So would the rest of the developing world.

The rest of the story is the boring economic stuff that nobody outside of finance thinks about: highly functional capital markets, low costs of capital, strong investor protections, strong disclosure regimes, existing structures around accountability, a managerial class that's somewhat aligned with investor objectives, etc. Things that many countries are trying to develop, albeit without much success.

The US model is not the only path towards success -- there are many others, and many are hard to replicate. Singapore was invited to try duplicating their success in China and it was mostly a failure. Export-led growth has served many other Asian countries well, but many get stuck in the middle income trap. Truly climbing into a developed economy is a hard task, one that not many succeed in.

e63f67dd-065b··on Discord as a filehost will no longer be possible by the end of the year
I don't think that's the right conclusion to draw at all from the Freenode debacle. A pretty significant amount of the channels stayed on Freenode after Andrew's takeover, until Andrew kicked them off himself with the mass Libera purge. There were many communities still on a wait-and-see approach -- at least until Andrew was revealed to be semi-insane (Emperor of Korea, digital territory, alt-right conspiracies, etc).

Yes, the open source privacy crowd is a big chunk, but there were just as many social communities. Don't you remember the olden days of sex channels and all their drama? The roleplay, alt history, sci-fi, and just general random niche hobbies that populated the IRC landscape. IRC was, above all, a social space, it just so happens that many of these communities died or migrated to discord over time. Those that remain are those who are most attached to privacy/open source/etc, all the rest are gone.

e63f67dd-065b··on Pixel 8 leak promises 7 years of OS updates
I'm pleasantly surprised -- this makes pixels much more easy to recommend to relatives. Still not the top choice over iphones, as those are just much easier to maintain as the designated family tech support guy, but at least it's not completely out of the picture.
e63f67dd-065b··on I’m a climate scientist and I think society will collapse by 2050. I’m preparing
Doomers are plentiful, he's really not that special. His points are just plain wrong and the conclusions are doubtful.

> Grimly, McGuire likens the human population to bacteria in a petri dish. When these bacteria multiply, uncontrollably, their number becomes so big that they can no longer sustain themselves in the dish. At this point, they die

> There’s one projection that by 2050, crop yields may be down by up to 30 per cent, at a time when the world will need 50 per cent more food [to account for population growth],”

Oh wow, I take what I said back. He's even more wrong than the usual doomer.

For one, we're not running out of food. The old idea of a planetary "carrying capacity" for humans has long been busted: technology progresses and makes food production more efficient. We might run out of good agricultural land, but that just means we need to clear out more or shift to hydroponics. As the price of food increases, that's a signal to the market to devote more resources to growing food, making more intensive methods of farming possible.

> Now in the countryside, McGuire’s house has walls that are two-thirds of a metre thick, to keep the heat out

Doesn't more insulation keep the heat in? Traditional construction in the tropics where I come from are not giant insulated blocks for good reason, that's how you die from overheating yourself without air conditioning.

> Then there is the water supply. He says it is important to be self-sufficient. “I am doing a lot of rainwater harvesting. The summers in Britain will be seeing consistent temperatures of 40 degrees plus, so water is going to be a huge issue,”

We won't run out of water the same reason we won't run out of food. Once the cheap water is gone, you turn to more expensive methods -- drilling, then desalination. Of course it's more expensive -- that's the whole point -- marginal increases in cost allows for more expensive methods of production. The same way increasing oil prices allows for deep sea drilling, increasing water prices will allow for desalination and other such technologies.

I really don't know why people keep thinking we'll run out of food and water. We won't. It'll get more expensive as cheaper sources run out, sure, but we won't run out.

e63f67dd-065b··on How Much Does It Take to Be the Top% in Each U.S. State?
Do you have any statistics to back this up? The census shows that top percentile household wealth at 11m, and top percentile income at 500k.

At current interest rates, 11m in cash would yield 500k/yr in interest, so the top percentile in net worth is also top percentile in income, just from first principles. Of course, over the last two decades interest rates were hovering mostly around 0, so that math doesn't quite work out. People also have their wealth tied up in investments, some of which fared better (S&P 500), some of which did worse.

Note that it doesn't really matter if that net worth is in real estate or in equities, both REITs and the broad market indices have returned about 11% per annum over the last 20 years.

More than one in ten american households have net worth >1m.

Your last sentence is more concerned about social status than about wealth. Poor families can spend in high-status ways, and vice versa. Social status is weakly correlated with both income and wealth at best.

You can't just point to billionaires and see that they have almost 0 income. There are only a few thousand billionaires in the world, enough for each and every one to be individually tracked. They are not included in these statistics, because again there's only literally thousands of them. They're so far into the top percentile they don't touch the statistics here.

e63f67dd-065b··on The US Military Is Getting Smaller, Cheaper and Smarter
I’m not well versed on the current state of military procurement, but this kind of vaguely resembles the NGAD idea/programme that’s been going on for many years now.

The article is also incredibly light on details.

> How would Replicator work on the battlefield? The initial step might be mobilizing two separate swarms of small, unmanned vehicles. The first group, numbering in the tens of thousands, would be focused on surveillance and reconnaissance, sending back uncountable millions of data bits to form a precise targeting picture. Second, the battlespace would be turned over to hundreds or thousands of vehicles large enough to accommodate payloads of explosives. Working alongside them would be drones carrying out cyberattacks to blind the enemy, effectively “cloaking” our own forces while destroying an enemy’s fighting ability.

Isn’t this already how the US mostly conducts operations, but without the tiny drones part? Take advantage of overwhelming air, intelligence, and logistical superiority to quickly neutralise enemy targets. I believe it’s literally in the latest edition of the released doctrine documents: use the clear superiority of US combined arms to coordinate multi domain operations.

This really feels like a mostly fluff piece without any real substance. I haven’t seen any major evidence of a doctrinal change to smaller drones and platforms; NGAD seems alive and well, leaning in traditional American technological and organisational capacity as a core part of doctrine.

e63f67dd-065b··on A case for dynamic scoring of high-skilled immigration
I recently read this https://www.forbes.com/sites/stuartanderson/2022/08/22/what-... by Forbes, which pointed out that the popular idea of exempting STEM PhDs from green card caps is primarily procedural:

> In July 2022, hopes were high the FY 2023 defense authorization bill would include an amendment on green cards for individuals with Ph.D.s in science and engineering. In what has become a familiar story, it was not to be.

> “According to a Congressional source, the House Rules Committee did not rule the amendment in order because the Congressional Budget Office (CBO) said the provision would cost $1 billion over 10 years,” as reported in July. “To address the issue and offset the cost, a $7,500 fee was added for the individuals who received permanent residence under the provision. However, the House Ways and Means Committee said the fee could not be included because it amounted to a tax and, therefore, violated Clause 5(a) of Rule 21 of the rules of the House of Representatives.”

That's just the first order reason. The second-order cause is that there is just not enough political will to get this over the finish line -- nobody wants to spend enough political capital pushing it through all the little things that can stop a bill, so a random reason pops up every time. All the hurdles here can be overcome -- just nobody in a position to do something cares enough about it compared to all the other legislative priorities they have.

As for the reason no congressman wants to spend political capital on this subject? I don't know enough about congressional politics, internal or external, to say.

The primary takeaway I have about this whole saga is that it's not enough for a bill to be without much opposition -- it must have people actively pushing it through all the little vetoes that different corners of congress built into the procedure of passing bills. It's not opposition that kills bills -- it's indifference.

e63f67dd-065b··on Ask HN: How to handle Asian-style “Family name first” when designing interfaces?
As others have mentioned, this is impossible to get right -- for the simple reason that there is no consistent order -- not for the same country, the same geographic area, the same culture, or any other marker you can possibly think of.

Example: in Malaysia and Singapore, Chinese people can have traditional Chinese names -- SAA for surname and 2 characters that's their "given name" so to speak. Some have an English name -- but it's written ESAA. Some only have one character names, so it's ESA. Some surnames have 2 characters, so ESSA/ESSAA.

When filling out US forms, ESAA usually turns into Englishname Surname, and SAA turns into Givenname Surname. So the display order for the first is firstlast, but the second is lastfirst. Same country, same culture, different display order depending on whether their parents gave them English names or they chose one later in life.

When there's a middle name field, often ESAA names turns into E AA S - first middle last. This display order is wrong. It should be first last middle.

As you can see, even within the same country, same culture, same neighbourhood, you still don't have a consistent order. It's simply not possible to do this in a consistent way.

e63f67dd-065b··on CLI text processing with GNU awk
Perl is much more terse for one-liners and has much more built-in for doing text processing in scripts. Stuff like implicit read loop, field separation, etc. I would say they're suitable for different jobs: if a perl script grows beyond a hundred lines (you can do a surprising amount in that space!), then Python may be the right tool.

Perl is also much more of a known target: some version of it exists on basically every single Unix, and the language really hasn't changed that much in the past decade. I have SSH'ed into multiple CentOS 6/SLES 11 (released 2009, and granted mostly to rescue data off them) servers in the past 2 years, and perl is just much more of a known target to write things against than whatever python release is on that system.

e63f67dd-065b··on CLI text processing with GNU awk
> free read loop

perl -n

> free field splitting

perl -a ... $F[1/2/3/etc]

> pattern/condition matching model

Not quite sure what you mean, but `perl -lane 'print if /abc/'` is might be what you're looking for

The boilerplate can be mostly eliminated with the magic incantation of `perl -lane`. The trick that makes all this work is that perl defines a whole bunch of pre-defined variables and populates them with things that might be helpful (see $_, @F, etc).

e63f67dd-065b··on CLI text processing with GNU awk
The old sysadmin in me says to forget both and just learn Perl.
e63f67dd-065b··on CLI text processing with GNU awk
Perhaps my old sysadmin hat is showing through, but I don’t quite see what the advantage of awk is over just writing the same thing in perl. I’ve seen my fair share of horrendous shell scripts from junior sysadmins, and every time I think to myself “the text processing portion would be so much cleaner in Perl”.
e63f67dd-065b··on Who is driving Germany's far-right poll surge?
That’s why I’m, unlike most pro-immigration people, very pro assimilation —- language proficiency tests, an intentional strategy of educating immigrant young in public schools, dispersing large immigrant diasporas throughout the country, etc. I agree that immigration without at least some policy of assimilating them is a recipe for perpetual social unrest and instability.

In the US, successful assimilations look mostly like this. Kids go to American public schools, start forgetting their parent’s language and culture, leading them to look pretty much like every other American. The Chinese do it much more harshly without a drastic increase in efficiency, so the soft route seems to be the sweet spot. There’s no reason to think that you can’t assimilate them. Discourage mosques, try to disperse high concentrations of non-assimilated cultures, it’s not new and it mostly works.

No policy lasts forever. We’re not looking for the next thousand-year solution to the demographic crises; immigration is the 100-year plan, so to speak. We can’t — and shouldn’t —- try to write policy for the next thousand years. There is no other area in which we demand solutions for more than a century. Eventually this policy too will wear out and our descendants will have to go back to the drawing board.

e63f67dd-065b··on Who is driving Germany's far-right poll surge?
Going by current estimates, immigration can plug the demographics gap (by accelerating the problem elsewhere mostly) for a minimum of 100 years, probably 200 and maybe even 300+ depending on growth patterns and economic developments.

So yes, it’s going to be a solution for as long as America has been a country, France as a republic, longer than Germany and Italy united, more than many modern countries have existed. If that’s not a long-term solution, I don’t know what is.

There are no thousand year solutions here. We can only pass the problem to our (many-times) descendants and hope that they figure it out.

e63f67dd-065b··on Who is driving Germany's far-right poll surge?
The death won’t be natural or peaceful. You can see this in current EU politics already — seniors expect pensions sufficient to maintain their existing standard of living, but this assumes that there exists the same amount of working people to sustain a sufficient surplus for the elderly to consume without meaningful contribution to the economy. What your policy actually means in reality is a well-known pattern — increasing taxes on the young to shift their consumption power to the elderly, as the elderly become an increasingly larger voting bloc, perpetuating the cycle. There’s only a few ways this can end — 1) retirees must have their living standards decrease as fewer working adults sustain them, 2) they maintain their standards at the expense of taking it from the young, who 3) will leave for other countries and make the problem worse, or 4) retirees exit retirement and actually work. You’re seeing the fight between options 1/2 happen in France, 2/3 in Spain, al four in Germany, etc.

The option that policy makers see is 5) bring in enough migrants to maintain healthy demographics and 6) making more domestic babies. 6 is, however, a mirage; no policy we have tried worked (see SK, HK, SG, TW, JP, and many more).

e63f67dd-065b··on Learn AutoHotKey by stealing my scripts
It won't run the program because the code is not correct, the same way a C compiler will refuse to compile a missing semicolon. If different indentation changes the meaning of a program from wipe the disk if condition true to wiping regardless of condition, that's a pretty bad change, so the interpreter rightly refuses to run incorrect programs.
e63f67dd-065b··on LK-99: Team of Southeast University observed zero resistance below 110 K
You're exactly right on the extraterrestrial part: the speech to text transcribed it as "最近我们这个有关世外超岛的话题非常的火热"
e63f67dd-065b··on LK-99: Team of Southeast University observed zero resistance below 110 K
The actual video is here: https://www.bilibili.com/video/BV1pM4y1p7u5

Short summary (not a physicist and consume most of my science in English, so may be slightly wrong):

- The material is fragile, and hard to form into a regular shape

- There's a weird drop in resistance around 230K; could be an artefact of the measurement process or instrument

- On 1 Aug they found a sample with almost 0 resistance, which got them really excited and led to them searching for more

- They tested 6 samples, only one of which exhibited zero resistance at below 110K

Basically, results are inconclusive, but it's promising.

e63f67dd-065b··on Tarsnap outage postmortem
Can you say a bit more about the log-structured S3 filesystem? I wrote something very similar recently (https://github.com/isaackhor/objectfs) and I'm curious what made you settle on that architecture. The closest thing I know of that's similar is Nvidia's ProxyFS (https://github.com/NVIDIA/proxyfs)
e63f67dd-065b··on Sinkholed: A DNS Horror Story (2019)
This is actually terrifying to me. If my domain goes, so goes my email -- everything is @myname.com, so I won't receive any email. If my email is gone then I just instantly lost access to a big chunk of the internet services I use, and my online identity might as well be dead. Sure I still have my work email but that's no consolation for all the services registered to my domain email.

I'm not sure what the right course of action here is to mitigate such a risk. Add backup emails on a different domain registered with a different registrar to services that support such a thing? Stop worrying because it's rare enough that it's a waste of time? Is this what happens to people that get banned on gmail?

e63f67dd-065b··on Stanford Graduate Students Won Their Union Vote
Do you disupte any of the things I mentioned? Yes, obviously PhD students with kids and health issues exist, but on average, PhD students are younger, healthier, pay less taxes, qualify for more benefits, and don't have children. The MIT estimate is based on all the above not being true, but that's simply not reflective of the PhD student population.
← PreviousPage 5 of 11Next →