The CS job market is very location specific, so I don’t have much advice in that regard.

I’ve got quite a bit of experience with Linux and I manage a home server, but that probably doesn’t differentiate me much.

That said, I have been on my company’s rotating interview panel for about a decade now, and while my company’s line of work involves a lot of Linux development, I can say that most of our college hires do not possess very much Linux background at all. Sure, they might have used Linux machines for school projects, but rarely do any of them assert to be “experienced” with Linux.

By that, I mean deeper knowledge than just using Bash. If a candidate can tell me why they prefer csh over Bash, or any syntax difference between POSIX sh compared to Bash, that is definitely a distinguishing quality. It speaks of an operator who has enough usage under their belt that they’re annoyed by the typical distro’s defaults, and more importantly, assessed the available tools, and picked the right tool that works for them.

I cannot understate how valuable it is to us to find a candidate that understands their tooling, especially right out of college. Considering that we assume most new hires have to be brought up to speed over the first few months, a candidate that saves us that effort is at least one rank above their peers.

Deeper functional knowledge comes in other forms as well. It’s one thing to know how a C program’s main() function is invoked by an OS, but anything which shows a fuller understanding of, say, system architecture and how a timer interrupt leads to a context switch in an assembly ISR, to a returned service call to load an ELF, to a CPU privilege ring change, to crt0, to main(), that is another level entirely.

I’ve interviewed candidates that had side projects involving retro game disassembly. So maybe they couldn’t give me the above level of detail for x86, they could describe the same for MIPS. And that’s good enough, because most architectures do roughly the same thing, with a few different semantics and names.

Circling back to managing a server, if you had to deal with PAM, NAT and port forwarding, tunnels and VPNs, compiling from source, or abything like that which is non-trivial, do not sell yourself short. All that stuff is resume material, because if you can relay to an interviewer that you’ve dealt with real network or machine security tasks, it is distinguishing.

The best part is that you have all of college to learn the CS curriculum, but it’s also time that you have to pursue any particular focus that excites you. I’ve written earlier about how embedded engineers don’t really get caught in the hype cycle, so jobs don’t suddenly appear then disappear a few years later. If you wanted to do that route, getting started with any microcontroller (eg Arduino, STM32) would help, with a goal to understand all the “magic” that the IDE and compiler are doing. Maybe instead you like das blinkenlights and find yourself drawn to hardware design. It wouldn’t be too late to consider a switch to the Computer Engineer (CE) major, so you have a small taste of the EE life.

CS as a field is so large that there are many routes between “I want to work with computers” to a declared major and to a career thereafter. Fortunately, time is on your side; this would be a very different conversation if you were a 4th year college student.

Humans that successfully visit would be honored as a Discman.

IIRC, the IAU’s definition of planet – infamously applied so that Pluto fell off the list of Solar System planets – requires that a candidate planet be large enough that its own gravity is strong enough to force it into a rough sphere, whatever it might be made of.

So a disc-shaped planet could not ever meet this criteria, because if it were made of something strong enough to remain a disk, then it’s too small to be a planet. And if it did exceed the critical size for gravity to make a sphere, then it wouldn’t be disc shaped anymore.

But setting that definitional quibble aside, we will focus on sizes and materials that allow a disk shape object to exist and be large enough for humans (or Mario) to visit. So no Wensleydale cheese. If we say that this object is mostly uniform in its mass distribution, then it would have to be the case that for any disc shape (including cylindrical), different points along the surface will be farther or closer to the center of gravity. Thus, inhabitants would experience gravity differently depending on where they are.

Note that we haven’t even considered whether the disc is rotating. If it is, then there’s a chance that the centrifugal acceleration at some points will completely negate the gravitational acceleration. At such points, one could hop up and off the surface, linger for a bit, and then get pulled back down once the disc has rotated to a position where there’s a net force upon you again. Or if the centrifugal acceleration is too strong, it might repel visitors on the surface altogether.

Alternatively, there would be a danger of playing on a trampoline that accidentally crosses into a net-zero gravity region. Here, a double bounce could send someone very high up, only to then plummet back down to their death when re-entering a downward gravity zone.

I have almost no citations for the above, but I thank you for posing an interesting question.

I agree that RPM isn’t the primary quality to assess here, but I think angular momentum doesn’t really matter either. As a quantity, momentum describes a conserved capacity to store kinetic energy. As in, it could tell us how long the fan would stay in motion when the wind stops blowing. But that’s not the objective for a wind turbine, which is supposed to convert lift from the wind to do some work.

At bottom, none of these quantities are the useful metric for what makes a good wind turbine. And that’s expected, because we have few details about the fan itself: the blade pitch, stall speed, and fan diameter, to list a few parameters. The only firm detail we can see is the blade count from the picture.

BCP 177 consists of: RFC 6540 : IPv6 Support Required for All IP-Capable Nodes

Given that FOSS licenses are premised on copyright, yes, the same ails would still exist: 1) AI washing of licenses (including transforming one license into another), and 2) the vagueness of whether LLM outputs can be copyright, which threatens the validity of a FOSS license upon that output.

The first ail can be seen even without LLMs: the BSD variants have gone through great pains to remove GPL-licensed code from their base repositories. This basically involves reimplementing utilities and functionality from scratch, using only the ideas that are in common with the equivalent GPL code, but never copying that code directly. This is properly considered a reimplementation, which can then be licensed permissively (eg MIT license).

If an LLM were to train on GPL code but the output were licensed with MIT, then that could be a GPL violation because GPL mandates that remixes continue to keep the GPL license.

Maybe you could avoid this fate by limiting the LLM to only train on permissively licensed code. So that it would be permissive licenses going in, and permissive licenses coming out. No GPL problems here. But that brings us to ail #2.

Some jurisdictions have rules against granting copyright for computer-generated works, in the same vein as works generated by non-humans (eg a macaque). If this LLM fell into this situation, then the output is not copyrightable. And if there is no copyright, a license like MIT or GPL simply cannot apply, because its terms couldn’t be enforced.

Well, to be clear, the copyright parts of those licenses would be unenforceable. Some parts of the license may still be enforced under a contracts claim. But in any case, the things we refer to as “FOSS licenses” cannot attach to uncopyrightable works (with the possible exception of the CC0 license, which is essentially the absence of any license whatsoever).

EDIT: you did say “consenting projects” and consent is key. If such consent came in the form of a license grant, then yes, that would be enthusiastic consent for the LLM to generate output, which solves ail #1. But for most multi-person projects, getting consent from everyone is difficult or impossible. The Linux kernel is one such example, having so many contributors that some of them are already dead. Death means they cannot consent, but their copyright grant lives on. And so practically speaking, obtaining enthusiastic consent for whole projects is a challenge, which drastically limits the prospects for such an LLM from the very beginning.

There’s a fascinating piece of banking history embedded here, since for a fairly long time, the economics of USA banking meant that long-term banking relationships generated the most revenue for the bank. So they would go out of their way to establish new relationships that would hopefully endure for decades, thus keeping the bank in business. This applies even if a small percentage of those attempts would be losses, because the average was still mostly upsides.

Today, the banking landscape is much more transactional, where lots of people have no qualms jumping ship to an online bank. Sure, not everyone will go through that inconvenience, but enough will that it upsets the prior calculus that prioritized long-term relationships to achieve long-term value. That of course means that there’s less upside to extending a $1k unprotected credit line and all the same downsides. Hence, credit issuers are more stingy now than in the past.

The obvious problem that this exacerbates is that people who never had access to banking have poorer chances of ever getting served once they do need banking services, because that first account needs to meet strict credit approval requirements or whatever. The very fact that a steady 4.5% of American households can’t meet the bank minimum balance requirements, while “non bank” services like payday loans and buy-now-pay-later are on the rise, is a reminder that opportunities from yesteryear are disappearing rapidly, concentrating to those that already have plenty of opportunity.

Past institutions and past people did what made sense at the time, but it’s quite clear that in the present, the situation is not serving the people’s needs. The corporate desire to grow bank market share or shareholder value is precisely the opposite of what sustainable banks should be doing: right-sizing their scope to match what their clientele can afford. If a bank can only stay in business by having a $1500 minimum monthly balance, then they should close and a smaller, nimbler bank or credit union that only needs $400 to be sustainable should take their place. Banks grew to match the needs of the public. If the public’s needs have changed, so must they.

IIRC, there’s a difference between entries expiring from a credit report, and whether a debt is still legally collectable. If the holder of a debt can prove the debt is still outstanding, or has it ratified as a court judgement (and renewals the judgement, as local rules may require), then it may still be pursued once someone returns to the personal jurisdiction of the court.

The thing is, though, since the original creditor will have wanted to get something rather than nothing, they may choose to sell the debt to a debt collection service, conveying the legal rights to pursue the debt. In return, the original creditor gets a fraction of their money back. But even the most prodigious debt collectors may give up and sell the debt to someone else, for pennies on the dollar. As the value of the debt keeps shrinking, those fixed costs to keep the debt collectable (eg maintaining paperwork, renewing a judgement) become disproportionate and thus skipped.

At the very end of the chain are debt collectors with such a huge catalog of bad, uncollectable debt, but they’re hoping they can schmooze or bamboozle the debtors into paying one day. As with all debt collectors, they have no right to add renewed entries to someone’s credit report, so the debt is essentially in limbo: unenforceable yet still on paper.

Your profile bio mentions Arch, Thinkpad, and SDRs. Just from that, you can likely expand your skills section to spell out the operating systems that you’re familiar* with (including others like Windows or Mac OS, if you’ve used those), possibly with inclusion of anything related to coreboot on Thinkpad (if you’ve done that), and can include radio technologies in its own section under either skills or interests.

Don’t sell yourself short, because if a job posting is upfront that they’re not looking for hard skills as a prerequisite, then the assignment is to present yourself as a candidate with depth. They’re clearly looking for candidates in a non-conventional manner, so you should showcase all your non-conventional aspects. Don’t be afraid of including “niche” details like RTL-SDR, because they can easily do a web search to figure out what that is. But do contextualize it as a tech interest, because if maybe the job requires working with complex, one of a kind tech, then they might value the candidate that self-learned.

Regarding “familiarity” on a resume, to be “familiar” with XYZ just means that you at least know what it is and have probably used it. Listing out some skills on a resume means, at least to me, that you’re familiar with all of those skills. It’s different if you’re actually an expert or fairly accomplished with something, in which case you’d describe that skill in detail, usually giving an example that you’ve worked on. I recommend including everything you can think of, even the mundane stuff like Microsoft Word; you’d be surprised how many people don’t know how to use Word or Excel.

I’m also poised to agree: setting the output air regulator to a low value does not change the reserve pressure in the tank. I suspect maybe Peppycito thought OP was talking about the governor, which is what controls when the pump turns on and off.

But even then, the most common design of governor only adjusts the peak reserve pressure (aka when the pump turns off) but the minimum reserve pressure (aka when the pump turns on) is usually a fixed value subtracted from the peak reserve. Usually the control for this is inside a box and turned by a screwdriver, because of how rare it is to adjust the reserve pressure.

That said, there is genuine merit in reducing the reserve pressure, if there’s an explicit objective to prolong the life of the pump as much as possible and the air loads are not very significant. But this usually isn’t the case for small residential air compressors, because those pumps tend to be built very small and thus oversubscribed.

IANAL. If a bank signs the copy they receive back from the counterparty, then generally yeah, it would be enforceable. Contract law demands that result, because otherwise there’d be no point to the words on the document. Working in the other party’s favor would be any lack of notice shortly after the bank signs the contract, because the longer the bank takes to notice a problem, the easiest it is to prove that they did not exercise the appropriate care when signing, incurring all consequences as a result. Missing more and more opportunities to recast or renegotiate the contract, that’s a poor position to defend.

That said, the other party must not have made any implied or explicit statements that the bank could have relied upon. Returning the modified contract to the bank in an email with the words “please find the signed contract attached” (my emphasis) could ambiguously imply that the original contract has been unmodified except that it has gained the other party’s signature.

Whereas the words “please find my signed proposal attached” would dispell any and all ambiguity, because it would clearly be a proposed contract by the other party, not the bank’s original proposal. Alternatively, a blank email with just the attachment is no statement at all, implied or explicit.

To be clear, a bank would almost certainly contest the contract, even if they don’t have a leg to stand on. And the usual reason for this – besides litigiousness – is that it’s the only way for the bank’s business insurance to pay out. Or at the very least, an attempt to slightly lower the damages by opening settlement talks.

This looks amazing. That said, you may want to have a look at rules 7 and 8 for this community.

Contributors: Claude

Please kindly see Rule 8 of this community and its associated post.

I’ll offer the contrarian answer: FOSS does not guarantee secure software; in-fact, neither does proprietary software, nor government software, nor anonymous software.

As others have mentioned, assurances about software come from audits or by trusting someone reputable who has done the audit. Delivering security guarantees is not what most FOSS projects are meant to do.

So what exactly do FOSS projects do? Why are they any better than proprietary software? The short answer is that FOSS is about continuity. You can and will find FOSS projects from 30 years ago, which have been kept updated so they can run modern machines. The folks doing that didn’t need anyone’s permission to do that; they can just do it.

Meanwhile, if Adobe of MSFT declare that a certain proprietary software suite is going EOL and will not receive any more security updates, then the user base is SOL. FOSS gives a potential route away from this fate, if someone or some group is willing to put the elbow grease into security updates. Even if it’s just one person.

So in the short term, there are no security assurances for either proprietary or FOSS. In the long term, all software cannot escape the unstoppable march of time and vulnerabilities. But at least FOSS has a chance to be corrected, years or decades later.

For handling international traffic under ICAO rules, ATC is indeed supposed to be done in English.

But if we take a look at some of the terminology in aviation, the French words cannot be ignored. Mayday and pan-pan are from the French venez m’aider and panne.

When radio silence is required, a station might command “SEE-LONCE” but this comes from the French word for silence. Though I think this is a rarity in aviation.

And then outside of aviation radio, the French left their mark on aircraft parts names: fuselage, empennage, aileron.

Wine? https://en.wikipedia.org/wiki/Wine_(software)

An intriguing question. I do agree that even the ultra-rich had an eye-opening time during the pandemic, but I don’t agree that their wealth acts as a singular shield, which if pierced would herald their downfall.

When someone has the resources of tens of billions of dollars, the principle of defense in depth is both practical and is most prudent. For each of your points that suggest the ultra-rich were vulnerable, it might actually prove the strength of their fortifications.

they were conceivably vulnerable to the same indiscriminate illness as everyone else.

The thing with disease is that it cannot teleport: the transmission vector must convey person to person, through airborne particles, or some other physical means. What the ultra wealthy can – and do – buy are exotic islands or chalets in faraway places from other humans. Meaning that once there’s notification of a pandemic, they are best positioned to flee to whichever of their overseas fortresses as they may choose. Meanwhile, the masses have to hunker down at home, which is their only refuge, and hope for the best.

the government can and will shut down your business

The ultra-rich leave the daily operations of their businesses to professional managers. It has been a staple of prudent management for decades to have disaster preparedness plans and business continuity plans. You may have seen some form of these through wildfire or earthquake notification systems, so that a company can confirm which of their employees are accounted for. But such systems also serve as a planning exercise, in case a natural or manmade disaster takes out an entire industry.

The businesses of the ultra-rich already plan for things like the loss of most of the world’s hard drive production capacity due to flooding and the closure of EU airspace due to volcanic ash. These aren’t as detailed as those specific situations, but are about the business impacts: what it products can’t get to customers? What if required materials can’t reach the factory? What if war breaks out domestically and the production line is taken over by the government?

Not to minimize the impacts of the pandemic, but the difference is that it activated multiple continuity plans simultaneously, an unprecedented scenario but otherwise not unrecoverable. Fortune favors those who make plans.

Indeed, a part of many business’s plans during a catastrophic situation is to – unsurprisingly – beg the government for aid or a bailout. After all, if they can get the taxpayer to partially implement their plan, they will. And business lobbying for a bailout only requires a telephone and a contact list, so the ultra-rich’s lobbyists were kept busy in the summer of 2020.

What is most telling is that the USA stock markets recovered in the latter half of 2020. The ultra-rich have plenty of resources to survive a few months of disruption, as they have more runway to wait things out than, say, a typical working class household that needs rent relief or faces eviction.

universal income model is not only possible, it is completely affordable and can be quickly implemented.

One of the most under-assessed aspects of the ultra-rich is that they find ways to make money whichever way the wind blows. Even in a UBI scenario, they can still make bank if they are the sole vendors of certain commodities.

In fact, a UBI system which automatically tracks inflation is essentially a license for vendors to also increase their prices with exacting precision, perfectly in-time with inflation. This is why a UBI system should not be implemented on its own, but alongside other social safety nets and regulation on the “demand side” of money. That is to say, business regulation (eg rent controls, anti-monopoly rules, banking reform) are all part-and-parcel of a long-term plan that defeats economic inequality. They are not separate pieces, but the ultra-rich can still take solace that not everyone recognizes this yet.

The war of information would still be on their side even if UBI were voted into effect today.

The luxury services that their money had always effortlessly bought could be quickly ended by decree, scarcity, and the loss of human labor to illness.

I didn’t exactly care to track the happenings of the ultra-rich during the pandemic years, but did they actually suffer such scarcities? Their business empires recovered by the end of 2020, and it’s not like rich people don’t just find other intrigues to spend their time on. Rhetorically, what is a few billion here or there?

Were the ultra-rich terrified? Probably, but only for their paper valuation. They themselves would have been physically fine, and since there was no proletariat uprising during this time, they basically just made money during those years. Essentially, a financial blip but their lives move on. Maybe they’ll build more bunkers for their holiday homes.

In the space of mini PCs/NUCs, this does seem comparable. But bear in mind that the entire space of “home server” is a lot bigger than NUCs. After all, not everyone just wants to host an NVMe drive or two; they might want to fit some cheaper SATA SSDs or even spinning rust, as part of a cascade of stroage, with caching on SSDs and other nice things.

Likewise, the dual LAN ports with 802.3bz support is nice, but the reality is that most networks are only 1 Gbps with some devices supporting 2.5 Gbps. Quite frankly, for server use that actually needs one or both port’s bandwidth, the link would likely be better served with SFP+ slots, because of better compatibility with cheap enterprise switches available online, and because SFP+ transceivers run less hot than their twisted pair counterparts at 10 Gbps. Heck, you could even use a 40 Gbps QSFP+ NIC, which only requires a x8 or X4 PCIe slot (depending on PCIe generation of the NIC). On that note, a single PCIe slot is nice, but it does preclude the use of a double-wide GPU for transcoding use.

I would sum all this up as lacking expandability. Yes, it can be expanded somewhat. But for folks that want the capacity of an ATX-based build, this ain’t it. And for folks with abundant solar power, electricity consumption for older gen chips isn’t as much of a dealbreaker. It certainly checks off a lot of people’s boxes, yours included, but that doesn’t mean that people are “sleeping on” this, because at the end of the day, if it doesn’t serve their purposes, it’s not the right box for them. In terms of price, never underestimate the cost of $0.00 that is an old, unused laptop; they make remarkably decent servers for light loads.

Also, as of this writing, only the 32 GB RAM variant is available on the USA website.

Would cert-manager for Kubernetes be approximately what you’re looking for? In that particular case, it’s a service that runs in a K8s cluster, that can auto-request CSRs and send them to Lets Encrypt using one of the challenge types. It takes care of managing the returned certificates, such as renewing them before expiration and registering the private keys within the K8s key store.

I presume that you’re not already using K8s, so maybe that project will help you find the keywords to describe the sort of certificate manager that you want.

If you’re already using Debian on desktop, then Debian on a VPS makes sense from a maintenance perspective: you only have to track one set of security updates. When there’s suddenly a critical update for some vulnerability, you can update both at once because it’s the same upstream repos. Whereas with Ubuntu on your VPS, you’d have to be mindful to make sure to check back when the same vulnerability is patched there.

Do not underestimate the effort to upkeeping a server, and do try to unify your procedures so you’re spending less time doing the necessary things, and more time building up your self hosting stack.