Someone who makes this statement this really want us to equate it with racist chattel slavery - the most heinous form of worker exploitation in modern memory. "Oh, I just meant: {the most common type of work in human history}" as a retreat is not very compelling.

If you're unaware of who he was, or what happened to him:

* He was one of 12 others who helped to spec out RSS' standard (when he was 14 years old).

* He is widely considered a co-founder of reddit, after merging with reddit, he rewrote the majority of its codebase from Lisp to Python.

* He was one of the reasons Reddit protested SOPA back in the day (the same day that good 'old Kim Dotcom was raided and arrested by the US, the irony of which being that they swore up and down that they needed SOPA in order to do what they did to Kim Dotcom).

* I didn't know this before: he also developed the markdown publishing format with John Gruber.

What happened to him?

He was charged with felonies for stealing PDFs for publicly funded research papers, he was tried as if he hacked these systems. He took his own life as a result of them throwing decades of prison time at him for what's hardly a serious hacking offense (seriously, Kevin Mitnick was way worse). Kevin Mitnick served five years, they were looking to throw up to 35 years at him, and a minimum sentence of seven years if he went to trial.

It is abhorrent to me that we have people doing more malicious hacks that get no repercussions. Someone at OpenAI / Anthropic should be 100% liable for criminal charges of what their AI models do. Need I remind everyone, that these models can run way faster than what we common folk can see, the speed at which they run these models internally can go so quickly that by the time the model is discovered to have hacked someone, it could very well be too late. Just try the "fast" feature of Claude Code, now imagine Claude launching thousands of sub-agents at top speed internally, the capacity for unprecedented destruction is the most concerning thing to me, I say that as someone who uses Claude every week.

Somewhere along the way, tech stopped asking questions like that, and now, people are hostile to the suggestion that the question should be asked.

It's got a bright and colorful design where a Mac mini and any third party monitor would just be silver/grey + black plastic. Everything works together, no decision making about compatibility, no extra purchases, it has a webcam, speakers, one power cable, only one customer support call if you need help with any of it, one warranty etc. It's quite far from the "same" computer as a Mac Mini.

24” or even 27” feels like such a downgrade.

Also please make the iMacs usable as external monitors agains. Would buy one as a “livingroom computer” instantly if it was also possible to occasionally use it as a docking station for an air..

They do:

https://openjdk.org/jeps/197

https://openjdk.org/jeps/515

Not every optimization requires a JEP, but certainly something as big as "add or remove a JIT" would 100% require a JEP.

JITs are big nasty things that by their nature have to do a lot of things. You are basically saying "Give us a PEP for a new python interpreter and compiler".

What they do is fuzzy by their nature because optimizations are very rarely universally beneficial. It's a game of measuring, looking for regressions, and tackling the biggest problem next for whatever sample applications you have available. You always run into the risk of overfitting on top of all that (You may have made django faster, but you've made flask slower).

This is the sort of thing that, IMO, should not go through the PEP process. It's an implementation detail for the runtime.

Imagine, for example, if the JVM needed a bunch of JEPs for JIT improvements. They'd never get done. Or if any javascript engine needed ECMA acceptance before they could do a new JIT optimization.

On one hand, it seemed like JIT work was really progressing in high gear until SC requested a new PEP. Also unfortunate, as it seemed like part of the hope for the JIT was that it'd counter the performance penalty from free threading.

On the other, it did feel (to me) like with all the fancy (and definitely cool) optimizations it got (automatic refcount operation removal, new type-based specializations, inplace math operations), the macrobenchmark benefits weren't _that_ exciting for a JIT. It definitely has potential for more, maybe with even fancier optimizations like unboxing ints, but the complexity just keeps going up.

> It said that the PEP was "overly optimistic" about user-visible differences with the JIT enabled.

The JIT is technically generated from the same source as the interpreter loop itself, so originally the differences were extremely unlikely, but I imagine the risk just kept going up as new JIT-only paths and optimizations were introduced.

And politically, I feel like SC’s reasoning was just “this side project has landed in main branch almost by accident and has been stewing without oversight ever since, with no stabilization in sight”; and the latest PEP was probably still too vague, with a lot of “here’s more things we can add to the stew” - so I can understand if SC wasn’t satisfied.

Dan Guido, of Trail of Bits, claims they have an "universal" exploit [2]. Knowing their track record, I don't currently have a reason to doubt this claim.

[1]: https://stealmybitcoin.com/

[2]: https://news.ycombinator.com/item?id=50226010

EDIT: The HN title has been changed since posting this comment.

You are explicitly telling them "this is good / bad"...

The good ones are increasingly producing good solutions much faster than before.

The not so good ones, are doing the same bullshit they did before, but now the don't even bother to check it. For example one requirement to "not implement it in the frontend" led to 3 times PR rejected, because it still has been implemented (even if it's partially there) in the frontend.

I'm not sure what's the reason, but I assume it's

- inability to understand architecture choices and their consequences

- "don't bother me with shit, just pay me a salary" attitude

but the AI lead to amplification of the good and the bad in people.

The thing is it gets you the wrong answer so fast it’s now the best way to figure out the right requirements.

So the result is noticeably worse, but you basically must do it, because the efficiency gains are inescapable. This makes the work seem even more pointless, adding to the feeling that you're working more and accomplishing less.

(1) More tech doesn't save time if it's developed in the direction/dimension of more information flow. Because it seems that the best environment for creativity is a somewhat restricted environment of information. And if one needs to engage with information to take advantage of this new tech, it actually impedes creativity.

(2) A lot of the grunt work that we seem to be automating also encourages creativity. From what I've observed from many artists and programmers (I hang out with both), it seems that creativity strikes best when it's not forced, and the "downtime" doing grunt work actually accelerates it.

So I think that AI breaks these two principles (minimal information and the removal of the crucial grunt work), and actually makes us work more for less.

The only gains seem to be those short-term kinds (a program can be made faster, rewards can be reaped by senior people who have thought about stuff for a while), rather than encouraging a healthy environment of creativity....

> We'll be honest about where we stand. We don't like copyright as it exists. We'd like to see it radically reformed, and we wouldn't mourn it if it were abolished. Our community lives in its grey areas every day: many of the models on our sites are fan ports of characters from commercial games, made out of love and shared for free. When a studio sends us a takedown notice, we comply.

> While we as small artists and fan communities get copyright used against us, the world's largest companies are committing copyright infringement at a scale unseen in human history. This has been the case since the invention of the internet of course; digital media make duplication, remixing and resharing trivial, but copyright law never caught up. Filesharing stabbed copyright in the gut. AI is just twisting the knife, and completes the creation an unequal system where copyright infringement is only a crime if you're a low net worth individual.

Rich companies are infringing copyright at industrial scale in order to build their LLMs, and even if caught and challenged, they don't suffer any catastrophic punishment. The infringed party gets enough money to shut them up and everyone moves on. A fine is a price. They'll keep doing it with impunity. Meanwhile the small guy who doesn't have copyright lawyers on retainer is getting shafted, time and time again.

There are lots of reasons I believe this. One is that frontier AIs are very clearly superhuman on shorter problems (as can be seen on benchmarks like FrontierMath, or just by playing with them yourself if you have a source for difficult problems)--so these achievements are "just" scaling that advantage to research-sized problems. Second, AI is also clearly superhuman in its knowledge across fields, so it can make connections and combine techniques that humans won't be able to access. Third, these problems are famous and have withstood efforts of many of the best humans, in many cases being considered unapproachable.

Altogether it seems like AI is not only smarter but also more cost efficient that humans for math-adjacemt problem solving. I'm not saying that AI is superior at everything in math, eg theory building, but I think it's really unlikely you could get any group of humans that could compete with OpenAI's latest results.

1. Biology has a very nasty habit of not cleanly abstracting. It's not like we can treat every protein as a "diffuse mass" without losing a ton of important interactions. This is unlike something like CFD where you can treat a mass of water molecules with certain constants. Think of something like cancer: a small mutation to a DNA strand can cascade into a tumor. You have to deal with the mutation happening at an atomic level while scaling to a whole organism (yes I know there's models that can approximate this, but they can't capture all the interactions).

2. This brings us to modelling at the atomic level. To model a system of atoms with the Schrödinger equation, it naively scales O(6^n) per particle. Note that means every nucleus and electron, so if you have say 10 carbon atoms and 22 hydrogen you're already at 10^127 operations. We can cut corners a bit and end up with something like CCSD which scales O(n^7), or cut more corners and scale O(n^3) for DFT. DFT is pretty much the last method where bond forming and breaking is modelled, so after more corner cutting you lose losing reactions, which are important for understanding why something is happening. Most protein folding simulations (moreso prior to alphafold) run a very crude approximation that models atoms as balls on springs (things like AMBER and CHARMM), which obviously can't handle reactions.

3. If this wasn't bad enough, you also can only move through time at about 1 femtosecond (10^-15), so it takes a ton of simulation steps to get even a nanosecond in. Forget millisecond or second lengths where things start to happen.

So, while the math of the physics exists, computationally it's practically intractible. There's some exciting work with something called Machine Learning Interatomic Potentials (MLIP) which approximate DFT using a neural network which are about 1000x faster than DFT, and iirc scale much more linearly (look at things like MACE and orbmol if you're curious). So we're making progress, but we've gone from so ridiculously impossible to pretty much impossible for large systems.

I have a couple ideas about developing custom microchips to make this faster, which is why I'm pursuing this, but there's still a massive way to go!

Thanks. I think the very response you wrote here is a better form of marketing for the software engineers than what openAI puts out there. When everything is groundbreaking and everything tops charts, nothing is.

I suspect a lot of target demographic (developers) prefer simpler and honest communication about not just the models high points but low points too. With the linked Terra post, if I saw 1 chart that clearly highlighted the usecases for which Terra is _worse_ than higher tier models, I would be more convinced overall.

Everything trying to occupy the top spot in my brain's limited "top model" representation space is not helpful.

What's confusing to me in addition to the naming is that there are no published guidelines I can use, as far as I can see. Put yourself in the user's shoes. On what basis should one pick a model for a given task?

I can understand smaller models being dumber and faster. That aligns well with scenarios where you are willing to compromise; e.g. you're classifying content or asking a model to format text into a table.

But when you want "maximum intelligence, cost and time be damned", there's the implication that bigger models are smarter on a linear scale, and every provider's model release naming is currently conflicting with that expectation.

This isn't just a naming problem, but a communication problem. If I knew that Astra was the best at coding but a bit bad at writing, and Sol was almost as smart but better at technical writing, then that's the kind of thing I expect to be documented.

One possible way: A simple matrix with horizontal columns like "writing", "coding", "math", rows are the models, and each box is a number of stars from 1-5.

Instead, this sort of thing becomes lore, shared by word of mouth.

In general, LLM performance is incredibly multidimensional. There's no guarantee of a fixed ordering where the model that is best at math is also the model that's best at jokes. Despite that complexity, much of the variation in performance is explained by just a single variable: model size. In general, models that are bigger will (a) be more performant / reliable, (b) cost more per token, and (c) have better recall of esoteric facts. As you note, our models are ordered:

Luna < Terra < Sol < Astra

Even today, I'd recommend this as a general starting point for picking models. If you care about quality only, pick Astra. If you're a bit price sensitive, pick Sol. If you're very price sensitive, pick Terra or Luna.

Of course, model size doesn't capture all variation. Models that are trained later might have new data or new techniques or new architectures that earlier models lack. You can see this with models from all providers - Sonnet 5.5 outperforms Fable 5.1 in ways, and Gemini 3.8 Flash outperforms 3.1 Pro in ways.

In the case of GPT-6.1 Terra, I'd say Terra is probably better than Astra and Sol at technical communication, similar on vision and computer use (but much faster!), and a bit worse on most other things. For API developers, it's a nice midrange option if you want something cheaper than Sol but a lot smarter than Luna (especially for vision / computer use / communication). For subscribers, it's not a bombshell release - if you're hitting usage limits, maybe it helps you go a bit further, and if you want clearer communication or faster computer use, it can be a nice option to select.

If you have suggestions on how to name our products better, I'm all ears. Our intention is to make things clear and simple, even if we miss the mark sometimes.

Unfortunately, a lot of the complexity is somewhat irreducible and due to to the actual complexity of the products. Scaling model size is different from scaling reasoning effort which is different from scaling multi-agent count. So if we just swept all of those under a single name of "more IQ", it would make the naming simpler but the experience much worse, as important details would be obfuscated from end users.

First we had:

> Luna (Small), Terra (Medium), Sol (Big).

Which was fairly smooth after you got used to their planets.

But then they added Astra as the "Bigger," which was mostly still self-coherent:

> Luna (Small), Terra (Medium), Sol (Big), Astra (Bigger)

But then things go confusing, they released Sol-6.1 which sometimes is superior to Astra, and discontinued Terra which made it look like this:

> Luna (Small), Sol-6.1 == Astra

But with Sol-6.1 costing the old Terra-6.0 price, while, as I said performing as good as the old "Bigger" model in many non-3D/UI situations.

Now they're re-introduced Terra, as a 6.1 release, making this even more confusing.

Did Microsoft's chief branding officer move over to OpenAI?

Finance. Everything is a number to be optimized that must constantly grow. The US abandoned antitrust, worker protections, social safety nets, and more in the name of greed.

For workers to care you need to treat them as humans. People care about a team and a family that care about them back.

The mythical story of the person who started in the mailroom and worked their way up is gone. The mailroom, secretary, custodian, and as many roles as possible have been replaced with contractors without benefits or any future in the building they currently work in who will facelessly cycle out in a few months for another contractor.

Employees have been abused too much to believe the lies. If I work twice as hard, my reward is a sub-inflation raise and no realistic hope of promotion. Feedback employees give is ignored, so there's no motivation to push to improve things. The management class has a desperate dream of all people being fungible so they can throw them out and get cheaper ones, and fungible is what you get - if you want a generic interchangeable cog, that's what you get, but don't complain about how the cog doesn't "care enough".

I feel like the affluent just love to invent new words to show the “others” that aren’t using it how backwards they are, and how the affluent are truly enlightened.

Someone reported him for not warning the room about suicidal ideation before saying it and not considering that it might be triggering. He was actually reprimanded.

Unfortunately typical of the place. I left soon after.

I remember reading about some similar thing with French and the French president’s wife responding that the language was beautiful the way it was and they don’t need outsiders telling them how to gender (or not) their words.

I’m sorry that you find it exclusionary. What I find exclusionary is people telling me that it’s justifiable to think less of me for using that term, or comparing continuing to use the term to their grandmother who still uses the N word.

For what it’s worth I have stopped using “you guys” at work except for in “safe spaces” because I don’t want to deal with this sort of judgement. But I find the “it costs you nothing!” defense of word policing to be completely unconvincing.

She was told to not use "Latino"; but to use "LatinX" instead. What hurt more was the implication that she was being insensitive.

When I learned of it, I asked my Mexican friends which one they preferred. None of them had heard of "LatinX" and were as confused as I was.

As a woman in this industry, being told that using “guys” to address mixed groups is non-inclusive/insufficiently considerate/worth thinking less of me for (just wow) is much worse than being addressed by “guys” when I’m in a mixed group.

Like… we have real problems with equality in tech, right?

and then

> I was encouraged to use "y'all" [..] I thought that was a bridge too far.

I wonder if there might be a small bias on your part. The criticism you deem constructive everyone shall endure, but the other criticism you deem a bridge too far you yourself shall not endure.

I did used to work for a consulting firm about a decade ago, and I was leading a staff augmentation project with another company and I was once chastised for saying "hey guys" in a chat because there were women in the group. I was encouraged to use "y'all" but not wanting to sound like I was from the south, I just used "peeps" which was somehow okay.

I'm all for inclusive language but I thought that was a bridge too far.

You have the guy who wrote code in the 90's, became a dev manager, maybe is a PUM now, or still a dev manager, and basically "does emails and attends meetings", goes to the "daily triage" (which has a whole bunch of people in the room), is a member of the "war team" (they probably don't use that term any more). All these people are not "doing" anything useful and there are a lot of them at MS.

This used to work out fine, because the way the August reviews always worked (stacked rank etc.) it's very much a favorites game, "who you know". So you get a pass.

But suddenly there's a reorg and new management comes in (or a flattening; no more dev leads with 2 reports!). And they have their own set of dev leads, dev managers, etc. they want to install. So the existing management gets kicked out, those who still have buddies in other orgs will transfer there (but this gets much rarer now, as most friends from 'old Microsoft' have retired). But the "does emails & attends meetings" guy is out of a job (and is going to cry "age discrimination" because he's 50+, but the reality is he hasn't done a whole lot for a decade, and has zero useful skills now).

PMs; there are WAY too many of these, with their entire parallel org of management. There used to be almost as many PMs as devs back in the day. Other companies don't have anywhere near so many. Most of them can go; back in the day a lot of them would go to the dev's office, ask about a topic someone asked about in email, the dev says "Oh I can just reply to the thread", PM says "Nooo... let me do that" (PM wants to be the one replying to be 'visible' even though they are just a glorified status collector/forwarder).

There ARE a lot of incredible people there, and the best people I worked with at MS are just as good as the best I worked with at the FANGs. But unlike other companies (FANGs, startups) there are a LOT of a really bad/useless people there too. A lot. Although I have no faith that MS is going to lose THOSE people in this buyout.

Engineering discipline is absolutely non-existent at MS. Everything is date driven. The source tree (Source Depot) gets locked as of X date and only "test fixes" can be checked in after that. But the next ship vehicle is months away. So people keep checking in regular code as "test fixes" after the branch lock. One time it got so bad that the build was actually broken for the entire day of the branch lock, and people were force submitting in the evening (everyone expected whoever broke the build to fix it, but... they didn't, so force submit!) to make sure their changes got in before the lock.

Random unrelated funny MS story: The wiki that everyone in Bing used, got wiped out, because it was running on a machine in someone's office that got wiped when they left. We had to have everyone run a script to salvage what they could off everyone in the org's browser cache.

I also think there's been a drastic shift in the working culture of tech/engineering spaces, in general, over the last 15 years. From my experience, it's definitely drifted away from a bunch of nerds solving problems to something more towards emotional/personal/politics. I recently saw someone cry when a person very kindly asked them to explain a discrepancy. "Back in my day" people would just state facts, point out flaws, and argue a bit, and have fun doing it, for better or worse.

So there is some truth in the claims of the MS President.

https://www.linkedin.com/feed/update/urn:li:activity:7518428...

And that is going to have predictable consequences like increases in suicides, substance abuse, etc.

A job might "just be a job" but I'm growing tired of this trend online of treating work like it isn't important for most people's sense of worth, place in society, and overall mental health.

I’ve heard this claim many times and is not backed by any more evidence than “it works on my machine”.

If the company really did continue to operate in the same manner, it wouldn’t have been worth 80% less in 2024 (https://www.cnn.com/2024/10/02/business/elon-musk-twitter-x-...), nor would it need to be rolled up and financially engineered into being part of an AI company that then gets rolled into a rocket company.

It has been losing users, safety standards have dropped and overall the platform is worse off than before. There’s a reason why no other company has done something similar with their employee counts.

There is a big HOWEVER. At least 2 in fact.

1. The company had been growing in the previous years, 2019, 2018, 2017 (they shrank somewhat in 2015 or 2016). In 2014 they had 29% workforce increase. So the significant growth in 2021/22 is not an aberration but rather continuation of the growth in previous years.

2. What IS an aberration is the workforce count after 2022. No growth in 2023 or 2025 (as in ZERO YoY change), and measly +7,000/3% growth in 2024. And some YoY loss in 2026 (-5,000, or 2.2%). With 2027 YoY likely showing at least some (and possibly a lot) of these 60,000.

So no, "it's just overhiring due to pandemic" was supposed to die years ago.

Headline is making me laugh cause it sounds like a press announcement for a new product

"The new Microsoft Voluntary Exit Offer, thinner, and lighter, with 20% slimmer bezels than before"

The buyouts were taken by superstars and low performers. The superstars moved on to other things, like working in other forms of media.

Dad didn't take several rounds of buyouts and was fine.

After suffering a stroke, he was "put out to pasture", which for a journalist means he was sent to cover city hall meetings in some suburb. Eventually, he retired with some much smaller payout than what the last buyout would have given him. The sense in the family was that the collapse of the industry denied him a dignified last decade of work. (When the industry was flush with cash, older journalists were sort of like "elder statesmen" in the newsroom. Post-internet, no longer.)

I saw it happening in a funny way not long time ago - company offered buyouts, but got insanely surprised when high percentage of people actually accepted it. Management thought acceptance rate would be much, much lower, but when numbers started to pour in on a 2nd day, oh man... Panic spread very quickly because people were accepting packages in fear of layoffs without such package, so management said if any layoffs will happen in 3 years period after this, same package will be offered no question asked. Cooled things a bit, but what a mess...

- Early CS researchers like Donald Knuth entered CS in the days when regularly someone would come up with a surprising faster algorithm for some problem, and their intuitions were that most problems had faster algorithms than you'd expect.

- Later generations of CS researchers, as all the easy algorithms were discovered, and post Karp's definition of NP-completeness (and the Garey & Johnson book, say), began to specialize in proving hardness, and complexity theory with a “zoo” of complexity classes became more about separation and lower bounds, e.g. I've seen algorithms like Arora's PTAS for Euclidean TSP being described as a “failed attempt to prove a lower bound”. So an entire generation of researchers' intuition was mostly in the direction of an elegant universe with neat separation results (even if we couldn't get them).

- So e.g. surveys of computer scientists showed most of them believing that P≠NP, and someone like Knuth has always seemed a bit of an outlier for saying that maybe P=NP: his argument was simply that there are so many possible polynomial-time algorithms that we haven't explored, and maybe one of them with a large exponent could solve NP-complete problems. (In response to arguments that many researchers have failed to find any such algorithms, he would say that many researchers have failed to prove that P≠NP.) E.g. in 2017 when Scott Aaronson wrote his survey (https://www.scottaaronson.com/papers/pnp.pdf = https://www.scottaaronson.com/papers/pnp-kindle.pdf) he wrote:

> Admittedly, it can be hard to tell whether declarations that P = NP are meant seriously, or are merely attempts to be contrarian. However, we can surely agree with Knuth and Lipton that we’re far from understanding the limits of efficient computation, and that there are further surprises in store.

- What we've seen in this month is surprising unexpected algorithms, like o(n log n) algorithms for FFT and integer multiplication, and now this sub–exponential-time algorithm for 3-SAT. (Yes the ETH has had less support than P≠NP but I still gather it's unexpected.)

So (though I don't expect intuitions on P≠NP to change, and probably that's still the right intuition… but I'm just an interested amateur), maybe there will be a bit of a swing of the pendulum back to the idea that maybe clever algorithms exist that we just haven't thought of.

This precisely coincides with the web developers being allowed to wrap stuff into "apps" in Electron and the like.

SpaceX was the only one mentioned as cancelable. These are listed under "Infrastructure Service Arrangements" on page 170.

There are moonshot projects related to superconductors, materials, and biology that these companies can claim progress on to keep valuations stratospheric.

The value of money, and the ability for average people to have influence on future events, ultimately is tied to the fact that the labor of an average person has inherent value.

Once that is no longer true, all bets are off: All promises of compensation become meaningless, and it's easy for those in power to rationalize to themselves why they no longer are beholden to their earlier promises. They will say to themselves "The situation ended up more nuanced than we expected"

(Which is exactly what Google said in February 2025 when they reneged on their pledge not to use their AI for military applications)

Sun made bank 25 years ago because their servers was needed for websites, today we run larger workloads on 5$ VPS servers (well maybe the VPS servers has gotten more expensive due to HW shortages but they're not intrinsically expensive anymore).

If you're going to make a point, make your point - don't just throw numbers around without context.

By the way who are they? They claim tracking provenance of training data and poisoning detection, which everybody does. Why should I trust them and not Alibaba? Because they're marketing themselves to be "sovereign" to Canada? "Globally sovereign" is an oxymoron. (that was a rhetorical question, I know who they are of course)

> Ripple effects from the war in Iran and the closure of the Strait of Hormuz continue to widen. There's yet another brewing shortage, this time in helium. While most people associate helium with balloons and funny voices, the element is used in a surprisingly wide variety of industrial settings, including semiconductor production, where its role in advanced lithography has been growing rapidly. But helium mining and exploration in North America has been practically non-existent for a variety of reasons. And while the US used to have a strategic helium reserve, the government started selling that down in the late 1990s. On this episode, we speak with Nicholas Snyder, the founder and CEO of North American Helium, which does helium mining in Canada. We discuss the properties of helium that make it so useful, as well as the difficulties of expanding global production and distribution.

* https://www.youtube.com/watch?v=bjc6MgUY0BE

* https://omny.fm/shows/odd-lots/now-theres-a-helium-shortage-...

* https://podcasts.apple.com/us/podcast/now-theres-a-helium-sh...

The US had a strategic reserve, but it was sold off:

* https://en.wikipedia.org/wiki/National_Helium_Reserve