To use RLHF you need a dataset that includes instructions with good & bad answers - do many of those exist? I know there are a few datasets of just plain instructions-with-responses, but I'm not aware of any that have both good and bad (or ranked) responses. Is that trivial, or an important missing element here?
All of the UX interface have little up/down thumb icons... that's where the boolean feedback comes from. If people stop using that, sentiment analysis on the human responses will likely go a long way.
If I understood correctly, the OpenAssistant team wants to open-source their community built RLHF dataset.
On the other hand, if you're being cheeky, I bet there's a way to datamine from websites like ShareGPT and profit off shared ChatGPT <> User interactions.
I agree - LangChain is totally awesome and unlocks amazing new capabilities of LLMs (easily!), but the code itself is pretty awkward. No logging framework, for instance, and the strong typing enforcement with pydantic seems un-pythonic and brittle to me (what if I want to use a different kind of Mappable instead of a dict for something?).
Still, great work overall and congrats to the team!
Whatever time pydantic is supposed to have saved me from making dumb mistakes, I've wasted 100x more trying to shoehorn stuff into some poorly thought-out schema.
> what if I want to use a different kind of Mappable instead of a dict for something?
You can use custom field types either by using a `validator` with `pre=True` or a define a class with a `__get_validators__` classmethod.
But pydantic does have problems, imho it isn't strict enough & has given me the wrong types when using `Union`s a bunch of times. Defining custom encoding & decoding behavior is harder than it needs to be. The purely narrative documentation is easy to learn but difficult to reference.
I would agree that strict typing is unpythonic but in this case I think this is an outdated opinion of Python, one that it's been backpedaling on with type annotations &c. I think Python made perfectly reasonable decisions about this some 30 years ago, and they haven't aged well, which isn't even really a criticism as much as a consequence of it's enormous success. Python had outlived a lot of the ideas & attitudes about language design that went into making it, and carries them as scar tissue.
Somewhat OT, but I feel like this is such an underappreciated aspect of recent LLMs: Not just their ability to generate text - but their apparent effectiveness in making use of arbitrary tools to interact with their environment to achieve some goal. It seems like we're just at the beginning of this with the MRKL and ReAct papers, there will be a ton more awesomeness coming in this area I'm sure.
Yes I think if this works it's one of the strongest signals yet that LLMs as they currently exist have a fairly general form of intelligence.
I am pretty sympathetic to the field of "AI Safety" and I worry a lot about the implications of agent-like general intelligences. This post gives me a lot to ponder.
What are the implications of the fact that even AIs that are not agent-like at heart can apparently be told "please simulate an agent-like AI"? I really don't know. Should we consider it as an "inner AI" with its own objectives? How can we determine what those objectives are?
Instinctively it feels much less concerning than an AI with a direct action->perception feedback loop but who knows. AI is fucking weird.
I'm still a little unclear on the benefits that fusion offers compared to things like wind and solar. I understand that we need to develop better storage technologies for the energy produced by wind and solar, but that seems so much easier than the challenges currently facing fusion. Wind and solar just seem so far ahead of fusion already - they're pretty cheap and very widely deployed on a global scale. In comparison fusion seems very expensive and unproven and even when we get everything to work it might not be much better than a solar farm with a big battery pack. But maybe I'm missing something important about the economics?
It's not controlled. It does help boost some fission weapons. But it's not the hard part or critical piece of producing a nuclear weapon, and you can get by without it.
To illustrate how little it's controlled-- I have a little bit on my keychain as an alpha source with a phosphor so my keyring always glows.
Depending on the scale and reactor design, we have really good examples of run away fusion reactions. Run away reactions are easy, controlled ones are hard.
And whilst I won’t doubt that if fusion ever becomes commercially viable the reactors would be walk away safe it doesn’t mean that you don’t need to account for that in your design.
That is a run away fission reaction that ignites a short lived fusion reaction. We don't even talk about neutron populations or k factors in fusion because there is no avalanche effect possible.
Wind and solar have a max theoretical output that is constrained by physical space and competition for its use, in addition to weather patterns, etc.
Fusion energy has a theoretical max that’s orders of magnitude higher.
Wind+solar is the path to decarbonization and sustaining our current world.
Fusion is the path to post scarcity. If/when we get scalable commercial fusion, it’ll be like the transition to oil - society will radically change, in ways we can’t predict.
> Fusion is the path to post scarcity. If/when we get scalable commercial fusion, it’ll be like the transition to oil - society will radically change, in ways we can’t predict.
Except this is also true for fission. So if fission has failed to transform society, why do you think fusion will?
Uh because building fission reactors, despite being fairly safe, is still a risk compared to fusion? We don’t want to put fission reactors in every town, but we could one day with commercial fusion. And the sheer amount of energy we could harness would allow us to do insane things.
> despite being fairly safe, is still a risk compared to fusion
I recommend you read up on the Gen IV reactor designs. They’re totally safe - meltdowns are impossible because of the way the reactor is built. If anything catastrophic happens, the reaction stops and can never get to a runaway reaction (physically impossible). Look up Gen 4 reactors. Those will be available before fusion even gets off the ground (and I’ll note that fusion has 0 reactors built so who knows what kind of safety issues actually come up when engineering theory hits the road).
Even Gen III reactors are fine to put up everywhere (20x margin over Gen II) and Gen III+ reactors continue with the theme of adding passive safety measures that would prevent accidents like Fukushima and Chernobyl. Critics who rate any possibility of accident as unacceptable will never be pleased but that’s not a reasonable position to take because nuclear energy isn’t built in a vacuum and global warming and existing coal power poses a significantly higher threat and renewables and batteries simply can’t scale no matter how hard we believe.
Fukushima and Chernobyl were Gen II designs which do have a cost advantage and EVEN WITH THOSE ACCIDENTS those designs are safer than existing coal and LNG power plants we are fine with having all over the place (nuclear is slightly safer than wind). Even Gen II designs built today are a fair bit safer than Chernobyl and Fukushima. Fukushima also ignored many and repeated safety warnings from internal and external reports although critics will generally point to this as a general criticism against all reactors (even though Fukushima still failed comparatively harmlessly all things considered).
Even with all of that, the death rate per kWH generated is drastically safer than coal and on par with wind and solar. Also construction costs tend to go down when the regulatory environment doesn’t inhibit building reactors due to political fears that aren’t grounded in the actual engineering.
I’ll also note that China is building many many nuclear reactors and Russia is also following suit. So from a competition/national security perspective, China and Russia both have access to significantly more clean energy and more energy independence than we do.
Look. I understand there are problems with fission reactors. They remain the only feasible way to generate nuclear power in the next 60-100 years at scale. Yes there are downsides and risks. However there’s one big upside vs fusion: it exists. It’s possible to build these plants now without physics and engineering breakthroughs we haven’t made yet. The advantages of fusion are safety, nuclear waste management, theoretical proliferation concerns. There’s no reason to believe construction costs will be significantly lower. Even if they are, we’re not even close to the first real commercial power plant even with this achievement as impressive as it is from a progress perspective.
I answered why fusion would change the work in ways different to how fission changed the world.
> Look. I understand there are problems with fission reactors. They remain the only feasible way to generate nuclear power in the next 60-100 years at scale. Yes there are downsides and risks. However there’s one big upside vs fusion: it exists.
If you re-read the original comment, it supposed that fusion exists. You can't criticize something in development for not existing and use that as a point against why it won't be beneficial. That's circular reasoning.
> The advantages of fusion are safety, nuclear waste management, theoretical proliferation concerns.
Yeah, just nuclear waste management. No big deal.
> They remain the only feasible way to generate nuclear power in the next 60-100 years at scale.
You absolutely cannot predict with that level of certainty over 100 year time scales. You severely underestimate how much we can achieve over timescales as long as that.
Wind and solar only provide power during wind / during the day. Fusion can provide 24/7 power.
Battery packs can only store so much energy, and Lithium is a contested resource as most of the Lithium produced is required by the automotive industry these days, and the largest deposits are in regions where you maybe don't want to get your Lithium from (child labor, unsafe conditions, politically unstable countries, etc.)
But yeah, future energy will be a mix of available technologies, not a single technology alone. So you need e.g. fusion (or fission) for "baseline" power and wind/solar for peaks
Solar and wind have massive environmental impacts. Fusion's foot print is much smaller for the same output. Batteries are rather dangerous. Fusion is -- as far as I understand it -- much less likely to escape a reactor due to how difficult it is to sustain the reaction. Moreover, it's more dependable.
So in sum, the advantages are (1) dependability, (2) safety, and (3) small footprint.
My source is the fact that solar panels cause shade on the ground and squander energy that would normally be going towards developing biomass into developing energy instead. It just doesn't seem healthy for the animals and environment that live there. Especially with the talk of in ground installation, which basically destroys entire environments and soils and covers it with impermeable membranes. That's not great for soil health.
Even the great deserts of the southwest have life. In fact, I challenge you to drive through these tens of thousands of miles of landscapes in the hour or two after a rainstorm and tell me they're dead. You're missing out if you've not seen the desert in bloom.
I grew up by the desert, and I don't know why people think it's dead. There are some extremely fragile ecosystems there.
Apart from the safety improvements and environmental benefits, it's a way to produce a ton of energy. I believe it's about 4 times as much energy from fusion compared to fission with the same amount of fuel. I'm a fan of solar and wind, but it's going to be way easier to power the entire world sustainably if you've got fusion in the mix.
I think when comparing PV/wind to nuclear (fusion or fission) generation, we should include the cost of storage for renewables in the comparison.
Renewable generation + storage gives a system that's capable of meeting base load needs, just as nuclear generation does. Cost comparisons among base load-capable technologies is a better way to evaluate the economics, IMHO.
If we get an order of magnitude more energy, we can do an order of magnitude more things; fossil fuels gave us the Industrial Revolution, and nuclear fusion may unlock something similar.
> But maybe I'm missing something important about the economics?
I think you've understood it.
Imo fusion is never going to be able to compete with renewables+storage with the energy being captured from neutrons. Maybe reactions that release energy in charged particles or photons could, but they're even harder to do.
Could you elaborate on your point a bit more? If you're talking about utilizing the weak force vs. the residual strong force then I'm not sure this argument holds up.
Also, when comparing to renewable+storage you have to consider how much land has to be dedicated to energy use in these scenarios. Wind and solar require orders of magnitude more than a potential fusion reactor (or an existing fission reactor).
Just referring to what particles the released energy is carried in.
The easiest fusion reactions to make happen release most energy as neutrons. But neutrons are, from a practical standpoint, a huge pain in the ass to deal with. They just fly off until they hit another atomic nucleus.
They irradiate the structure of reactor, making it radioactive and weakening it, neccesating periodic replacement. This means handling radioactive materials, which as the existing nuclear power industry demonstrates, is hard to make cheap.
Reactions that release excess energy as charged particles, though all harder to actually do, leave you with charged particles that can be directed by electric or magnetic fields and can be used for direct enerergy conversion.
Yes solar requires a lot of surface area, but fusion power is just not looking like it will be anywhere near cheap enough for the real estate savings to matter.
Neutrons aren't that hard to capture. They are certainly harder to capture than charged particles but there are plenty of materials that are dense enough to reliably capture neutrons. This is how heat is extracted from the reaction to use in a generator. The activation of the containment material is a problem but it's not even close to the level it is for fission reactors where you're forced to deal with spent fuel rods.
At the moment fusion is obviously not cheap but no one is planning on using the technology in its current form for actual power generation. The processes involved will all get more efficient and given the astronomical upper limits of energy output from fusion it doesn't take a big stretch of the imagination to think that it will eventually be preferable to solar and wind power. There's no guarantee that will happen but hopefully this breakthrough will trigger more investment and momentum to make it a reality. I also want to add that I'm very pro solar and wind, especially in the short term.
Fusion brings the power of the stars directly to us, without it capturing the energy millions of miles later.
It unlocks a Star Trek, post-scarcity future that PV and wind cannot bring due to their space requirements.
Also, you could eventually put one on a spaceship or other planet. For that Star Trek future.
High power density. Start and stop on demand. Abundant fuel is another advantage, but in our neighborhood sunlight is also abundant. Fission also has good power density, but not so good on the start/stop flexibility.
Industry does not run on solar and wind and sad to say it, current storage energy is not green.
The cleanest energy available now is nuclear fission, but there is no money in it for the energy industry. It is too plentiful and cheap if implemented properly and capitalism does not like plentiful and cheap.
France has had cheap electricity for decades and it seems it has been so cheap that they don't want it anymore.
Most of the discussion here seems to assume bioinformatics / genomics jobs are academic, but I work for a clinical testing lab where production-quality code is a must. We're probably a 10/12 on the Joel test.
Just to clarify, exons are the portions of genes that code for protein sequence (they are expressed). Introns span the distance between exons and may contain regulatory or splicing information. Areas between genes are referred to as intergenic, and also may contain regulatory sequences that affect how genes are expressed.
Part of the issue with centromeric or telomeric sequence is that not only is it hard to sequence (being super repetitive), little is known about what a sequence variant in such an area might mean. It's kind of a chicken-and-egg issue: it doesn't get much attention because no one knows how to interpret variants there, and since no one knows how to interpret variants there it doesn't get much attention
It continues to annoy me that fusion companies make claims like "enabling a future with unlimited clean electricity.". Current, broadly used technologies such as wind and solar already deliver unlimited clean energy. What fusion promises is clean energy that is delivered more consistently - at night, or with no wind. But there are multiple paths to consistency, and improving energy storage technologies and improving the grid seem easier to do than inventing an entirely new and extremely expensive new class of energy production.
I'm honestly not sure why there seems to be so much interest in fusion these days. Wind and solar seem to offer a limitless, carbon-free energy supply with relatively cheap, well understood technology that is already price competitive with coal and gas. By contrast fusion seems super expensive and technologically very complex - even fission plants take 10+ years to bring on line. Does fusion offer some advantage over wind + solar that I'm missing?
My armchair/spectator interest in fusion comes from the idea of significantly increasing the energy available per capita, without increasing carbon footprint. Traditional renewables might meet our current energy needs eventually. Fusion could let everyone on the planet enjoy the same, or even a significantly increased, lifestyle. As the saying used to be long before I was born, "energy too cheap to meter" (though obviously it still would be metered).
With enough surplus energy, you could run entire reactors just for carbon sequestration, or nation-scale desalination, or climate engineering, or what have you.
A joke some friends of mine who study fusion make to this kind of comment; wind and solar are fusion powered.
In a simpler way of looking at it, (1) what is the source of the energy of solar/wind? (2) how much land and materials are required to linearly increase power production?
There is a finite amount of space and resources on the planet to continue to scale power production with humanities consumption.
Fusion, preferably MCF/tokamaks in the style of smaller sized ones like SPARC @ MIT and less like ITER (behemoths that take decades to build and maintain) offer two things (1) the fuel is comprised of the most common elements in the universe, (2) power per square foot is much greater than solar or wind... And bonus (3) once developed it should in theory require less material per watt generated. And less materials mean less processing and fabrication which in turn reduces the environmental impact on the planet.
Excuse my lack of knowledge on the subject, but from a few comments/articles it seemed that although Tritium can be recuperated through contact of the charged neutrons with lithium "blankets" surrounding the enclosure, it is still incredibly hard to achieve self-sustainability, to the point where we might need fission reactors just to produce that isotope. Am I missing something or isn't Tritium not so readily available?
The margins for 'breeding' tritium from lithium are slim, yes. I don't think fission reactors can help in the long term, but right now that is how we get tritium for our experiments.
How about advancing the frontiers of science? That alone is worth it, it is not?
You seem to want to halt all further research into alternative energy and settle with the current state of our solar/wind capabilities, which is strange.
Fusion is a poor kind of science. It's very inward focused, answering questions that are relevant to fusion but not much else. As pure science, it would not merit the $$$ being focused on it.
As stronglikedan said, spaceflight propulsion/institutional inertia. (Stationary space facilities will use solar, just like on Earth. Space transport, unlike Earth transport, will be an awful combo of slow and expensive. Space stations will need to be simple. One type of computer. One type of microcontroller board. Maybe three sizes of screw. Solar panels are simple, identical, and interchangeable. And not radioactive! (Fun fact: every bolt on the outside of the ISS uses the exact same head size: 7/16" hex))
Seasonal variation with solar is a bit of a bummer. If we need to fully electrify everything, (Transport and heating) then winter will be a problem. Either we massively overprovision solar in order to still have heat on the shortest day of the year, or we run thousand mile cross-country transmission lines and enormous battery banks.
Even so, the economics are such that heavy industry might become a seasonal job. Right now we run aluminum smelters 24/7 because baseload power is fairly consistent, but if solar power is free in July but dear in January you might see multi-month shutdowns. This gives headaches to central planners, and makes them inclined to pour billions into fusion if it can preserve some of the status quo.
Something of a question mark of how long the tropics are going to remain habitable. We're still on the "business as usual" emissions curve, with no signs of meaningfully changing that. That puts us 5C higher by the century, and will keep going up after that.
Many equatorial regions will either be too hot for human life, (https://advances.sciencemag.org/content/6/19/eaaw1838) or be active combat zones as a result of refugees escaping heat. Some cities in India are now routinely hitting 50C during the summer, and we've only had 1C of warming.
If the tropics threaten to become uninhabitable countries like India will go ahead with direct climate engineering by dispersing fine particles in the stratosphere to reflect sunlight. This need not be at all expensive, even for them. No credible threat from other countries would deter them, as they'd face mass destruction otherwise.
I think wind and solar are plan A at this point, if for no other reason than they're hard to beat on cost. Battery storage is still expensive, though.
If fusion works out, it could be used to make up the difference when the sun isn't shining and the wind isn't blowing (though if we eventually get high-capacity transcontinental HVDC lines to buy and sell power from practically anywhere or batteries become really cheap, that becomes less of a concern).
Fusion would also would require far less land, and some people object to having a landscape covered in windmills and solar panels.
Fusion might be useful in places where renewables are less practical, like on ships. Naval vessels might conceivably replace fission reactors with fusion. If they're safe and relatively simple to run, you might even see them on civilian ships. Or you could have fusion reactors in remote places, like floating on a buoy in the middle of the ocean, to serve as a charging station for battery-powered ships.
Fusion may be useful for establishing a human foothold outside of Earth. For instance, methane production on Mars (for rocket fuel) will require enormous amounts of energy, which could be supplied by a fusion reactor. (A fission reactor would perhaps work just as well, but there are legitimate reasons why people get nervous about launching hazardous materials into space on a rocket that might blow up before it achieves escape velocity.)
We might also begin engaging in projects that require enormous amounts of energy. For instance, if certain CO2 absorption strategies are energy-intensive, and we can't practically generate that amount of energy from renewables.
At this point, we really don't know if it'll work much less what the practical limitations will be, so perhaps the best we can do is say "if a fusion reactor can produce X amount of energy and weighs Y tons and requires such-and-such amount of cooling and requires an overhaul once every N months at a cost of D dollars, we might want to use it in these applications".
Hydrogen made from renewables and burning in combined cycle turbines would likely be far cheaper than fusion.
Fusion would be horrible for ships. Ships are volume constrained, and fusion reactors are very large.
Land constraints are not globally significant at current energy demand. The world is constantly hit by 100,000 TW of sunlight; average global primary energy demand is about 18 TW.
In space, DT fusion reactors will be inferior to fission reactors, which will be much smaller and lighter for a given power output (and also much simpler).
It's very difficult now to make a case for fusion. In the past, the case was something like "fission will be a big winner, but then we'll have trouble with uranium availability and safety and waste, and fusion, while slightly more expensive than fission, will still be cheap and solve these problems." But that's not how it turned out -- fission failed because it was too expensive, and fusion being even more expensive than fission makes it a nonstarter.
Fusion doesn't have to be big. ITER is huge because it was the smallest it could possibly be given the superconducting magnetic coils that were available at the time it was being designed, but we have much better high temperature superconductors now. (This is the basis of MIT's SPARC and ARC projects.)
Currently, we don't have any practical working fusion reactors, so it's hard to say what the attributes of such a reactor would be. We have some designs that according to our understanding of physics might work, but the designs are likely to go through many iterations before we have something that can be mass-produced and deployed in volume. Rebco tape probably isn't the best high-temperature superconductor that will ever be discovered. And so on.
It does, actually, with neutron producing fuels. The problem is that volumetric power density is limited by the areal power density limit on the wall of the reactor, and by the need of a sufficiently thick blanket to absorb neutrons. The inferiority vs. fission is roughly (thickness of fusion reactor blanket)/(diameter of fission reactor fuel rod). This is independent of any details of plasma confinement.
Something like ARC has much higher power density than ITER, but it's still very inferior to fission reactor. ITER's power density is just so incredibly bad.
Does it matter? Fusion doesn't have the same power per unit volume as fission in order to be usable on a ship, it just has to be good enough to be usable in that application: i.e. able to produce maybe in the neighborhood of a couple hundred kilowatts continuously without being overly bulky or expensive.
There might be limits though on how small the reactor can be made. ARC is apparently meant to produce hundreds of megawatts, which sounds like it maybe be two or three orders of magnitude more powerful than what even a large container ship would use for propulsion. SPARC is a physically smaller reactor, but not intended for continuous or long-term use. If the basic design works out, probably the first real-world designs will be optimized for utility power generation, where size doesn't really matter except to the extent that "bigger" tends to mean "more expensive". Minimum-size designs might take longer to show up.
If I understand correctly, the ARC design uses FLiBe to capture the energy from the neutrons. It takes up the space between the vacuum chamber and the outer housing. The FliBe heats up, and is pumped out into heat exchangers that produce steam to run a turbine. At some point there's a practical limit to the amount of heat that can be removed that way, but it seems like a low-output reactor should be easier rather than harder to make from that standpoint.
"Hundreds of megawatts" may be oversized for ship application, but maybe more like 2x to 10x oversized rather than 100x to 1000x. Or maybe not. Apparently they use some pretty powerful reactors in aircraft carriers: https://en.wikipedia.org/wiki/A4W_reactor
It's a direct rebuttal to the false statement "Also for the same reason deployment would be faster allowing a faster phase out of fossil fuels."
Fossil fuels had better be phased out soon, and fusion cannot be available in that time.
IMO, the prospects of fusion reaching a practical state are so remote that even the current level of funding on it is difficult to justify. There are fundamental engineering constraints that render it inferior to fission -- and fission is now going extinct itself, being too expensive.
Fusion (and fission, if we are willing) would be the only way to actually reverse climate change by using CO2 capture and then electrolysis to put carbon back in the ground. Solar and wind might be able to just sustain human consumption for energy but only fusion would give us enough energy to take CO2 out of the atmosphere.
It doesn't scale or do 24/7 power like fission/fusion would. Not sure why everyone is still against nuclear but that's fine too. I certainly have nothing against solar or wind and think we need a mix.
And, honestly, my life would be way better if I could spend twice the energy I do now, without destroying the planet. Energy use is basically a synonym for quality of life for humans. If we can expand it without destroying the world, we should.
In response to your final question: To my ear using "men" as a substitute for "all people" doesn't flow better, I found it rather jarring and antiquated and, yeah, a little insensitive. I'm also not convinced it's a purely stylistic decision, as you claim... There's a rather long history of using men as a substitute for everyone in, say, medicine, politics, finance, and many (most?) disciplines, with outcomes most of us agree were not so great. It can also muddy the interpretation of your arguments, as (asterisks aside) it's not perfectly clear when you say "men" if you're referring to everyone or just actually just males. It's an easy thing to change, and your rhetoric will be clearer, more persuasive, and less likely to be immediately dismissed by a fraction of your audience.
I thank you for your take; yet I refuse to yield to your sensibilities.
I think where we differ is in our approach to language. I see it as a medium of art -- like a song or a painting. "Man" is but a certain, evocative hue of brown that I believe fits best into the feelings I'm trying to elicit; and "human" is a lesser, albeit passable substitute.
I think this scientificization, making it more rigorous and "comprehensible," has done the opposite. Words have connotations, denotations, and all sorts of deeper meanings behind them. "Human" is such a sterilized, unnevocative word; and I refuse to use it.
However, certainly you've felt something from its usage -- even if that feeling was not the one I felt (compare it to: human -- which only the most scientifically-obsessed would have their hearts sing from its utterance).
"Human" is a perfectly evocative word that does make people's "hearts sing". You have the "human touch", humanitarianism, treating things humanely. It reflects our highest aspirations for ourselves, and your rejection of it reflects upon you.
"Man" connotes... well, men. It depicts a society where the real movers and shakers are men. No one really hears "man" and thinks "men and women" (or "women and men" - why should men come first?). Notably, they say they do, but this has been shown by experiment to be inaccurate. If you say "man", people can't help but picture one. Indeed I also notice you use gendered language everywhere, not just in the word "man" to mean humanity. You say "men of letters", for instance, and you take care to match the pronouns. Am I really meant to imagine women when you write that? Would "people" not do just as well?
Lastly - far from being a "fad", this debate was active in the 70s, and has been entirely resolved now; nobody uses "man" this way anymore, and it sticks out like a sore thumb when you do. It sounds archaic to the point of comical. Consider if the subtextual message you want to send with your word choices is "I stubbornly refuse to adapt to the times".
Human as its own word, not as a derivative for others---and in its own, separate, unmodified context---has a specific, insipid connotation, compared to humanitarian, "human touch," etc.
Frankly, my experience has been that men are the real moves and shakers (and women impelling them to move, shake, and writhe around). It takes a certain amount of ego, and internal and emotional drive to "shake" and "move" (which I must assume you consider to be "good" aspects of humanity) the world -- one that most women do not have, and the ones that do require external resources to maintain that "drive" (mostly food).
I don't understand why you would rally around for this point. Being someone of substance is a worthless affair, compared to being someone of culture. The first comes easily to men, and arduously for women. The last comes easily to women, and arduously for men. It's an atrocity to eschew woman's gifts, to pursue men's. That's how you destroy a nation's culture, tradition, its children, and thereby its society. There must be a duality, with a strict boundary, else you get muddied people that don't know a single thing about what it means to be human. (in this context, it fit just right)
I use man in this way. I don't care what anyone---justifying their schooling and existence---has to say about it. Progress without purpose is wasted breath. Conforming for its own sake is death. Anything Western Intellectuals have written in the past 70 years I believe to be without merit.
...
Yet, I'll admit Hofstadter got a chuckle out of me with that riddle: I thought the surgeon was the male spouse of the deceased father! I yield.
I agree with your take, after it's been shoved in my face, and forced me to self-reflect. However, I'll still use "man" as a more archaic synonym of "human," albeit with a more apt note this time.
You got me. Take your upvote, and let me live in my crotchety, curmudgeon fantasy-land.
