What are the big/important problems to work on?

If we lose the war on general-purpose computing, authoritarianism and corporatism are likely to rule the world.

Winning that war is prerequisite to everything else.

If you're worried about climate change, remember that authoritarian regimes don't care about climate change.

If you're worried about injustices, remember that authoritarian regimes don't care about them and actually manufacture them.

If you're worried about any particular political thing, remember that authoritarian regimes don't even let you speak your dissent.

If you're worried about economics and making a living, remember that authoritarian regimes don't care if you have enough food.

Authoritarian regimes only care about themselves and their power. They'll make sure they're insulated from climate change. They'll make sure the law doesn't apply to them. They'll make sure they have megaphones for their propaganda. They'll make sure they always have food and shelter.

So what can you do?

Write Free Software. Get people to use Firefox and Ladybird (once it's ready). Refuse to work for companies that are locking down computers. Shift the culture until it is shameful to even do so. Reject remote attestation. Accept inconvenience in the software you use. Preach to everyday people about using Free Software. Teach them about privacy and control issues. Help them install Linux on their laptops. Root their phones for them. Be their tech support when things go wrong.

Most of all, learn UX and make your Free Software more usable and convenient than the freedom-snuffing software. Until Free Software is more convenient, there is very little way we can win this war.

This is something Free Software enthusiasts are really blind about: we can't just make software how nerds want it; we need to make software that will be useful and convenient for everyday people.

The world's hardest problems are generally uninteresting because they aren't problems you personally can do anything about without years of training.

"What is the highest-impact problem you are capable of solving?"

^^ I think this question is more powerful and much more difficult.

Learning your limits is incredibly hard. You have to border between optimism and pessimism -- between ego and humility.

After you understand your capabilities, follow your curiosity to find great problems. I think the discovery of good problems is itself the important trait. If you rely upon others to tell you which problems are important, you will spin directionless without understanding the underlying principles.

I would take a different approach than many other problems here.

- Healing damaged relationships in your life.

- Addressing unresolved internal traumas and fully integrating them.

- Replacing addictive and avoidant behaviours and taking care of the problems in your life head on.

I don't have "the answer" to any of these, obviously, but they are certainly big and important.

Apologies for the USA centric views:

Tackling climate change mitigation and remediation.

Getting more people off social media.

Protecting libraries.

Making the police more professional. Raising the bar for police candidates. Stopping the racism and fascism embedded in some police department cultures.

Overhauling the US political system to work for the people rather than the rich.

Protecting the rights of women, including bodily autonomy.

Protecting the rights of queer people.

Protecting the rights of non white people.

Reducing plastics use.

Planting more trees.

Improving housing construction standards.

Giving everyone a permanent place to live.

Decoupling healthcare from employment.

Universal healthcare.

Increasing the number of doctors and nurses.

4 day working week.

Forcing healthcare professionals to work normal number of hours.

Teaching good parenting at school. And various other life skills.

Improving teacher conditions.

Preventing fascism from continuing to rise in America.

I have the feeling that we are looking at climate change wrong.

All the discussion is about reducing CO2 emissions. But we have nothing to show for it. Even the during the Covid lockdowns, CO2 emissions were reduced by only 10%:

https://ourworldindata.org/co2-emissions

CO2 stays in the atmosphere for hundreds of years. That means even if we permanently take the extreme measures of the Covid lockdowns, global warming would be in 11 years were it would be in 10 years if we did nothing.

Yes, we should continue to push towards reducing CO2 emissions. A carbon tax would probably be the most effective way to achieve that.

But that will only give us more time to prepare for the inevitable: Living on a planet with higher temperatures and all kinds of problems that arise from it. It will happen.

What technologies can we develop to cope with it?

1. Books or e-books, send books to countries with low literacy rates.

2. Get non-fossil fuel energy to poor countries. They are burning coal or without power due to cost/availability. They still need to go thru all the ages richer countries have gone thru.

3. Clean water. Feel free to research water wars and Wikipedia doesn’t list all of them.

4. Weather forecasting. It is not great especially for disasters. I think the goal is 2 weeks of accurate forecast globally.

5. Getting people together without social media. Like a church but not religious, it is good to have a network of people to speak to.

6. Wicked problems: https://wikipedia.org/wiki/Wicked_problem

7. Natural language processing. We are repeatedly doing work that has already been completed or solved but we do not know about it. Ex: underground flowering and underground fruiting palm “discovered” but known to locals for a very long time.

8. Communications. It seems to be the problem almost always. Too much data to digest, not enough, frequency of data, etc.

9. tried to get to 10 but just search: difficult problem, Unsolved problems in (math, physics, philosophy, etc)

Tearing down the political circus, big government and achieving a pocket of anarchism powered by free, unconstrained trade.

Green, sustainable, plentiful energy. Turning trash into products.

Achieve greater and more harmonious integration of technology and nature.

Fostering small decentralised groups, communities, instead of large institutions, companies, projects. Doing more research on how small groups of people behave, as opposed to what sociology has been doing for the past 75 years.

Making technology work for us. Dedicating 40 hours a week for 50 years to full time employment, for the privilege of hiding from the rain, is a level of slavery most people are happy to defend.

Declaring the current computing stack as bankrupt and unsustainable and explore novel ways of expanding human intellect through technology. The current code-compile-run model was already ancient in the 1970s. We have reached a local maxima, time to try something different.

--

These are the things I care about, and not enough time nor connections to make a significant dent to any of them, just yet. Happy to chat about any of this with other restless souls.

I feel like this can really vary by person. Some things may be more important to others.

I personally think seeking ways to improve transit and reduce car dependency is an important task. A lot of other commentors mention electric cars. These are conflicting goals but both solve similar problems and may be worthwhile to work on.

If you are interested in seeing what jobs are actually out there, I think "80,000 Hours" is a good place to start: https://80000hours.org/start-here/?int_campaign=2021-12--pri.... I don't think they are always perfect, but I think the idea is in the right place and their job board often has roles that are working on important problems for humanity.

Provding access to electricity to people in less-econmically developed nations.

South, East and West Africa still have massive issues with providing one of the most transformative technologies ever, to their populations, particularly in rural areas.

Average rural electrification rate across Africa is 29%, with places like Zambia at only 11%. https://www.usaid.gov/powerafrica/zambia

Access to electricity obviously has impact across many aspects of life, but for many of the people I have spoken with, their immediate priority is improving their kids education. If you don't have electricity you can't study after dark.

This particuarly impacts girls, who not only don't get the light to study after dark like the boys, they are also responsible for collecting firewood, which is used to get even a little (crappy) light.

So after a day at school, the girls go out to collect additional firewood to create light, then are obvisouly exahsted, don't have very good light to study and fall behind in school.

That's just at the bottom of the ladder. All the way up industry, they are all impacted by poor electircity supplies.

In my opinion, it's a transformational area to be working in.

In non-software domains:

0.De-acidification of ocean water and rapid fossil fuels phaseout in all industries, preferably before the marine foodchain collapse and de-oxygenation of oceans.

1.Methods for clearing microplastic, PFAS and heavy metals(both of which accumulate in microplastic).

2.developing a sustainable plastic/rubber alternatives without fossil fuels.

3.Development of better supercapacitors and decentralized/distributed electricity storage(as alternative to fossil fuels).

4.Development of non-chemical space launch methods, space factories and asteroid mining.

5.Development of cheap/effective desalinization methods.

6.Development of environmentally friendly re-usable building materials and replaceable building components/parts.

In software: Open-source decentralized alternatives for proprietary AI APIs/tools would have biggest long-term impact. Its probably the most important field in software engineering right now, having this power locked in proprietary corporate walled gardens like "Open"AI would also force mainstream AI use to depend on whims of centralized API providers.

Solving the trust and bad actor problems on the internet, without turning the internet into a set of restrictive walled gardens.

Personally I think it's partly solvable with a combination of cryptographically signed HTML elements and a web of trust similar to Raph Levien's https://en.m.wikipedia.org/wiki/Advogato or https://keybase.io/

But it's going to be a big piece of work.

Richard Hamming gave a very influential lecture called “You and your research” — there are a few versions of it online and it’s been discussed here a few times before.

He poses the same question in pretty much the same way, “why aren’t you working on the most important questions in your field?”

I would say that what he defines as “important” is along the lines of “what will be the most influential?” - the sorts of things that bring Nobel prizes and glory.

But no one can tell you what is important: it depends on your own values; it’s a question you have to ask yourself.

Here are my two:

1. Software supply chains. Our civilization is built on top of a layer of software that has a gaping security hole the size of a gas giant planet. It's also recursive - software dependencies themselves have further dependencies and so on. As we get into a world with autonomous trucks driving down highways at 60mph and cars that collect a complete profile of location data history of unsuspecting drivers and upload them for "analysis", this problem will explode in ways that very few people can even imagine much less predict.

2. Linux distro upgrades. An Ubuntu LTS release or an RHEL release will eventually come to end of life. At that time, performing an upgrade across a fleet of servers is a complete fucking nightmare. You test everything that you know to test, hit deploy and watch as your servers go down and reboot with a new software version and restart various services and containers and hope everything will be ok. This is not sustainable. A solution is needed and is worth $$$.

Not really answering the question, but…

Ignore those people - it’s perfectly OK to work on small problems, or even just to work to support a family or put food on the table. Little things add up to big things. Don’t beat yourself up about doing “just” earthly tasks. Enjoy the moment. Quit comparing yourself to a straw man aggregated from the best slivers of imperfect people. There’s a good chance any given person’s grand vision of what’s important turns out to be misguided anyway.

My thoughts:

* Some new kind of wholesome work that gives anybody with a mobile phone a source of income, enough to pay for shelter and food.

* Invent a way that an organisation's operations that cannot fail because the operations are desired by people and it is not dependent on funding or property.

* Abstract the problem of costs so that we can indeed have good things and they are sustainable and kept alive.

* Heal the world wide web.

"What does it mean to be human?"

We are about to see machines surpass human performance at every conceivable task.

I therefore consider the problem of understanding what humans are at the fundamental, existential level (besides second-class machines) to be the deepest and most important problem of today.

And I suspect it will be the only problem left for the humans of tomorrow.

Climate change.

Biggest unsolved problem I'm aware of is product discovery. Given an intention, I want to buy a thing, say a TV; finding an actual good product (according to whatever criteria) has gotten absurdly difficult due to a confluence of effects, including the closure of brick and mortar stores, the death of magazines, the rise of e-commerce market platforms, dropshipping, paid reviews, straight up scams, etc.

This might be helpful: https://80000hours.org/problem-profiles/

In 2015 the UN created 17 ‘Global Goals’ (https://www.globalgoals.org/) that are meant to be a "shared blueprint for peace and prosperity for people and the planet, now and into the future":

Goal 1: No poverty

Goal 2: Zero hunger (No hunger)

Goal 3: Good health and well-being

Goal 4: Quality education

Goal 5: Gender equality

Goal 6: Clean water and sanitation

Goal 7: Affordable and clean energy

Goal 8: Decent work and economic growth

Goal 9: Industry, Innovation and Infrastructure

Goal 10: Reduced inequality

Goal 11: Sustainable cities and communities

Goal 12: Responsible consumption and production

Goal 13: Climate action

Goal 14: Life below water

Goal 15: Life on land

Goal 16: Peace, justice and strong institutions

Goal 17: Partnership for the goals

Each one is broken down into subtasks and targets. Making a dent in any of them would be worthwhile - https://en.wikipedia.org/wiki/List_of_Sustainable_Developmen...

Local government SaaS.

There's so much business processing in local government whose cost could be cut by an order of magnitude if extracted into external software. In an inflationary world with stretched budgets, this could add much needed resources to public services.

For example, instead of each police force building their own video reporting system they could share an external one. Similar for booking GP appointments, managing jury duty, applying for council tax discounts and managing parking permits.

People might critique this comment by referring to SaaS solutions which already solve these problems. However, none of these are adopted at scale.

The biggest barrier to widespread adoption lies in the sales process. Marketing to governments is hard and the challenge of effectively addressing privacy/ethics/sovereignty concerns is bigger than with most private sector actors.

I don't think big and important are synonymous. One of the most important problems in many communities is the vanishing of so called 'third places'. The places people go that isn't their home or work. Many communities lack them them nowadays.

This is something everyone can work on, you don't need to be a climate scientist or solve global peace. If you're a programmer as many people here are you can work on buildings things for your community, your neighborhood, a civic organisation, and so on.

The alternative to building some soulless ad platform isn't some world changing thing, it's simply to build small scale things with good incentives for people around you. That is important, and if enough people do it, collectively it will have a big impact.

If you can solve human greed then all other problems will resolve

Rutger Bregman recently twote about a "Hogwarts for do-gooders" which is an incubator for charities/non-profits for problems that are important, neglected, and tractcable [0].

Because there isn't a profit-incentive, there are many gaps in the market with a lot of low-hanging fruit with high ROI for every charitable dollar.

[0] https://twitter.com/rcbregman/status/1688526388004077568

I think there’s a lot of value if it’s possible to make saas as low cost as software.

Software that I write and you run on your machine costs me nothing extra per user to run (assuming I provide zero user support). So I, and others, don’t mind sharing so much.

That’s not the same for saas as there’s a cost for each user.

It seems to me that there are more saas utilities that are needed and that’s resulting in less free software.

Figuring out some kind of super low cost grid/distributed compute that doesn’t get abused to death seems important to having more software shared and thus a wonderful future state for humanity.

So even basic stuff like a password manager service is costing $5/month/user and I don’t think they scales. I run hundreds of little utilities thatI accumulate over my life. I can’t afford to pay $5/month for all of them and so would like to find some way to solve this so when there are real non-zero costs per user it can be covered and scaled. (Obviously some companies and people don’t want to give their stuff away for free and wish to sell their service, more power to them, but there’s not really a way for me to offer something that needs saas functions without me standing up and running a server and that’s a very different task than just writing software, giving it away and buggering off).

The internet seems to be a really good tool for sliding some people into ideologies of hate.

If you could avoid or correct that course, like at an architectural and protocol level that'd fix a big problem of the internet.

I'd only accept that it's not possible if someone could somehow demonstrate a logical proof (such as, presume a system existed, then prove by contradiction that it's not possible, etc). That would be an achievement in itself.

Social media fracturing relationships.

This hurts on a personal level (families, friendships). More importantly, what is the effect on society when we can't listen to each other?

Our contempt for each other has increased civil strife and violence, and we simply can't tackle large problems. Think about COVID-19, where political issues have prevented us from even agreeing on the cause of this plague - to say nothing of preventing the next one. And now imagine coordinating to prevent possible AI catastrophe.

We don't need to all think alike. But we better repair civil discourse if we're going to address any other problems.

  - Sustainable *and* scalable agriculture
  - Energy sector transition to renewable

I quit finance and work on EV chargers. The energy sector seems like an interesting place to be as a software engineer. A lot of the software problems has to do with customer acquisition, user interaction, and pricing, but there are lots of real engineering problems, too.

Did really nobody answer "raising kids" yet? Then I'll do.

Top priority: ensure our continued existence as a species/planet against all existential threats.

Next most important: human aging and longevity.

Next most important: AI, energy

After that: we're into sci-fi very, very quickly. It's hard to imagine what will be important, we're going to be sprinting in every direction. A million things are incredibly important to support what we're already doing, too. These are just the most important frontiers. There's some other incredible problems falling by the wayside, like loss of biodiversity that is billions of years of code written by nature.

For most of us as individuals, I think the most ambitious work is to help solve human aging and death in our own lifetimes. We're probably the first generations who can call that possible, if it is possible.

https://www.youtube.com/watch?v=RwEKg5cjkKQ

I think an inevitable but mostly untapped direction of research is to leverage biotechnology to manufacture materials to replace plastics, metals, chemicals etc., that either don't have renewable sources or can't be biodegraded back into source material. We can go even further and create new biological machines, bioengineered motors and computers and all kinds of shit scaled up from existing models in nature.

The technology of nature and life is far more advanced than anything we have conjured up ourselves, and the more we learn from it and harness it, the more we will be able to advance and evolve our technological progress without creating all the geopolitical and environmental problems that usually come with these advancements.

As some others have mentioned in this thread, the most important problems are often not the most interesting.

For example, Code for America has several "civic tech" projects where volunteer software engineers can make apps to make it easier for recipients of government benefits to navigate the system. These kinds of "civic tech" apps make government more efficient. They have a direct and measurable positive impact on poverty and inequality.

But these projects are not particularly interesting from a technological perspective. They are not rocket science. They are just web apps. A lot of the time, solving important problems is not particularly complicated or difficult. You just have to care. In other words, you have to be able to prioritize your values over your ego. Will you be able to put on your resume that you learned a fancy new technology? Probably not. But will you be able to say that you had a positive impact? Probably yes.

If you are a software engineer looking to make a positive impact, you should search for terms like "civic tech" or "humane tech" to find projects that are driven by humane values, not profits. Or you should reach out to a nonprofit organization and ask if they need help with their website. They could often use help with making it easier to get donations, communicate their message, contact a politician, or do other stuff related to their website.

The other thing to remember is that solving the most important problems is not just thankless, most of the time. It also doesn't pay well, because it often involves serving people and communities who don't have the ability to pay much. But if you are one of those patient people who prioritizes your values over your ego and your finances, I respect you, and even though I don't know your name, you are my favorite kind of person.

I think the world would be a beautiful place if the most important problems were also the most financially lucrative to solve. But we don't live in that world. We can't expect the financial incentives to line up with what's important. If you serve a nonprofit that aims to make the world a better place, you probably can't do it for the money. It has to be because you care.

Looking at the responses here, I see a pattern: generality in the form of downstream consequences rather than specificity of upstream causes, e.g., “world hunger” rather than “lack of low-cost drought-resistant soil-regenerating edible plants” (I’m making that up — I have no idea why hunger still exists, and I’d assume it is mostly political).

“Make something people want” sounds simple, but it is really hard to identify specific gaps in people’s lives that are the primary, but many-step-removed, cause of the problems in people’s lives.

Maybe it is possible to go from general to specific by asking, “Why? What’s missing?” until you get to something very specific that is missing that is the root problem.

Scaling worker cooperatives. Giving tools/automation that makes it viable.

Building HR-alternative software for workers to organize with and provide for themselves the things that HR denies them (pay transparency, mutual aid, feedback accountability, worker rights training, etc)

I think you need to think of it from 2 perspectives. What if I only had 5 years to make an impact and what if I had 20. The UN has its 17 sustainable goals [1] which can give you a better idea of what challenges face us at a macro level.

Start with your skill set, what do you have a unique advantage in that others don't? Are you a biologist? Work on creating more nutrient-dense banana or something. Know how to code? Solve a simple problem that wastes tons of time at a large scale (help your DMV or help people navigate something cumbersome using tech).

[1] https://www.globalgoals.org/take-action/?id=1

I do fear that as climate change evolves and we get to face more severe consequences of that the democracies will turn to authoritarianism as there would no longer be a win-win game to play. Together with surveilance and deception tech this might cement and worsen conditions of the poor as their power to organise would diminish whereas they would be most affected by the worsening climate.

Thus one of the important problems in my opinion is a solution which could circumvent surveilance and deception tech and help people to organise more effectivelly. Personally I am interested in publically verifiable software independent remote voting systems.

Non-exploitative business models.

Wonderful question!

The most important problem at the societal level is: sustenance, stability, avoidance of complete ruin. And by extension, all the sub-processes that contribute to those goals. Education, economy, law, healthcare, training -- all these contribute to the resolution of social problems.

The most important problem at the individual level is: that of realising the nature of self, the mind, the system called "thought"; moving towards an understanding of what the actual position of the individual is in the greater universal order.

In most popular culture discourses, one rarely sees any focused activity towards these goals in meaningful ways.

Pushing IPv4 to end of life. If we can get even just half way there our approach to online technologies will change wildly. For example, we can do so much more with so much less without reliance upon a server between two people trying to message each other. That means superior capabilities and you get things like inherent privacy and end-to-end encryption for free (and with less effort).

I would call understanding what's the underlying mechanism behind LLMs and other deep models such a problem. I.e. what's the internal representation? How it emerges over training? What its failure modes.

There're two reasons:

- In order to substantially improve these models we need to understand what they do. Something similar happened with steam engines when thermodynamics helped to explain what they do and improve them substantially.

- We need to understand their failure modes in order to understand their safety profile. I.e. it's AI safety problem.

I recommend taking some time to consciously assess your own values. When I went through this exercise several years ago, I wrote down things like “family”, “helping others”, “pursuit of knowledge”, “health”.

Once you understand what is important to you, set out to find ways to align your actions with those values. For instance, if I value family and don’t think I’m spending enough time with my family, I should make a goal to spend a certain amount of time with my family. The “real problems” are whatever you decide they are.

The worlds is full of problems, no shortages there, the evolution engine is still churning.

however, the problems occur with what is meant by "big" [more quantitative, utilizing science + engineering] or "important" [more qualitative, utilizing humanities].

without a common framework of tools, standards and ways it's hard-to-impossible to agree on the existence of 'the problem' and/or how 'big/important' it is.

I have moved from scientific research to the tech industry working on inane problems to working in health data science. I get infinitely more reward working on what I feel are important issues than I did working for something I didn’t care about (even though the money was much better!).

The important issues as I see it:

* improving health care, access to health care and generally helping people to live healthier lives.

* helping in some way those in poverty around the world.

* Climate change

There are more big & important things than can be listed. I think the best idea I ever heard is to pick the one that you are most energized about and do that.

> I keep seeing critiques against working on tasks that are meant to maximize user engagement.

Well, sure, because that's work against the best interest of users. Even if there were no big & important things to be done, that's one that should not be done.

With regards to global warming, one important task is to more accurately and convincingly measure the cost of climate change. People are already working on this topic, but all the work is quite napkin-mathsy at this point, and of course heavily politicized. If one could very convincingly predict the future cost of climate change it would do a lot to convince people to pay today to avoid the worst in the future.

It's an interesting dilemma; while maximizing user engagement has its place, focusing on solving pressing societal challenges is equally crucial. Issues like climate change, healthcare accessibility, and social inequality are significant problems that require innovative solutions.

Perhaps a balanced approach can lead to meaningful contributions that address both user needs and broader global concerns.

The biggest problem IMHO is preventing AI research from killing all the people.

(In contrast, people's becoming enslaved is a much smaller risk because the kind of AI capable of enslaving people will probably also be able to create robots that are more reliable and efficient than people at whatever task the AI is contemplating enslaving people for.)

I think Maslow's Hierarchy of Needs is good place to get inspiration for this. It's guided my last two job roles in agritech and biotech. But I'm becoming aware that there seems to be an (inverse) correlation between how much money/funding is available as you go down the triangle.

Self driving vehicles will eliminate the leading cause of death for grownups under 40.

For those concerned about CO2 (as they should) don't forget that were it not for human CO2 we would very soon be on the cusp of another ice age. The point is that we're at a tricky balancing point either way.

We need new ways of living in large-scale societies.

Current political systems at almost every scale are still essentially just monkey-troup dominance hierarchies.

Market capitalism does not work at this scale, or rather, it works too well, converting everything to paperclips.

We want the dynamism and innovation that markets bring, without the rentier-ing and enshitification. I notice a strand in recent reporting trying to distinguish 'rentiers' (bad) from 'capitalists' (good), but it's a false distinction; capitalism almost by definition is people with money using it (by owning the means of production, or of existence) for the goal of making more money. All other goals seem to have become incidental, and the environment, social harmony, individual wellbeing and sanity are all victims.

So many problems stem from the incentives these broken systems create. Unconstrained fossil-fuel use, deforestation, pollution of the commons, junk products designed for obsolescence, junk addictive diets, corruption, rash emotive decisionmaking, and on and on.

See the recent "Ask HN: Problems for the next decade?" https://news.ycombinator.com/item?id=36651506

Affordable electric vehicles

Get all humans on earth to ponder deeply on the question "Who am I?" - not until they find an answer - but until the question disappears.

This is the end of misery.

The biggest, most important problem to work on right now is any and every problem that's currently leading to the destruction of our planet's natural systems. Unchecked greed, rampant extreme inequality, resource abuse/waste, dis/misinformation, political extremism, all of the general stupidity that many (most?) of our richest, most powerful humans are guilty of encouraging/contributing to in the general populace of Planet Earth these days. If you can "increase user engagement" in anything that improves one of these critical areas, or anything related, you're doing humanity (and all life on Earth, really) a service.

Diseases of aging

If you hear it on the news everyday, it is not a important problem

LLM context window limitation.

Making child care affordable.

Cure for cancer

universal access to internet, universal access to hover powered transportation devices.

Make star-trek replicator.

Being a good person.

Caring for others.

Resource allocation.

The housing crisis.

We need to widely adopt Operating Systems implementing capability based security.[1] It doesn't help that there is widespread confusion with the horrible permission flags systems in phones, tablets, etc.

Without capability based security, our current situation is analogous to having a power grid with no breakers or fuses, anywhere. (One bad load would take down the grid, or set things on fire, etc).

We've improvised systems to make up for the deficiency including Virtualization, Containerization, and systems like WASM. Mainframes hold on partially because they run processes without ambient authority, which is also a form of capabilities.

The general inability to secure general computing, in turn, leads to practices like blaming the users, or applications, or operating system vendors. It leads to band-aids like virus scanners, and immensely draconian IT management. It threatens to lead to government regulation of IT systems and US.

It also leads to users playing it safe. Since you can't safely run a program, no matter how clever you are, you're going to avoid any novel software, or web sites, to try to avoid compromise of your system. This leads us to Facebook, and all the other walled gardens. It also leads to app stores and all the evil that entails.

This leads, eventually to the loss of the war on general purpose computing.

---

Next, the Von Neuman architecture has hit its limits, we need to explore other techniques to get the most out of the billions of transistors we can put on a chip. I have some ideas[2], but welcome others.

As Kevlin Henney states, when you introduce concurrency, you change the laws of physics for software. I don't think this is widely understood.

---

We need a Memex, but copyright restrictionists will fight extremely hard against it, as the primary purpose is to copy information and the context it belongs in.

---

We need to push for internet connectivity, instead of internet "access". Everyone should be able to run their own hosts, servers, and services.

---

We need to build a second supply chain for everything, with full open source documentation. There should be laws to allow commercial trade secrets to be escrowed for some period of time, perhaps as long as 50 years, with the national archives, or other suitable organization.

Nothing, anywhere, should have a single source.

---

[1] https://en.wikipedia.org/wiki/Capability-based_security

[2] https://esolangs.org/wiki/Bitgrid

IMO the #1 biggest most important problem is the design of institutions.

Specifically we need to design institutions that can accept higher information throughput from their members. This applies to corporations, governments, academia, etc.

When I say information throughput, I mean, accept information from its members and change its own behavior in accordance. Our current institutions were designed when people had to travel days to go contribute to law writing etc. That's no longer the case. The amount of information going to and from our institutions is much larger, but the institutions themselves have not scaled. Institutional processes can be very similar if not equivalent to software. And there are a lot of preventable problems just from scaling.

For instance look at our red/blue culture wars and how obviously it corresponds to population density... maybe we need to have variation in policy based on population density. Instead whoever is in power pushes their policies on everyone. When a republican is in, the cities suffer, when a democrat is in, the countryside suffers.

We also have corporations making decisions that would not be made if the shareholders and employees were given all the information and polled for the best decision.

We have policies that are made "for the greater good" that don't compensate those who lose out. This makes them unpopular. Any policy that is against fossil fuels should also focus on providing compensation to the millions of people who depend on it for their livelihood.

We have insane amounts of waste in the government and 20% of people think their jobs that they're forced to work are meaningless.

The fact that there are some readily identifiable, strong patterns for some of them indicates that there could be a relatively straightforward cause. IMO these problems aren't impossible to solve. They just need a systematic approach. They need to be studied in academia like we study physics. Imagine if we spent the amount of money we do on particle accelerators to instead create a few dozen "test cities" where people can move there and experiment with different economic and social systems.

Until then, though there's still a whole lot of theoretical work that can be done understanding how information flows through an institution and ultimately affects and constrains its behavior