Some ideas...

Autonomous Management/Governance

The main function of managers of all stripes (according to me) is to solve coordination problems. However, there are also “autonomous” systems for solving some specific kinds of coordination problems - markets, in particular, are a major example. But these are limited to problems which fit their interface well - i.e. a high degree of standardization, in the transactions if not the products themselves. A requirement for large amounts of context, or difficulty recognizing success even in hindsight creates problems for markets.

I could imagine systems which solve (some) more context-heavy or alignment-limited coordination problems, with the scalability and decentralized “autonomy” of markets. Management/governance via prediction markets is one class of ideas in this vein, but it seems like modern information technology should open up many other possibilities. The way I picture it, it's not so much "build an AI to run a company" as "organize the company so its constituent people together implement the same algorithm the AI would". Just like competitive markets perform a variant of backprop and gradient descent, other AI algorithms should have equivalent structures in human organization.

Information Indexing

In recent years, availability of information/data has grown much faster than our ability to index it. Matching particular people with the highest-value information for them - i.e. indexing the information - is largely a social technology problem, though one which benefits a lot from computer science.

Suppose I'm researching mutations of mitochondrial DNA. Perhaps the topic has been studied by a particular group separate from the main field, with different aims and entirely different jargon. A text search for the normal jargon won't turn up their work, no matter how helpful and relevant it might be. In order to find something like that, we need some kind of jargon-independent indexing. For instance, if we had a detailed world-model, then we could link the information directly to structures in the world-model (i.e. modelled mitochondria). If we had extensive augmented reality, we use object recognition to recognize mitochondria visually, and then index information to that - so that even research using a different word for "mitochondria" would pop up when looking at a picture of a mitochondrion.

More generally, it would be useful to be able to index information based on the real-world patterns or mathematical structures they involve. This seems like something which could, in principle, be implemented purely socially, if we knew a good way to do it. (I'm sure people have tried things like this before, but usually projects like these try to use data structures which are convenient for programming, rather than data structures which represent the world well.)

More Efficient Notation

People joke that Einstein's greatest invention was tensor index notation. Good mathematical notation makes a surprisingly large difference, as anyone who's used both assembly language and python can attest.

It seems to me that the main trend in mathematical notation from ~1850 to ~1950 was to introduce really bad notation for things which previously had no formal notation at all. (This is part of the more general trend of introducing not-very-good foundations for things which previously had no foundations. Set theory and Turing machines and the like were better than nothing, but they are more like assembly than Lisp.) I expect in the future, progress will come more from improving notation, as a side effect of the general project of refactoring mathematical foundations.

Continuing improvement in programming is, of course, a similar story. In that department, I think progress will come from better understanding the structure of the world around us, and incorporating those insights into programming languages (abstraction is one potential example).

Model-Based Financial Markets

Similar to Abram's suggestions for model-based prediction markets, as automated trading gains share of the financial markets, there's potential for financial systems which take advantage of the legibility of automated models.

In particular: modern markets direct capital to existing projects. Traders probably have some views about what projects would be really good investments, but the market does not have an easy mechanism to share those views with business founders/managers during the planning phase. Instead, it's guess and check: a founder/manager tries a project, and then finds out how much the market likes that project. Part of the problem here is that project-space is very high dimensional, so it's not easy to compress all the information from the traders into a low-dimensional price signal.

But with transparent automated traders, we could in principle use a backprop-like mechanism to identify projects which the traders think are promising, without the slow guess-and-check process.

So let me give a babble answer.

1. More visceral history education with VR. Literally experience the past. Maybe even role play it before you gt spoiled. Get answers like would you have resisted nazis?
2. Consent tracking. Have politicians tht can be literally voted out of office midtermby not having terms but a instantly transferable "representation" relationship that voids if not actively reassigned in a period that would correspond to the term.
3. Measure and recognition of creative input. Track who composed, remixed, balanced, commented on live-stream.
4. Explicit argumentation that can be explicitly checked. Find fact and fiction speakers. Courts can trace whose opinion was followed. What adverts claim can be vetted for what kinds of emotinal appeal are ok.
5. Universal addressing system covering all physical and conceptual locations. math/geometry/angles/trisection. Milkyway/sol/earth/australia
6. Language covering visual and tactile experiences. I think notes cover a lot of sound already and conceptual space is already part of having a language. UFO sighting would not be vague as any claim could be maid explicit. 
7. Large coverage emotional language.
8. Codifcation for reasons to deny answers or lie. I have never killed an animal (scope tag A4). Improper questions by employers would be answered but woudl be amarked to be answered as business matters.
9. Being able to personally microfinancde abstract values like science. Ie put 2$ out of your monthly paycheck into language nobel prize.
10. Being able to address private financial power with values restrictions. 100$ usable on non-guns. 50$ usable on non-pornography. 30$ usable on food.
11. Ideological conflict group therapy
12. Codification of will so that it can be stated as legacy so that off-spring don't have to imagine or speculate what would have grandpa tought about issue X.
13. Large coverage peer support groups. Have a anonymous group for any status or property. People with houses. People of parents of ethnicity X and Y. People with medical conditions both X and Y.
14. Codified old sayings. Have a ledger about new and statements that have stood the test of time.
15. Large coverage personality trait identifiers. Use on dating servicdes to assess compatibility. Use on police to have psychosocial identifiers. Use on psychology to asses risk factors and make diagnoses more precise
16. Have sources listed in speech/thought for truth verification
17. Any publicly broadcasted speech as autobet for its veracity.
18. Microlaws for households
19. Fines for talking over people. Fines for micro-aggression. Explicit notifiers of small degrees of botherance.
20. Standised job descriptions for automated matching of talent and need.
21. Standardiser speculation. Explicit threats of action on conditions.
22. Broadcasting plans to enable strangers acts of kindness. Match picture takers before either party has set of on the travel. Fill empty car seatas. Find series of empty car seats for automated hitchiking.
23. Spot education on single words or needs.
24. Catalogue of microcultures. Use to manage mismatches in expectation. have a reliable way to point to rare cultural conditions.
25. Ownerless property that any person can try to upkeep. Designate a park and allow other than founder to upkeep it without selling or employment.
26. Explicit palettes of social choice with large coverage. Have lots of people have opinions on lots of questions and ways of registering more. With consent people can see how friends or enemies they poltically are and maybe can automatically stay clear of foe waters.
27. Large coverage language to be able to describe body motions in detail. Mechanical skills can be said out loud instead of learned from experience or training.
28. Micro-employment. Have tasks which have small rewards attached. Have it viable to be an odd-job filler without having a job description.
29. Recognition and tracking of socially desirable non-work that can be appriciated like work.
30. Being able to describe and compare meaningfullness of issues across persons. Rate art. Make economic decision on wish fullfillment.
31. Ask and design for weird reality tv formats. Have a scheme to give consent for weird things without spoiled. Make consenting for Truemanshow practical.
32. Boost empathy with AR and emotional monitoring and broadcasting
33. Make new gestures for information age needs (sauce face)
34. Schedule roads to not have traffic jam hours
35. Topic interest referral networks. Have a desk where you can go for weird facts.
36. Active community builders that solve and benefit social well being. Detects upcoming arguments, unhealhiness in dependencies, opportunities for romance, friendship. Have it be non-creepy, non-authoritarian and respect privacy.
37. Systemized collection of drunk movie pitches. Services to get feedback on promising new songs.
38. Policy deliberation via roleplay or reality tv show like settings.
39. Being able to establish new rights that you personally respect and have them efficiently be broadcasted for others. "I will provide directions for any body in my home town". "I will always end questions with please"
40. Literal telepathy with brain to machine to brain interfaces
41. Shared recollection with stored memories from brain to machine interfaces
42. Society level cognitive processing with holidays designated to particular cognitive tasks. (Predict next year day, imagine better next year day, plan for better next year day)
43. Common coginitive shared facet detection with averaging over indidvual imaginative inputs (read everybodys dreams and recognise the median to inform zeitgeist)
44. Being able to place financial incetives for opinions how production should be done (ie 0.30$ for good liveable wage employee products, 0.50$ for sustainable energy usage)
45. Scheme to push memes by paying money. Pay 10 000$ to increase people that wear your design jacket. Receive money for not making personal clothing choice but adopt a pushers design.
46. Abolish and outlaw poverty
47. Ways to distinguish good critics (workably define what that even means)
48. Being able to represent earth and sue people in the name of earth. CO2 emissions as crimes against the planet
49. Virtual personalities which can be influenced/defined by multiple actual people but are their own legal personality.
50. AIs in equal or comparable footing to in legal terms to corporations
51. Legal status for principles that do not have any hardware support (ie laws that don't need to be interpreted)

In jurisdictions where there's an exam required to graduate high school, let students of any age take the exam, but have a sufficiently higher cut-off to graduate early. Anyone who graduates early is automatically admitted to a university (e.g. in the U.S. it might be to their choice of state university) and received a tuition subsidy at least equal to the amount it would have cost the public school system to keep them in high school.

• Some way to publish a book as an excludable good. For example, you could have a movie theater, but instead of a movie screen, you have temporary access to a book. Someone watches to make sure you don't copy the book.
• Here's a sci-fi-ish one: A system of safeguards and filters around your attention. A user scrolls the internet while advertisers, narrative-pushers, and propagandists try in vain to slip their subtle manipulations past the user's gatekeeper.
• Where's my flying car? Get on it scientists! Where's my Conditional Proof of Stake Hashcash? Get on it crypto devs!
  • "The idea here is that we set up a smart contract mechanism where along with an email the recipient gets a secret key (the preimage of a hash) that allows them to delete some specified amount (eg. $0.5) of the sender’s money, but only if he wants to; we expect the recipient to not do this for legitimate messages, and so for legitimate senders the cost of the scheme is close to zero (basically, transaction fees plus occasionally losing $0.5 to malicious receivers)."
  • It's a way to price spammers out of your attention. Awesome.
• Dominant Assurance Contracts?

Better community/meetup advertising via large scale text analysis to find unique features in text outputs and make sure those people become aware of each other. Like imagine if goodreads suggested friends to you based on rare book overlap but for non user curated data.

Digital Rights Enforcement Agencies

Given a desire for digital rights in the face of Crypto-Anarchy, market-based polycentric law might yield a solution.

David Friedman's model for market-law involves defense agencies and arbitrators who mediate between those agencies. The system is stable as a repeated game, wherein the cost of fighting other agencies is higher than the cost of peaceful negotiation.

In the digital world, a 'defense agency' might look like a professional hacking group. This group would maintain a public identity and offer it's services to clients that won their case in court. Occasionally groups fight each other, in epic Neuromancer-esque style.

Friedman's theory uses social-norms (e.g. property rights) as the basis for efficient negotiation between agents. Therefore we might predict that a demonstrably neutral arbitration protocol could form the basis for this new market-based law.

Prosthetic neocortex that permits dunbars number to increase dramatically for the individual using the prosthesis.

Quantum trading: you get access to some information. If you do anything meaningful with it, you have to share the result with the owner of said information (or do a profit share). At any time you can delete this information from your mind (as well as any derived insights).

Maybe you only have a limited time to consider whether or not you want this info.

Maybe it’s more of a barter system. Both of you exchange info and if you both like it, you keep it. But if either party doesn’t like the deal, you’re reverted back to the before deal state.

Expertise tracking via credence calibration is a possible technology that would be helpful in areas where an expert makes a lot of decisions that have clearly measured outcomes.

I wrote prediction-based-medicine to describe how the system might work in that context. 

Another area where I would expect this technology to be useful is reducing recidivism. Parole board members should put down their predictions for the recidisvism of individual prisoners before those get released. Various actors in the prison system should make predictions about the recidisvm of released prisoners. 

Better conlang for regular communication

I think there's a good case to be made for Esperanto already being better then the naturally evolved languages we have. It's easier to aquire a large vocabulary and speak it proficiently. At the same time it was developed by one person in the 19th century. It's likely possible to create a language that's even better.

Esperanto focused on being easy to learn and not on actually improving in areas where our European languages were lacking. Having a few more stems and structures might increase the work to learn the language but it would make it more powerful.

When Google Maps tells me to move into the direction of northeast I feel that our language fails us. In the aviation industry they try to patch that flaw of our language by speaking of X o'clock. It would be good to be able to express more fined grained directions more natively then through the X o'clock hack. 

European languages force speakers to constantly communicate information about gender and Esperanto copied that flaw.

Our languages are often more ambigious then desireable. 

You likely want an easy way to mark a word like between as being meant inclusive or exclusive. 

In our Eurocentric languages a few words like "to be" and "feel" are heavily overloaded with various different meanings. Spreading that meaning over more words would help remove ambiguity. 

Multidimensional pricing

The benefit of a single scalar price is that it is super easy to compute, and lets you compare different types of value in a like-with-like fashion. The downside is that in order to compare different types of value, you have to sacrifice all of the information about it; there are probably arguments where this is considered a feature between two simple agents, but in meatspace there is a large problem with organizations not having good internal knowledge which means prices conceal information from yourself just as much. Secondly there is the eternal matter of externalities, which by long standing tradition are impossible to address outside of regulation. This is because we have no good way of assigning value to things not happening.

I think a multidimensional pricing model would help with this, because it would allow all the elements of value to be evaluated independently, including negative prices for things to be avoided. Then the whole thing could be computed as a single eigenprice, giving us the same overall benefit as regular dollars, but keeping all the relevant information. We are very good at automatic computations of this type now, so it seems feasible as long as something as sophisticated as a smartphone or engineering calculator is available. There is no reason a bank or transaction platform could not have this built in.

Generic strategy models

I've never been able to let go of Jaynes' notion of using the phase-space of macrophenomena for prediction, which gives me a strong feeling that there is a fully general and scalable method of generating strategies. If we think of strategies from an action perspective as being a combination of predictions+interventions, it seems like any desirable end-state could be approached.

This should help a million and one things, because at least the United States is pretty strongly anti-strategic. If it became more consistent, cheaper, and effective, then I expect adoption to go up and overall efficiency as a result.

Full spectrum group alignment methods

At some point in the relatively near future we will start putting together the pieces that allow the entire spectrum of human experience to be brought to bear in building group alignment. This will include tools traditionally shunned in developed nations, like violence and fear. Of the three ideas I suggest, this one feels the most inevitable and simultaneously the most double-edged; most of the legwork to date has been done by professional militaries and religions, but there is plenty of interest in investigating things like trauma, abuse, and political movements gone haywire. As far as I can tell consent is really just one more variable in this space, because otherwise conscription would not be viable as a military method, and people seldom set out with political or religious radicalization as their goal. As a consequence, once these methods are well established I expect them to work on whoever happens to be within reach, rather than requiring voluntary association.

Good Ones

A better version of social media that fixes its flaws. Causing less outrage but fostering constructive collaboration. Paul Graham repeatedly hints at this e.g. here and here (these are not the best quotes but those I could find quickly). Maybe something with a distributed trust or social credit system like Advogato aimed at. You will feel safe there while getting valuable input and feedback with a balance of confirmation and relevant new information. 

Schoocial Media (name suggested by my son). A combination of the good ideas from 

• social media (getting in contact with the people that let you grow most), 
• Lambda School (incentives aligned between educator and educatee), 
• Khan Academy (free high-quality lessons), and 
• Dragon Box (teaching elementary education in a fun a natural way).

I'm not sure how this would actually look like but my son seems to have some ideas. One is that memes can summarize actual core scientific insights in a memorable way and can lead to exploration of what is behind it - e.g. on Khan Academy. He has shown me quite a few cool ones e.g. about relativity. 

General Public Contract - A growing cooperative society - by some called a cult - that applies a positive-sum mechanism to build a better society on top of - or embedded in the existing society at large. It does so by using a valid legal contract among all parties. A contract like the "General Public Virus", that requires participants to contribute to it to gain its benefits. One extreme example would be a contract that requires you to effectively give up your private property except for your immediate belongings in exchange for access to the net utility of the property managed under the contract. 

Bad Ones

Real Social Engineering. Technology that models motivation and communication of large numbers of humans accurately enough to predict the impact of communication acts or to select the most promising communication scheme for some target metric. Today's click optimization will look like child's play against this. Most likely it will not be capable of predicting global trends like Asimov's Psychohistory but good enough that everybody not having access to this will be the future Third World. Among those having it, it will be zero-sum of course - and tricky, because fixed-point algorithms will need to be developed. I once read a short story about something like this but can't find it. Any takers?

A Caricature of Science. One of the very bad things that could happen is that significant parts of science could lose public trust by not delivering or even worse by falling prey to partisanship (esp. beyond a single country). I fear a religion-like science cult where results are more influenced by group expectations than by the experimental method and hard statistics. I think some people would say that we are not that far away from that but I think it could be much worse.   

Fake Solutions - I expect a lot of structures that promise to solve real or imagined flaws of society and/or that fight the good solutions that are resistant to exploitation (like the GPC above). Organizations, companies, NGOs that appeal to moral values but don't deliver. Mostly because of the mechanisms outlined by Robin Hanson e.g. here and here. Or by SSC here. 

To foster the good ones and counter the bad ones, the Long View would suggest starting early at finding and categorizing and creating transparency about them.

A school where students spend much of their time in mostly-unsupervised independent work and/or socializing, and they are (individually or in small groups with similar ability) matched with tutors on specific topics. I think this would work much better than the one size fits all model we use now.

An elected office where there's a term limit, and some length of time after someone leaves office (e.g. in the following election), voters vote on how good a job they did, and the former office-holder receives a cash payout or pension based on the result of the vote.

The benefits of this system would be:

• Politicians would have an incentive to focus on how their performance in office will be judged on a longer time-horizon.
• If the payout is large enough, it would incentivize politicians not to take sinecures and "speaking fees" that would be seen as corrupt after leaving office (and thereby make them less useful as an inducement to corruption while they're in office)
• During an election, the question "how did it go last time this party was in charge?" would be top-of-mind, encouraging political parties to optimize more for the long term.

Equity markets for individual products. Essentially, a company with a design for a new product creates a Kickstarter-like campaign where you can back the product, but in addition to (or instead of) getting the product when it's released, you get royalties on each one sold (e.g. a company promises $5 in royalties per unit sold and sells that for $250k, and investors can buy fractions of that).

This would enable companies (and individual inventors?) to de-risk and apply wisdom of crowds to new product development.

The most important social technology we have is government. Government has a very important role: to prevent members of the same country from killing each other. The biggest flaw in our current government technology is that we have no good system for coming to agreement among the 200-ish countries in the world. We have had the UN for the past 75 years, but there have still been plenty of wars and plenty of fears of nuclear war, and there's still a good chance of nuclear war in the future. Nations still spend large amounts of money on armies that are essentially wasted from the point of view of humanity overall.

A more advanced social technology would prevent war among countries, at far less expense than the current expenditures on armies, but more importantly with a lower chance of global nuclear war. We need some political way to consistently resolve international conflicts without having them spill over into war. It's somewhere between legal technology, mechanism design, and coordination technology on your list.