Hacker News Reader

Comment by gvicino | Hacker News Reader

Hello,

Thank you for the link you've shared. I read it with great interest. While I can conceive that the aforementioned areas might survive in a post-industrial civilization, it is unclear to me how the technology that 100rabbits is researching could do the same. They might use Raspberry Pis or old PCs donated by others, but:

- these are impossible to manufacture without industries, and they are limited in number

- the only means of communication they're using to share their work, the Internet, inherently requires an industrial society to exist

This seems somewhat akin to the naive reaction to the financial crises we experienced a few years ago: "Oh dear, banks and finance can be evil, let's decentralize with Bitcoin."

I believe that we cannot address the problems that technology has brought us (nuclear weapons, global warming) without resorting once again to technology and science. This technology and science must necessarily be funded by the market.

View on HN

>This technology and science must necessarily be funded by the market.

Why?

We are a planet of 8 billion people. Through what other supply model could we ensure the production of computers and technology for everyone (or even just for some) if not the market? Here I found an infographic that shows how the supply chain for the production of a PC works:

https://scmresearch.org/wp-content/uploads/2018/09/supply_ch...

This also applies to science. The Covid pandemic is almost certainly a result of global warming caused by human activities, but the vaccine for Covid was found thanks to a shared effort of humanity, encouraged by both government policies and, above all, the profit of pharmaceutical companies.

I am well aware of all the flaws of capitalism, but I can't understand how we can have an "information age" without a market. We can try to reduce our impact, learn to decentralize, learn to fix things that break (all things I personally do), but I can't envision a decentralized technological and scientific world that wouldn't set us back 300 years.

Your covid point reminds me of the Jonathan (oops forgot his last name) inventing rna tech on his yacht :)

And trying to create a market, also from his yacht, but alas also not very successful

>I can't envision a decentralized technological and scientific world that wouldn't set us back 300 years.

And yet you are here arguing for more capitalist market-based solutions, which are inherently decentralized and have a tendency to create unrelated, wasteful (duplication of effort), and often pointless competition and profit-seeking?

Very odd.

Let's not mince words:

- The decentralization of markets (criticized by some, though I'm not among them) is often condemned because it can lead to elements of economic power centralization. Critics argue that it makes people dependent on tools and technologies controlled by a few, and driven by profit.

- The decentralization proposed in the technological field is closer to anarchist ideals and suggests a kind of decentralization implemented by individual people and groups on a voluntary and mutualistic basis.

I don't believe the latter can truly benefit humanity because I don't think we are that evolved. Individuals primarily seek their own personal gain and are often so uninformed as to not understand that this personal benefit is only achievable with the benefit of the community. I believe that the market (controlled and monitored by nations and supranational entities) functions better because it is an imperfect model that, by leveraging people's selfishness, compels them to collaborate and communicate with each other.

>The decentralization proposed in the technological field is closer to anarchist ideals

Anarchist ideals? From right-wing billionaires like Musk and Thiel? Or by centrist billionaires like...?

Because that's who is driving the technological field, ultimately — people who have more money than entire countries, to whom we are basically just bugs to be squashed if we bother them.

>I believe that the market (controlled and monitored by nations and supranational entities) functions better because it is an imperfect model that, by leveraging people's selfishness,

lmao

This doesn't feel like a serious discussion. I'm out.

Please reconsider..

gvicino has important points regarding the fine structure of collaboration regulation.. he says it's a "belief", which means it could still be open to discussion!

(Imagine, OTOH, if he started with "reason tells me.." :)

Anarchism isn't a belief, but rather a political-philosophical framework with some very specifically-defined foundational aspects.

Imagine discussing math and programming with someone who insists that pi=3 and GOTO is the best control flow mechanism.

pedantry:

a) A friend made their pile developing Atari 2600 cartridges. In those days, doing a pythagorean distance calculation would've been way too expensive (especially as they had to work in between scan lines), so in their programs hit regions were octagonal, which is not pi=3, but not far off from pi=6.

b) As someone who generates object code, GOTO is the best control flow mechanism in that domain. [see also "Lambda: the ultimate GOTO" (1977) https://dspace.mit.edu/handle/1721.1/5753 ]

Thank you for your curiosity!

Decentralization is a very long term effort -- Grothendieck said it would take generations, we can't expect immediate results. So while 100R certainly can't survive without relying on industry, it does do research into technologies that may be easier to "deindustrialize". See the "off-grid" and "sustainability" pages on their site.

Finer point: even before the industrial revolution, humans have had access to supercomputers -- their own brains!

It's conceivable that in the not so far future we will be able to grow computers of our own design, from just air and water.

See promising efforts in this direction, it's not an empty dream! https://youtu.be/bEXefdbQDjw

I guess the moral of the day with regards to banks and finance is, don't jump the gun! We will need them until we can grow raspberry pies on a tree!

I understand your point of view, but I'm not sure we're that far ahead in progress. It seems very solar punk science fiction if we're talking about the near future. If we're talking about a more distant future, then I can still hope for it.

I wouldn't say the horizon is 100 years, but 250 years. Mainly to re-architect finance to support relevant research. You know how psychological obstacles are the hardest!

The first milestone is to minimize the use of metals in our society, use as dopants or catalysts only. Second milestone is alumina. (Money OTOH will probably always be necessary!) What I wanted to explain above is that for computing, arguably our most advanced tech, metals are not strictly needed.

You would also probably be surprised at the true state of our solarpunk tech. There are water antennas, solar-pumped lasers.. that's almost 1950 already.

Thank you. It seems like you have a clear path for this transition. Do you have any reading suggestions that explain this transition?

Seconded, appreciated. Would love any materials to peruse ...

For the research portion, there are (unfortunately) lower impact journals like where the following paper -- about replacing aeronautical alloys with alumina-- is from:

https://www.sciencedirect.com/science/article/abs/pii/S22149...

For organoid intelligence, more hyped so you have higher impact, free journals https://www.frontiersin.org/journals/science/articles/10.338...

Japan has a lot of interesting not-quite-industrial family run high tech businesses. Solar pumped lasers I mentioned, but there's also crystal growing, small scale almost cottage level hightech paper and wood makers. Few reading materials though. Some of this "movement" will eventually diffuse to China, Korea and finally India.

These are about nearer term substitutes, a lot of them still involve industrial processes, but it is much easier to tweak industrial processes to turn them solarpunk than to replace entire classes of materials.

For the finance angle, I'm afraid I won't be able to direct you to any good free roadmap for Year 50 to Year 250, but for Year 0 to 50 I recommend studying how governments in the asia-pacific are funding decarbonization R&D, especially in SMEs. (Although we might be more interested in Africa or India?) Pay attention to IP law reform.

E.g. japanese gov report here (I was especially interested in the optoelectronics, for reasons of "demetallization", but Japan has small very localizable optics firms, think Thor but localized).

https://www.mof.go.jp/english/policy/jgbs/topics/JapanClimat...

   R & D expenditures for optoelectronic fusion technology that is expected to reduce energy consumption by up to 40 per cent in data centers. Optoelectronic fusion is a technology that fuses circuits that handle electrical signals and optical signals. Calculations have been performed with "binary numbers" by switching on and off of the electricity in conventional computers.Electricity however generates heat when it flows through circuits, and energy is used to generate heat that is not originally necessary, and when it generates heat, the resistance of the electric paths increases,leading to decreases in the calculation speeds. Therefore, research is underway to replace calculation using electricity with processing using light.Power saving and low latency are achieved by connecting internal circuits in computers with light without using electricity as much as possible. The aim is to gradually introduce light into computing chips or peripheral equipment in computers that have been processed by electricity.The aims are ﴾1﴿ to establish a technology that connects chips used for calculation and peripheral equipment with light in 2024, ﴾2﴿ to connect between chips with light in 2024 and ﴾3﴿ to practically realize a photoelectric fusion chip that calculates with light in the final stage in 2030. There are some estimation that the spread of optoelectronic fusion technology will save more than 40 per cent of energy by 2030 compared to the current state-of-the-art data centers.

EDIT: found this: https://youtu.be/lzHqhNoyx2o

This is at commercial scale, but the process is very simple, you can even do it at home! (And of course, barely hinted in the video many mom and pop shops in Japan doing it. It's not very far from samurai just making swords from beach sand.)

used to be many small town companies in the US doing this, you can imagine, IF THERE WAS DEMAND from your cow rearing neighbours..

I wanted to really appreciate your comments. Thanks for taking the time.-

Thanks for paying attention. Central Europe also has a very similar low volume ceramics/optics SME networks, but they are also much more industrialized and energy consumptive.

Requires boots on the ground to see if the energy shortages will lead to these networks evolving "solarpunk" tech. (They are also similarly secretive). Let me know if you see any trends in that direction!

I can tell you from AP that people are trying to reverse engineer "secret" Japanese tech, reverse engineering is sacred work :)

> Central Europe also has a

Central Europe out of all places.

> very similar low volume ceramics/optics SME networks

By which you mean - excuse the question - "SME networks" as in "business-networked" mid-size commercial concerns doing low volume production in the electroptical field -or- production networks running optoelectronic hardware at a small scale? Or, both ...

Haha the first one, but also "small-sized" (in addition to mid-sized)

Also want to point out to academic spinoffs (which most of these SMEs are) like

https://www.kit.edu/kit/english/pi_2023_041_nanomaterials-3d...

They have been "incorporated" but still seems like they haven't got the attention they deserve? (Okay the temp/volume is maybe not low/high enough, but the process is almost domestic kitchen level simple)

Wanted to address your question about decentralization of the supply chain. We can't decentralize computer manufacture at the moment because there are too many specialized processes and materials in VLSI & beyond, many of which are secret/not published for no good reason. Okay I consider rent seeking not a good reason :) (Think Bell Labs vs post Bell Labs era) you saw the video of that kid making chips at home? It isn't remotely like cooking meth, AI isn't going to help. What is needed is radical simplification of chip design down to the atomic level.

Beyond the chip level, tech becomes more decentralizable -- soon you will replicate many Shenzhens in India, but at that point you hit the diffusability of polymer and metalworking tech.

No reference here, but money diffuses faster than technical knowhow, that is the issue, that's why markets seem like magic.

See the quartz growing video I linked, I'm surprised that I'm surprised that this almost stone age but highly used tech is largely "secret" compared to random JavaScript tricks.

Let's hypothesize that from tomorrow, the knowledge to produce everything at home "DIY" becomes available because it's free (and even today this is increasingly true as the concept of patents is decidedly in crisis). We now possess all the possible knowledge to do everything ourselves. However, one thing remains missing for us to become self-sufficient, and it will always be missing regardless of the economic or political systems, as it remains a finite resource: time.

We can afford to imagine new systems and models because we have significantly more time than our predecessors. We have more time than our ancestors because we are highly specialized, and we exchange this specialization on a local and international level for services, goods, and raw materials.

In the scenario you propose, would we have the time to search for raw materials, study to understand how to process them, create the objects and tools we need, and simultaneously dedicate ourselves to other things (i.e.: while we are doing all of this, we need to attend to children, family, and the fulfillment of basic needs such as food, leisure, sleep, illness, etc.)

In my opinion, it isn't possible. I would more readily understand forms of anarcho-primitivism in which people decide to return to nature and primitive values, either forcibly (which might be feasible) or voluntarily (which seems highly unlikely). At least these do not violate the spatial-temporal limits in which we live :)

Friend, I love being as independent as possible, fixing things by myself and maintaining a garden at home, but I'm aware that it's just a hobby. It's a hobby I can indulge in because for other needs, there are others in the market to lend a hand. If I were to have a toothache tomorrow or if my glasses were to break, in a fully decentralized model like 100rabbits, I'd be in trouble :) Or perhaps, I'm still missing some pieces of the puzzle they propose.

Hello. Thank you for engaging in good faith and friendliness..

I'm not sure what is this "scenario" that you think I'm proposing ;)

You seem to like spending time on HN, as do I :)

Are you a resident of Trieste, by any chance? rOtfl

Let me stress, anarcho-primitivism isn't the label I would like to have thoughtful people put on my "scenario", haha!

I will outline my "scenario" more carefully, promise! I realize what I have written here so far is still ripe to be misunderstood. Sorry!

But I also have a job (,i.e. something equally crucial to the delicate art of time management) and that takes a lot of quality time as well ;)

(Did you notice that rearchitecting finance is the top priority in my "scenario"? money=time=money?)

Thank you for taking time out to formulate these important questions.

> or if my glasses were to break

was that a https://en.wikipedia.org/wiki/Time_Enough_at_Last reference?

I think your reference to time as a limiting input may be very close to the objection I'm groping towards in https://news.ycombinator.com/item?id=39844527 , https://news.ycombinator.com/item?id=34160852 , and https://news.ycombinator.com/item?id=33952298 ?

Thank you! Keep them objections coming, especially if they have a geometric or siliceous slant. Helps me to focus :)

On second thought, these are more than enough lol

Just a quick and dirty counter-objection-- to satisfy a Grothendieck model as well as yourselves, perhaps in the decentralized future, your glasses get fixed purely as a side effect of research on silica?

So,really, the problem could be, as you implied, success in silica research is very random (Poisson?), but the need for silica manipulation is some other unfortunate mismatched distribution? You would think Russell already worked that out!

I.e. why are you guys unhappy. You're already getting well-paid for managing!! (For my part, my other research feels like a rachet and not a controlled spiral into nothingburger, which is what I have to label the unhappy ones)

Of course the tongue-in-cheek way to improve morale is to improve research until it feels more like labour :)

EDIT: Less tongue-in-cheek guess is that interacting time-utility functions (i.e. connected "counter-rachets") are poorly worked out,or at least, not grokked by the unwashed masses (but very well understood by your Grandma)

https://en.wikipedia.org/wiki/Time-utility_function

Not sure I understand the ratchets, so continuing with the theme of broken glasses:

Not so long ago (200 yr? 300 yr?) almost all people spent a significant fraction (2/3+?) of their time just keeping themselves fed. We've automated enough of that that even with distribution, etc. the time budget for keeping oneself fed is now actually dominated by eating (~1/7) and not by production (~1/20?).

So, 200 or 300 years in the future, will we have done something similar to other physiological needs? other safety needs? (note here that many aspects of keeping oneself fed in prior centuries involved long-lead time skills that are no longer necessary because we produce and distribute food in an entirely different manner)

Upon reflection, the broken glasses scenario is not so dire: distribution of new lenses (produced in bulk) according to Rx is low-enough skill that one might imagine easily finding volunteers who schlep packages for the variety of social contact it affords. The toothache would be more difficult, because there treatment (currently) requires specialised knowledge. (and fixing the lens-making machine might be easy enough if one takes the "DEC repairman" attitude of parts are cheap, time and knowledge is expensive, so just swap old parts for new until it passes the self-test again?)

Not sure if good faith is good enough to get me rate-undelimited, but here goes (:

[e.g. Maxwell's] Ratchets are part of the firmware debug, ideally a debug should be irreversible, yes?

Sorry about the TUFs, it was a lagniappe for gvicino (--assuming you've already seen my previous rant on it.) some (most?) have the form of a reverse rachet: discontinuity+monotonic. I surmise that it would be a good entity with which to model complexodynamic (economic) agents ("peers") trying to debug (cooperate with) one another.

(Each assuming an epsilon of bad faith in the other peers?)

For glasses, unbreakable or, in 30 years time, self-healing teutonic glasses.

EDIT: main problem with today's deployment finance ("VC"): it monotonically moves complexity/brittleness (eg cloud is arguably simpler than locally-served wares as a business model, but not as a technology) -- but not S-Sophistication.

The lust for funding software comes from the seduction of centralized simplification. The business of deploying atoms has no moat besides the violent one.

My attempt at drawing out the experts at

https://news.ycombinator.com/item?id=41167865

Resulted in this: https://marginalrevolution.com/marginalrevolution/2013/11/th...

  The FDA never approved it in the US, although thalidomide did make its way to the US market. It's the perfect example of seen-versus-unseen. There are probably lots of examples of benefits that unapproved drugs could provide but we don't see them, while we do see the effects of the bad drugs that the FDA did not approve. I favor looser FDA regulations, but if my side doesn't even know the history of the #1 argument against it then we should just give up.

Anyways.. in my mind, mRNA does really rhyme with anarchist-primitivism*, so y'all might want to check if FDA is supervising a market failure or not.

*Due to my bounded rationality, I shudder to discuss the technical reasons, with,uh, people who have not thought about it..

But your grandma probably knew that... AMP is a OTC supplement and GMP is a common food additive . (Compare with their d variants)

Reply on news.ycombinator.com