Hacker News Reader

Comment by mistermann | Hacker News Reader

These lessons seem very logically sound to me.

I wonder, if we were to think of (realize) human minds as nodes of a supercomputer, connected via wires and human "biology" (data to screen to eyeballs to cpu (brain), cpu to fingers to keyboard to central controller), and designed a system to optimize away the obvious efficiencies, what could mankind be capable of?

If we could wire in all the nodes from HN, SSC, LW, IDW, [insert communities/professions of your choice], and then pointed this "thing" at the problems we have on this planet, what might be the result?

But first, I wonder if you have to believe it's possible, if that makes any sense. I have a strange feeling that might be the hardest part.

EDIT: assume this is workable. What should be the first two problems to address?

View on HN

Curing disease. Not just the current pandemic, but cancer and everything else.

There are only 70,000 cancer researchers in the world. You could 1000x this and still only be 1% of people. 1000xing isn't likely to create 1000x the progress, but maybe you could hope for √1000x. So the progress of 31.6 years happens every year.

I wonder how much of the work those 70.000 cancer researchers do is theoretically unnecessary busywork like applying for research grants or dealing with office politics. Further I wonder how much of their research is optimized for looking reputable to the people writing the checks vs. what they think would really move the needle. Sure cancer is a complex problem but 70 thousand is already a massive amount of people so I can't shake the suspicion that instead of throwing even more people at the problem changing the environment in which they work would be much more effective.

I suspect that "helps cure cancer" is a popular touchstone for grant applications, no matter how indirect the association.

The first question I'd ask is: what if cancer is maybe too hard?

I doubt a whole bunch could be done on this covid thing, I suspect collaboration is probably going about as good as can be expected at this point.

I would propose the first project would be optimizing as much of the biological portions of the connection away, ie: design a platform that can be used for high bandwidth communication and management of projects at which this biological supercomputer can then be pointed.

Something like reddit would be an obvious component, it would require some new features here and there obviously, it works pretty well. The primary problem I see on reddit is rogue (delusional) nodes - these are not only not helpful, but incredibly harmful in my mind. A way would have to be found to see that these kinds of nodes are minimized, but my intuition tells me it would be much easier than we expect.

I think there's an impression that cancer is one disease rather than an adjective used to describe at least 70,000 different diseases (if one wants to be pedantic). Even the biggest killers of those probably have a 1,000 researchers max.

Speak of the devil, look what I found: "OpenCovid19 Initiative, we are currently running an open-source face mask challenge to identify all existing designs out there and have them evaluated by the global community and experts. The goal: Validate the good designs and send them to manufacturers for production."

Community: https://app.jogl.io/program/opencovid19 (will submit this separately in a sec)

Challenge: https://facebook.com/groups/opensourcecovid19medicalsupplies...

Database: https://docs.google.com/spreadsheets/d/1e81ceSrVT6Bl4UaS1SxI...

Very nice, initiatives like this are what we need way more of. But as per my earlier comments, I see this an example of an instance of a sub-project, that would be run inside a larger framework.

What are SSC, LW, IDW? Wait, Slate Star Codex, but the other two?

LW=LessWrong, and IDW = Intellectual Dark Web I think

Thanks!

Reply on news.ycombinator.com