- myostatin inhibitor gene therapy as an OTC cosmetic procedure -- would solve obesity
- cheaper rockets
- VR telepresence to eliminate the massive petroleum waste of the workday commute
- self-driving car (great open hardware + source project)
- "brain backup" using high-res MRI + in-depth biographical interview + DNA sample
- seasteading
- "bio lab on a chip" using microfluidics, flourescence-assisted cell sorting, nanopore sequencing, and phage-assisted directed evolution
If you read this list and say you lack the necessary skills -- bullshit. In the age of Wikipedia and Google Scholar you can learn anything you try to. The only limit is your courage.
Downvoter -- I don't normally let stuff like this get to me, but you better explain yourself.
It took me 4 years of manic polymathic learning (at the expense of my actual schoolwork) to come up with this list. I've worked in a genetic engineering lab (Drew Endy), AI, robotics, and now AR, so I speak firsthand that these ideas are all possible now.
> It took me 4 years of manic polymathic learning (at the expense of my actual schoolwork) to come up with this list.
I'm impressed that you independently came up with those ideas, but it seems like an average news day on HN. Those problems are regularly being discussed.
I think that highlights that, even in our inter-connected social media world, our biggest problem is still one of communication. You are quite right that anyone can easily gain the skills necessary to work on those projects. Maybe you can even find the capital necessary.
However, you are going to really struggle to catch up with what people already know. If you want to build a driverless car, for example, you are going to spend years reimplementing what Google has already figured out because there is no good way for the average person to build atop their existing technology. By the time you finally get to new problems, someone will have already figured them out too.
The first startup who figures out how to solve that communication gap will provide the most transformative technology to date.
Haha "come up with the list" meaning agregating the list; I'm not taking credit for first thinking of them all! The problem domains where I've thought of / sometimes written about original solutions are just these ones:
- camera drones to provide a "real-time Google Earth"
- genetically engineered supercrops for the developing world that provide protein, fiber, and every vitamin
>>> If you read this list and say you lack the necessary skills -- bullshit. In the age of Wikipedia and Google Scholar you can learn anything you try to. The only limit is your courage. <<<
I down voted you not because of the swiss cheese holes in your ideas but this sentence. Sorry but given that some of your ideas assume a lot of basic research and the devil is in the details such a remark didn't jive with me. Your ideas have been clearly based on extrapolation and well that doesn't always equate to actually implementing the system. If someone achieves the harder items on this list then I would be pleasantly surprised, but until then those items are just science fiction. I've talked about the ones that have caught my eye below.
>>> - camera drones to provide a "real-time Google Earth" <<<
You're forgetting the costs associated with maintaining a large fleet of drones, every drone takes up energy to not only stay aloft but to beam down pictures using whatever radio gear is being used at the end, drone by drone those costs will add up and the total cost of keeping an entire fleet in the air will be prohibitive. Of course one way around would be to use balloons, but then you have to make sure that drift in either scenario is as little as possible and you can correct for it. There are day-to-day variations in weather patterns and building something that can stay aloft and provide a steady picture within storms is quite challenging indeed. If you solve all of the issues then you run into the issue associated with manipulating all of that video and stitching it seamlessly to specific geographic locations. Remember your balloon etc. might be swaying so there won't be an exact match between what the video camera is recording within a certain set of pixels and what you think that area is. You'll have to geotag sets of pixels to figure out that this is the live stream of street X and so on, that's really really hard to do and imagine you have a city of these things and you're trying to do that in real-time. The sheer number crunching involved will be huge and again impractical. Then you have the optics themselves, you'll have to put up a very special camera indeed to cover a maximum possible area per drone while maintaining a usable level of resolution. Perhaps you will decide to use multiple cameras or something, but either way creating something like that within weight, costs constraints is simply difficult.
Then there's that other tiny thing, all of this needs to produce enough value to pay for itself... Given all of the costs that have been sunk in what economic use could such a thing find that couldn't be done in much more cheaper and pragmatic ways?
>>> - camera satellites for same purpose (see DARPA SEE-ME challenge) <<<
I don't know enough to comment on the practicality of this, but I do know enough to whistle at the amount of cost this will require. Either you will have to launch a satellite into geostationary orbit or launch a lot of satellites into overlapping orbits so that each area gets covered all the time... Then again there's the problem of achieving resolution in the video and having something large enough that could actually store all of the content and then beam it back to earth in real time. All of this sounds prohibitively expensive and doable only with military funding.
>>> - cheap laser scanners for self-driving cars and robots <<<
Agreed. Personally I think that sooner or later people will start forgoing laser scanning and use something cheaper such as an array of webcams to do hyper-stereopsis and then interpolate that with other data to get a good enough idea of the world.
>>> - genetically engineered supercrops for the developing world that provide protein, fiber, and every vitamin <<<
That's a nice goal and the world needs it, but aside from the difficulty of creating the crop there are tons of pesky real world issues that need to be overcome including the fact that few countries would allow such a crop to grow at all it would be probably banned until further "testing" which would again inflate the cost dramatically. There are lots of real world schleps over here, which can be overcome but will necessitate significant backing.
I think that solving the problem of multi-storey stacked farms might be a better bet. (but it will be really challenging as well)
>>> - "brain backup" using high-res MRI + in-depth biographical interview + DNA sample <<<
Given how little we know about intelligence and the way the brain truly functions and that this is still in the basic research stage, I would be shocked if anyone could create a "brain backup" from just these 3 datasets. There are several non-trivial challenges that need to be resolved before doing anything like this. How do we know that what we relate as consciousness isn't implicitly a certain configuration of synapses in a very specific manner that is influenced by environmental factors as well? Clearly DNA alone does not explain everything, you have several epigenetic factors within the womb that makes the person who s/he is. Just the timing of a few series of hormones at specific intervals can radically alter the end result. Then outside the womb you have day to day factors which take this really complex system and push it into oh so many directions. Each factor leaving a mark. How can you capture that with just an MRI, an interview and a DNA sample? Within the next 5 to 150 years we might develop some hypothetical technique that allows us to do this, but clearly until then this is unpractical.
>>> - seasteading <<<
Energy. Aside from the resource issues related to creating a seastead there are tons of issues stemming from energy, currently the plan is to just boat in diesel, but to create a truly independent and self sustaining system there needs to be a way to capture and efficiently store energy on the seastead itself. That is much harder than it looks and it will take some time to make that possible. This will get done sooner or later, but this tiny detail will bog down a lot of progress due to the usual mix of costs, quality, and last minute issues.
---
Most people's view of the future is off because of the tiny details and the things that they tend to overlook. We all tend to do that and forget the real world is a harsh master capable of bringing down the wrath of economics as well. Some of these things ought to be implemented, but whether or not they actually are at the end of the day is another matter.
I don't mean to deflate you or anything, but the truth is that you need to be cognisant of the challenges and do a reality check if you ever want to implement something like this at all. A lot of these systems are possible technologically, but actually doing them is another thing. I hope that you go on to achieve this list. It would be amazing.
>>> That ways to learn line should include getting subject matter experts to explain their field to you. E.g., many professors reply to a cold email. <<<
I wasn't actually aware of that. I'll keep that in mind. Thanks for the heads up. :)
>>> And yes, I'm only 22 and filled with naive optimism :-). <<<
I'm 20... I realised at some point that optimism doesn't always need to be naive. To get anything done you clearly need a mix of just the right amount of naiveness and soul crushing insight.
>>> But I'm channeling that energy into doing an AR glasses startup instead of some web thing. <<<
That's great to hear. :)
>>> I'll help the world with my research even if I have to spend my every penny doing so, and I wish more people would attempt the same. <<<
That's even better. :D Trust me you aren't the only person who wants to do so.
- camera drones to provide a "real-time Google Earth"
- camera satellites for same purpose (see DARPA SEE-ME challenge)
- cheap laser scanners for self-driving cars and robots
- synthetic meat
- genetically engineered supercrops for the developing world that provide protein, fiber, and every vitamin
- AR / VR glasses (shameless plug: http://vergencelabs.com)
- subvocalization microphone
- myostatin inhibitor gene therapy as an OTC cosmetic procedure -- would solve obesity
- cheaper rockets
- VR telepresence to eliminate the massive petroleum waste of the workday commute
- self-driving car (great open hardware + source project)
- "brain backup" using high-res MRI + in-depth biographical interview + DNA sample
- seasteading
- "bio lab on a chip" using microfluidics, flourescence-assisted cell sorting, nanopore sequencing, and phage-assisted directed evolution
If you read this list and say you lack the necessary skills -- bullshit. In the age of Wikipedia and Google Scholar you can learn anything you try to. The only limit is your courage.