Thiel: Technology Stalled in the 1970's
technologyreview.com
technologyreview.com
This is somebody who watches Star Trek and figures that if they can use dilithium crystals to control anti-matter plasma conduits, well gee how hard can it be? We should get right on that.
I'll cut him some slack because he wasn't actually alive in the 70's but seriously, cure cancer? Is he not aware of the billions of dollars of investment in that problem? Did he not even realize that the cell biology research that came out of figuring out AIDs wasn't even freakin' POSSIBLE in the 70's? You couldn't do analysis on DNA in the 70's because the machines that did that, would take until the present day to give you an answer. Think about that.
Famous quote from Marvin Minsky along the lines of you want to solve a problem that will take a million hours of compute time, you can either spend a ton of cash and buy a super computer, and start it on its way and wait 114 years, or you can invest the money you have in a savings account, wait 20 years and then buy a computer at Fry's that will solve it in an hour.
We couldn't even build the tools to build the tools yet in the 70's that we regularly use to put cool features into an iPhone today.
So basically I'm offended on behalf of every engineer who lived and worked in the 70's at his comments.
And then lets take the flipside, given that we do have all of these amazing tools at our disposal, do you think people are not trying to cure cancer, in all its forms? So Peter has some genius idea that the Memorial Sloan Kettering Cancer Center's staff of doctors isn't considering?
Technology did not "stall" in the 70's. It has been accelerating at an exponential rate for decades. I can't believe he really thinks that.
Has it really been though? Let me ask. What we usually perceive as progress comes down to few things. Mobile phones. Increasing hard disc drive capacity. Increasing memory capacity. Increasing processing speeds, and recently increasing energy density of the batteries. These things enables a lot of things to go digital/realtime and makes a lot of cool stuff possible.
Mobile phones can be thought of as amazing things. But we do that by putting big towers littered around the globe. Is that really a technological breakthrough? It is a massive engineering accomplishment. But I cannot see it as a breakthrough.
The progress we have now is mostly limited to making existing things better/more efficient. We are not inventing/discovering anything groundbreaking. Like electricity. Like Electromagnetic waves. Even like Diodes and Transistors.
Fusion is still 20 years ahead. We are still driving around in vehicles powered by burning fossil fuels. We don't really know what to do about climate changes. We don't really know how to cultivate crops economically, without heavy use of pesticides and such things. We don't even have an easy way to check if the food we buy at shops contain any harmful chemicals.
So where is the technology that has been accelerating for decades? In the duel cameras of your octa core smartphone?
Non Fusion things to think about.
RF systems, think about WiFi etc. Wireless data channels with the capacity to send hundreds of megabytes per second. Computational radar systems, magnetic resonance imaging and about a half dozen different ways to image things at the atomic level.
Materials science, building airplanes out of carbon and glue. Building skyscrapers as pre-fab blocks. Sapphire glass, tempered glass 10x harder than anything in the 70s.
Vision systems and vision equipment, from cameras that have megapixels of resolution costing just a few dollars, to dedicated object tracking and identifying sensors. Lidar as a thing.
Information retrieval on a scale that was not even imagined in the 70's. Trillions of documents accessible through a simple network protocol.
Precison machining, additive manufacturing, 3D printed rocket engines.
Privately funded space technology, from micro-satellites to booster systems.
Textiles with the feel of satin and the stopping power of ceramic plate armor. Carbon based conductors made out of graphene.
Internal combustion engines that go three times further, or have the same power with 1/3 the gasoline required.
Solar power systems that can provide 100% of the needs of a single family home for less than the cost of re-roofing.
One of the interesting ways to think about it are what tools enabled what technologies. For example the ability to make hardened glass at scale gave rise to the ability to have large screen televisions. You could imagine a 55" television in the 70s but you could not cost effectively manufacture one.
Another way to think about it is how computer games today on the PS4 or the Xbox would be inconcievable on the DEC-10 or the IBM 370 in the 70's. The GPUs themselves were simply to complex to imagine, much less design, simulate, and fabricate. Get your hands on state of the art "demo reels" from the Siggraph conferences in the 70's early 80's. Compare that to Renderman. There is no comparison.
I think as Peter gets older and has participated in bringing more technologies from concept to product, he will evolve his thinking where he understands the entire stack of things that have to be in place for something to occur, this "eco-system" if you will, which has to all be present before progress can be made. You can't study diseases effectively without microscopes, you can't analyze genetics without the polymerase chain reaction and a way to sequence DNA.
Fun to try and go through your house or office and not use anything that didn't exist in 1975. But sometimes you can only see the distance you've come when you look back to where you were before.
But when I look at something that is 'modern', all I see is a highly refined/glorified version of a 50 year old thing. Mobile phones are just a glorified radio. Computers are glorified calculators. Internet is a glorified telephone network.
Sure, these things have massive impact on our existence. But that does not prevent them from being mere refinements.
So, if we figure out something like, teleportation, or anti-gravity or a way to convert arbitrary matter to energy(mc2), that would be something. An innovation/technology that will be indistinguishable from magic.
Kidding, but I hear you, the truth is that there isn't all that much that cannot be equated, at least at a high enough abstraction, to things that have been around for a long time.
Anyway, I don't see what he said as being a statement like "I, Peter Thiel, could trivially cure cancer" or anything like that.. I see it as a call to arms to people who are working on, say, ad serving networks, to work on curing cancer instead (where "curing cancer" is a variable that could be replace with any "big, hard, fundamental problem").
Imagine you're building a castle, and you've got a quarry full of people cutting the stones to build it, and then you chastise the carpenters who are working on decorative woodwork because there is not enough castle yet to work on.
Sure, Google has spent a ton of money building an ad network and figuring out how to get people to click on ads. They have also spent a ton of money on building phone OSes, and build tools, and efficient data centers, and self driving cars.
Not everyone can work on curing cancer or creating durable political institutions that survive local corruption. So what do they spend their time on? Building tools, or using those tools to build interesting things which feeds back into better tools.
I didn't see any appreciation of the foundational work that has to happen in order for progress to be made in that interview.
No, a call to arms for working on big, hard, fundamental problems would be to say, "Let's restructure the scientific career path away from a pyramidal labor and triple funding to the NSF and NIH."
According to Wikipedia, he was born on October 11, 1967.
http://stackoverflow.com/questions/432922/significant-new-in...
Our whole industry got sucked into making a fortune and now suffers (ironically) from "The Innovator's Dilemma".
We have more money than Croesus now, but at the expense of any real progress.
Second, productivity growth would normally be good if the gains from that growth were distributed. Instead, all of those gains have been placed directly with the wealthiest %1. So instead your graph, sadly, simply reflects how the quality of life has degenerated over the generations. That's not the kind of progress that 99% of the world was hoping for.
Second, how those those gains are distributed is likewise irrelevant to that particular question. Not that they're irrelevant to other questions, or that they're unimportant, or uninteresting, or that this is the only question we can ask. The careful reader will notice that neither Thiel nor I really commented on the otherwise important and interesting question of how gains are distributed. That's just something you threw in.
Third, it's not "my graph." I linked to a Wikipedia article written by other people and which contains no fewer than four graphs at the time of this writing, and none of them were created by me.