Look for the Next Tech Gold Rush?

34 points

expand_less expand_more

Jul 19, 2014

In early 2000, I registered my personal domain name weidai.com, along with a couple others, because I was worried that the small (sole-proprietor) ISP I was using would go out of business one day and break all the links on the web to the articles and software that I had published on my "home page" under its domain. Several years ago I started getting offers, asking me to sell the domain, and now they're coming in almost every day. A couple of days ago I saw the first six figure offer ($100,000).

In early 2009, someone named Satoshi Nakamoto emailed me personally with an announcement that he had published version 0.1 of Bitcoin. I didn't pay much attention at the time (I was more interested in Less Wrong than Cypherpunks at that point), but then in early 2011 I saw a LW article about Bitcoin, which prompted me to start mining it. I wrote at the time, "thanks to the discussion you started, I bought a Radeon 5870 and started mining myself, since it looks likely that I can at least break even on the cost of the card." That approximately $200 investment (plus maybe another $100 in electricity) is also worth around six figures today.

Clearly, technological advances can sometimes create gold rush-like situations (i.e., first-come-first-serve opportunities to make truly extraordinary returns with minimal effort or qualifications). And it's possible to stumble into them without even trying. Which makes me think, maybe we should be trying? I mean, if only I had been looking for possible gold rushes, I could have registered a hundred domain names optimized for potential future value, rather than the few that I happened to personally need. Or I could have started mining Bitcoins a couple of years earlier and be a thousand times richer.

I wish I was already an experienced gold rush spotter, so I could explain how best to do it, but as indicated above, I participated in the ones that I did more or less by luck. Perhaps the first step is just to keep one's eyes open, and to keep in mind that tech-related gold rushes do happen from time to time and they are not impossibly difficult to find. What other ideas do people have? Are there other past examples of tech gold rushes besides the two that I mentioned? What might be some promising fields to look for them in the future?

34 points

expand_less expand_more

Comments

sorted by

Highlighting new comments since Today at 5:53 PM

All comments loaded

15 points

expand_less expand_more

While it is true that we (techies, rationalists, etc.) have the opportunity to catch a gold rush by becoming early adopterst, I suspect survivorship bias is at play. There are plenty of people who try to systematically 'grind' on such opportunitities but it doesn't pan out for many of them - I know some people who used to mass-register domains, and made a neglible profit in the end, people who jump on all sorts of altcoins, programmers who join a promising startup, where they sacrifice salary for equity, etc. Additionally, I also started messing with bitcoins in 2011, and while it has been quite profitable, I have made less than six figures, since I wasn't very serious about it at the time. And yes, in retrospect I can say that I should've put more money in (and kept them in bitcoin), but if I follow the same line of reasoning with all the seemingly-promising things I see, I might very well go broke.

I wish I was already an experienced gold rush spotter, so I could explain how best to do it, but as indicated above, I participated in the ones that I did more or less by luck.

Indeed, luck seems to be a big part of it, and the main action that you can take to facilitate the process is probably to put yourself in the right circles, so you can hear and look into innovations early on. This, however, is something that you and many people on here already do, and I doubt that you can easily find another intervention that will have as big of an impact on your chances to participate in a gold rush.

15 points

expand_less expand_more

6 points

expand_less expand_more

if I follow the same line of reasoning with all the seemingly-promising things I see, I might very well go broke.

Nobody has come up with another example of something that meets my definition of a tech gold rush, so these things don't seem to happen very often. I suggest keeping an eye out for them, but set your standard of "seemingly-promising" high enough so that it's triggered only rarely.

Another useful tip may be to look for entirely new asset classes, rather than something similar to the last big thing (altcoins or startups). Before domain names and Bitcoin, who thought that entries in a lookup table or a decentralized currency could be an investment asset?

6 points

expand_less expand_more

Mark_Friedenbach

3 points

expand_less expand_more

I would have thought you did, based on your research.

What information available in 2009 would have convinced you that bitcoin would become a billion-dollar market as quickly as it did?

3 points

expand_less expand_more

2 points

expand_less expand_more

I would have thought you did, based on your research.

No, my monetary policy views were firmly mainstream, which considers rapid unpredictable changes in prices, in either direction, to be a really bad thing for a currency. So I designed b-money to have a stable value relative to a basket of commodities, and until Bitcoin came along, never thought anyone might deliberately design a currency to have a fixed total supply.

2 points

expand_less expand_more

10 points

expand_less expand_more

It's weird to think that there are assets available, right now, which are dirt cheap but will be extremely valuable in 5 or 10 years.

Anyway, I've given this issue a decent amount of thought since it's reasonable to expect that advances in AI will cause a lot of economic disruption and there is a decent chance that people fortunate enough to be invested in the right assets will end up very well off. What will be scarce but desirable in a post-AI world? Land? Rare art? Taxi medallions?

The main idea I have had is to save money by investing in an index fund on the theory that it gives you a pretty good chances of hitting one or more companies that would benefit tremendously from advances in AI.

10 points

expand_less expand_more

9 points

expand_less expand_more

I thought about it for a while and wound up concluding that the Bitcoin thing probably happened because there was a barrier to entry that kept out the majority of people. Back in 2010 or 2011, you had to send cash to places like Dwolla to actually get BTC. It sounds stupid, but I suspect that if you had been able to (e.g.) invest in a Bitcoin ETF back then, Bitcoin would have appreciated in price fast enough that it would have been hard to get in on the rush.*

So the lesson I'm taking away from that one is to beware of trivial inconveniences.

*Of course, this presumes that the "rush" is over, but at least it's a lot harder for me to be sure that buying Bitcoin is +EV at $500 than at $1.

9 points

expand_less expand_more

15 points

expand_less expand_more

So the lesson I'm taking away from that one is to beware of trivial inconveniences.

I thought the lesson was look for trivial inconveniences as indicative of potential opportunities.

15 points

expand_less expand_more

4 points

expand_less expand_more

Sure, you can look at it that way. From my perspective, I think the lower-hanging fruit is to make sure that when I have a good idea and I don't do it because of a trivial inconvenience, I notice and do it anyway.

4 points

expand_less expand_more

4 points

expand_less expand_more

it's a lot harder for me to be sure that buying Bitcoin is +EV at $500 than at $1.

Serious question: how much analysis went into that?

I can come up with cases where the price goes down (massive regulations / bans result in no one using it), and cases where it goes a lot higher, but I'm having trouble coming up with scenarios where the price holds roughly steady (say, between $100 and $3000).

4 points

expand_less expand_more

3 points

expand_less expand_more

So the lesson I'm taking away from that one is to beware of trivial inconveniences.

Yeah, I knew about Bitcoin in 2010, thought to myself "I really ought to put $1k into that in case it explodes," and didn't because of the inconvenience. Oops.

3 points

expand_less expand_more

John_Maxwell_IV

6 points

expand_less expand_more

Would you purchase weidai.com for six figure if it were for sale today? Is having the same domain on a continuous basis extremely important to you? If not you might be suffering from status quo bias.

6 points

expand_less expand_more

10 points

expand_less expand_more

If not you might be suffering from status quo bias.

Yep, that has occurred to me. Here are my excuses for not doing anything about it. (And yes, I'm aware that these could just be rationalizations for my bias. :)

My main email addresses are on weidai.com. It would take a lot of time to inform everyone and change my registrations on various shopping sites, etc., and I'll probably still lose a bunch of emails.
Lots of links to my site from other sites, which I can't change easily or at all. This wouldn't be the case if I was considering buying weidai.com today.
weidai.com has historically produced very high returns. Selling it and putting the money into index funds might still be an improvement from a portfolio theory perspective, but I don't think the benefits are that high (given that I already have an otherwise well-diversified investment portfolio).
I do have an emotional attachment to the domain now that I've owned it for so many years, so holding onto it could be considered a real preference, not just a bias.
If it is a bias, overcoming it takes effort, which might be more productively directed towards other goals, like writing a LW post about looking for the next tech gold rush, or thinking about meta-philosophy.

10 points

expand_less expand_more

8 points

expand_less expand_more

It would take a lot of time to inform everyone and change my registrations on various shopping sites, etc., and I'll probably still lose a bunch of emails.

If they're willing to pay $100k, they're willing to forward emails for a few years.

Lots of links to my site from other sites, which I can't change easily or at all.

Linkrot is high enough on the Internet that in a few years, all the damage will be repaired. You also have few enough pages you may be able to arrange redirects for individual pages. (I've read through everything on the site - it's not that big.)

weidai.com has historically produced very high returns.

No, having a site has historically produced very high returns. Not having the exact domain name. (Unless you mean something like 'I made a fortune off someone who told me they randomly typed in the exact domain name 'weidai.com' and decided to name me in their will'.)

I do have an emotional attachment to the domain now that I've owned it for so many years

Can't argue with this one. On the other hand, I'd be thrilled to sell gwern.net for $100k; money makes an excellent salve for hurt feelings.

If it is a bias, overcoming it takes effort, which might be more productively directed towards other goals, like writing a LW post about looking for the next tech gold rush

How long does it take to sell a domain name and set up shop on weidai.net or whatever? Let's say 3 or 4 weeks of nonstop work. You really think you are currently producing ~$100k of value with your current lifestyle and work patterns each 4 weeks in which you ignore selling the domain?

8 points

expand_less expand_more

2 points

expand_less expand_more

No, having a site has historically produced very high returns. Not having the exact domain name.

Huh? I registered the domain for <$10, last year it was worth $50,000, and now it's worth $100k. How is that not "historically produced very high returns" just from having the exact domain name? Are you thinking of "return" as including just income and not appreciation?

You really think you are currently producing ~$100k of value with your current lifestyle and work patterns each 4 weeks in which you ignore selling the domain?

By not selling the domain, I still own it, and it's still worth $100k. The effect of selling it is just to increase the diversification of my investment portfolio (i.e., I would no longer have a substantial chunk of my investments in a single asset), which is a good thing since I'm risk averse, but that benefit is certainly not worth $100k.

(Actually, thinking about it more, I'm not sure that selling the domain is actually good for diversification. Domain names are a different asset class from everything else I own, so ideally I ought to have some exposure to it.)

2 points

expand_less expand_more

7 points

expand_less expand_more

I registered the domain for <$10, last year it was worth $50,000, and now it's worth $100k.

Oh, you meant the domain name itself as an asset, not the use you were getting out of it. Yes, I suppose it has on paper, but past performance is no guarantee of future returns. There's no reason to hang onto it. (How much do you know about domain names? Will additional TLD releases by ICANN reduce the value of weidai.com? Or is the reason it's worth $100k time-sensitive in any way? It seems like a boring domain name, why is it worth $100k anyway?)

The effect of selling it is just to increase the diversification of my investment portfolio (i.e., I would no longer have a substantial chunk of my investments in a single asset), which is a good thing since I'm risk averse, but that benefit is certainly not worth $100k.

You're taking on a huge amount of uncompensated risk here. Would you put $100k into a single stock? For example, buy a single share of BRK.A? No, of course not, even if that got you exposure to 'insurance' or 'railroads'.

Domain names are a different asset class from everything else I own, so ideally I ought to have some exposure to it.

Unless you're orders of magnitude wealthier than I think you are, your exposure to an opaque illiquid minor hard-to-price asset like domain names should look more like $10 than $100,000.

7 points

expand_less expand_more

2 points

expand_less expand_more

past performance is no guarantee of future returns

Technically, past performance is Bayesian information about future returns, which in an efficient market is already reflected in asset prices. The domain name market is not an efficient one though, given that I'm the only person who can even see the history of bids on weidai.com. But my main point was just that the domain produces returns like any other asset, so the financial cost of holding onto it can't be the whole $100k, which is what you were assuming. I don't know what the actual cost is. My intuitive estimate is that it's about $10k, which does not make it worthwhile for me to sell. I don't know how to do a more exact calculation. Do you?

Unless you're orders of magnitude wealthier than I think you are, your exposure to an opaque illiquid minor hard-to-price asset like domain names should look more like $10 than $100,000.

"opaque illiquid minor hard-to-price" are arguments against trading in the asset, either buying or selling, since these attributes tend to increase transaction costs. I don't see how they are arguments for keeping my holdings in the asset class to a low level, if I started off holding a big position.

2 points

expand_less expand_more

3 points

expand_less expand_more

The domain name market is not an efficient one though, given that I'm the only person who can even see the history of bids on weidai.com.

It may not be particularly efficient, although it's grown up a lot since the '90s. But being inefficient is not helpful for you, since you are not an expert on domain names and have no edge. As far as you are concerned, the domain name market is efficient. As I said: do you know how serious the offers are? Do you know why exactly weidai.com may be worth $100k? Do you know whether it's likely to continue increasing and what the limit is ($200k? $1m? $5m?) for it? Do you know whether additional TLDs would affect it (wei.dai would be a nice domain...) or whether the use for weidai.com would be affected by any increases in adoptions of Unicode or punycode domain names? If you don't know any of this, how on earth can you sit by and leave up to $100k of your money in such an asset? Such complacency baffles me.

My intuitive estimate is that it's about $10k, which does not make it worthwhile for me to sell. I don't know how to do a more exact calculation. Do you?

No, but my intuition (as a person with no sentimental attachment to the domain and not seeking excuses to not sell) is that the risk and opportunity cost are much larger than $10k. You have a bird in the hand, which you've never sold, don't know why it's valuable, and can easily replace. I would fling that away from myself like it was 2000 and I was holding $100k of Pets.com stock.

"opaque illiquid minor hard-to-price" are arguments against trading in the asset, either buying or selling, since these attributes tend to increase transaction costs. I don't see how they are arguments for keeping my holdings in the asset class to a low level, if I started off holding a big position.

All those attributes make it a very volatile and risky asset to hold, so by regular portfolio theory, you should be holding very little of that asset and in particular, should be rebalancing away from it now that it's recently doubled.

3 points

expand_less expand_more

4 points

expand_less expand_more

Such complacency baffles me.

Part of it, which perhaps you and most other observers are not aware, is that I have enough passive income, and enough dispassion for conventional status signaling, that my marginal utility of money is pretty low compared to my disutility for doing busywork. To put it in perspective, I quit my last regular job in 2002, and stopped doing consulting for that company as well (at $100/hour) a year later when they merged with Microsoft and told me I had to do a bunch of paperwork and be hired by Microsoft's "independent consulting company" in order to continue.

The reason for writing this post was that there seems to be opportunities "out there" for earning up to hundreds of millions of dollars (like the opportunity to mine Bitcoin at version 0.1 that I narrowly missed) while doing very little work. In comparison, doing busywork for a month to earn some unknown amount of money between $0 and $100k is not particularly motivating to me at this point.

4 points

expand_less expand_more

Mark_Friedenbach

2 points

expand_less expand_more

But my main point was just that the domain produces returns like any other asset, so the financial cost of holding onto it can't be the whole $100k, which is what you were assuming.

$100k is the opportunity cost, which is very real. You could be right now re-investing that $100k in some other venture.

2 points

expand_less expand_more

2 points

expand_less expand_more

I think you may have misunderstood what I was saying. Gwern had asked me, "You really think you are currently producing ~$100k of value with your current lifestyle and work patterns each 4 weeks in which you ignore selling the domain?" And I was trying to point out that by doing 4 weeks of work, I'm not earning $100k, but rather earning the difference in future returns between holding weidai.com and holding some other asset, presumably an index fund, and this difference has to be worth less than 100k in current value unless there's 100% chance that weidai.com becomes worthless.

2 points

expand_less expand_more

3 points

expand_less expand_more

Well, the good gold rushes are rare.

I think it's worth pointing out that the Bitcoin gold rush is probably far from over. It's by no means certain to continue, of course, but I think there are reasons to question how strongly the Efficient Market Hypothesis applies. In particular, the current regulatory murkiness that makes it difficult for institutional money to get involved, but there are others as well.

3 points

expand_less expand_more

2 points

expand_less expand_more

$100,000 is a very small amount of money for smart people in finance, certainly not something they would spend a lot of time trying to get. It's certainly possible that if you are very smart, have a bit of specialized information on some topic, have smart associates willing to give you advice, and are willing to devote, say, 100 hours you could find a $100k bill on the ground.

2 points

expand_less expand_more

8 points

expand_less expand_more

I think you make a good point, however I'm really looking for >$1 million bills on the ground. The examples I gave in the post weren't primarily meant to be examples of success at picking up $100k bills on the ground, but rather examples of opportunities for >$1 million missed because I wasn't paying attention. For example, when I was registering for weidai.com, why didn't I stop to think, "the Internet will surely be big in China eventually, so let me just register all the available two syllable pinyin domain names while I'm at this"?

8 points

expand_less expand_more

2 points

expand_less expand_more

I looked at the Bitcoin whitepaper, thought "this is new math", and put some spare cash on it. Unfortunately spare cash at the time was close to zero because I was living paycheck to paycheck on close to minimum wage. Oh well.

2 points

expand_less expand_more

2 points

expand_less expand_more

It's an uphill battle for an individual to catch on to these advances. I try to focus on two strategies, one of which you've mentioned which is to just "be aware" by reading, following twitter feeds etc. awareness of whats out there is a starting block. once you become aware of something you need to evaluate its potential for success. Ex. read an article on Bitcoin in 2010, is Bitcoin something that will be valuable going forward? Institutions can manipulate value through repression, over-buying, over-selling etc. Individuals can increase their odds through a form of higher-level thinking. Ex. relating a product to human values and attempting to place it on a futuristic timeline. Now, for an actionable example...

3D body imaging - lots of focus and investment around 3D printing to create concrete objects, however I am more excited with the scanning portion and the virtual representation possibilities of this technology. See Google's Tango project, or PrimeSense out of Israel. Now imagine the possibilities of having a 3D representation of your body and potentially the world around you. you could easily:

Track your BMI and your physical features (athletics, vanity, health) Try on clothing virtually (fashion, performance) Use your own body for online gaming avatar Virtual dating The list goes on and on..

Why would the above be successful? Because it decreases the cost of being inaccurate. (time, money, morale)

Now think about the value of the data to corporations that would exist from having information on everyones body size and type, the value would then compound the more they use the product to purchase new clothes, products, etc.

$$$$$

2 points

expand_less expand_more