4 ms·

Welcome to the Dot Com Bubble 2.0

25 points by jkush 20y ago

jkush 20y ago
From the blog post: Seven years later, we're now clearly in the throes of another dot-com bubble. You might argue that the new bubble has been in effect since mid-2006, but the signs are absolutely unmistakable now. The job market for software developers is every bit as hyper-competitive as it was in 1999. The idea that you can found a company on the internet-- and make money-- is taken seriously now. There's a new one every week.
- sabat 20y ago
  ... and this describes a bubble, how? This describes a boom.
ido 20y ago
I love this quote from one of the comments: When life gives you bubbles - take a bubble bath
mojuba 20y ago
This second bubble was made possible by better web programming tools and languages -- everyone becomes a PHP/MySQL programmer nowadays. These tools weren't that affordable in 1990's. Technology made many things easier, like copying books and music, automating everyone's tasks, etc. But it also made programming itself a lot easier. Programming improves everyone's lives and it improves itself too. So bubbles are inevitable (as well as inevitable are some oscillations).
- omouse 20y ago
  everyone becomes a PHP/MySQL programmer nowadays. These tools weren't that affordable in 1990's. They're free!!! And they're widely available now which is why everyone becomes a PHP/MySQL programmer.
  - mojuba 20y ago
    Sorry, I didn't mean "affordable" exactly in terms of money. The tools became much easier to use, which means a better learning curve, which means saving programmer's time, which means money.
- lupin_sansei 20y ago
  Back in 1997 when I started web programming perl was free, then around 1998 mysql and PHP started being included in cheap hosting accounts.
sabat 20y ago
I doubt that the guy who wrote this understands what "bubble" means. I think he means "boom", and booms are not fragile -- not like bubbles. Truth is, this guy needs to read PG's essays.
- mojuba 20y ago
  That first bubble was a "boom" until it bursted. And what exactly he's supposed to learn from PG's essays?
  - nostrademons 20y ago
    I dunno...I think the first bubble was a boom through 95, 96, and 97. Netscape, Yahoo, Amazon, Craigslist, EBay all had real products that made money and changed how people did business. It didn't get to be bubblicious until about 1998, when people started getting venture funding for pet stores. It was right around 97 that people started crying "bubble!", and I think that we're at about the same point in the economic cycle that we were then. Tail end of the midcycle phase. Assuming the housing bust doesn't tank the economy, the real bubble years are still ahead. If we are in the midcycle phase, it's still a good time to start a company. Capital's not flowing freely enough that anyone can get funded, so those who bootstrap good ideas now stand to profit handsomely when money managers seek to pull all their capital out of the failing real estate market and into the stock and startup market. It's not time to worry until VCs start giving out $15M to any idiot with a business plan. You don't want to be that idiot.
  - sabat 20y ago
    nostrademons gave a fine answer to this. The fine point here is: there is a difference between a boom and a bubble. Boom == increase in activity supported by actual innovation. Bubble == increase in activity based on marketing fluff or other unsustainable models. PG talks quite a lot about this kind of thing; in particular, he points out that a boom can sustain for as long as there is innovation (wealth-creation) to support it, and that there is no reason the economy can't support, say, 1000 people working on startups rather than 1000 people working for IBM.
danteembermage 20y ago
It was actually very easy to justify the valuations of many (but not all) internet companies in the first "bubble" although not by naive net present value calculations. The key is to model scaling up as a call option on a future larger version of the firm; you only get the larger version if your company ends up valuable. With a standard checking account type model the only ways to increase value are to increase cash flows or decrease your discount rate, but with an option other avenues become available. For example, the lower the strike price of the option, the more valuable. So Webvan = enormous initial outlay = high strike price = bad, whereas two college kids in the Yscraper = low strike = good. Another way to increase the value of an option, and I think this is the critical one for the first boom, is to increase the uncertainty of the value of the underlying asset i.e. the hypothetical larger version of the company. In the boom, that uncertainty was huge, we will all be buying pet food online next year, will Google manage all commerce of any kind in five, etc. So, interestingly, the very fact the wisdom of the crowd had no clue what Web 1.0 business should be worth may have driven the high valuations in the first place. In the second boom it appears 1. Strikes are lower, 2. Uncertainty is lower = affect on valuations is ambiguous. That's basically the conclusion of article as well as I see it, just approached from a different viewpoint.
jsmcgd 20y ago
A bubble can only be identified definitively in hindsight i.e. after it has burst. If it doesn't burst then its just industry growth. I think the lessons were learnt the first time around and because the stakes are now comparatively very low the chances we are now currently in a bubble seem remote.
- mojuba 20y ago
  Still, there was one clear and strong sign of a bubble before it collapsed: most public dot-coms gave zero dividends, or actually a tiny symbolic sum close to zero. That was a signal for any smart investor to make money now and run away.
  - dpapathanasiou 20y ago
    Dividends don't mean that much: Microsoft in its dominant years (when they had 40%-plus margins and were raking in the cash) never paid a dividend (they did so only recently).
    - mojuba 20y ago
      As far as I know Microsoft always did pay dividends, always, and never stopped doing that. Can be checked easily.
      - dpapathanasiou 20y ago
        No, the first time MS ever paid a dividend was in 2003: http://seattlepi.nwsource.com/business/104643_microsoft17.shtml http://seattlepi.nwsource.com/business/104643_microsoft17.shtml Not paying a dividend is not unusual in the tech industry.
        mojuba 20y ago
        I was wrong, sorry.
  - nostrademons 20y ago
Tichy 20y ago
Maybe it wasn't so much the business ideas as the greed that made the bubble burst. Most bubblers just wasted so much money on consultants etc., which really wasn't necessary. Since I heard this time around people are more careful, maybe this bubble won't burst. Besides, I remember Ray Kurzweill writing that the Bubble wasn't really a bubble. The companies that really created value created so much value that they compensated for the ones that went bankrupt.
- jey 20y ago
  Besides, I remember Ray Kurzweill [sic] writing that the Bubble wasn't really a bubble. The companies that really created value created so much value that they compensated for the ones that went bankrupt. Sure, overall the whole Internet thing has created a lot of actual value, and this is evidenced by the fact that the recession that followed was pretty mild. But it's still true that it was a bubble in the sense that companies were being invested in with little to no actual value, with unworkable business schemes, and a "hemorrhage money now, make billions later!" mentality. The perception was that the Internet was a money-generating panacea, all you had to do was open a site with a .com suffix, like RoofShingles2U.com, and you'd automatically put all roofing salesmen out of business and corner the roof shingle market.
mojuba 20y ago
After all, the guy coined a new term - Bubble 2.0, and I liked it very much.
create_account 20y ago
Shh... don't tell all the people applying to YC!

25 points by jkush 20y ago

jkush 20y ago

From the blog post: Seven years later, we're now clearly in the throes of another dot-com bubble. You might argue that the new bubble has been in effect since mid-2006, but the signs are absolutely unmistakable now. The job market for software developers is every bit as hyper-competitive as it was in 1999. The idea that you can found a company on the internet-- and make money-- is taken seriously now. There's a new one every week.

sabat 20y ago
... and this describes a bubble, how? This describes a boom.

ido 20y ago

I love this quote from one of the comments: When life gives you bubbles - take a bubble bath

mojuba 20y ago

This second bubble was made possible by better web programming tools and languages -- everyone becomes a PHP/MySQL programmer nowadays. These tools weren't that affordable in 1990's. Technology made many things easier, like copying books and music, automating everyone's tasks, etc. But it also made programming itself a lot easier. Programming improves everyone's lives and it improves itself too. So bubbles are inevitable (as well as inevitable are some oscillations).

omouse 20y ago
everyone becomes a PHP/MySQL programmer nowadays. These tools weren't that affordable in 1990's. They're free!!! And they're widely available now which is why everyone becomes a PHP/MySQL programmer.
- mojuba 20y ago
  Sorry, I didn't mean "affordable" exactly in terms of money. The tools became much easier to use, which means a better learning curve, which means saving programmer's time, which means money.
lupin_sansei 20y ago
Back in 1997 when I started web programming perl was free, then around 1998 mysql and PHP started being included in cheap hosting accounts.

sabat 20y ago

I doubt that the guy who wrote this understands what "bubble" means. I think he means "boom", and booms are not fragile -- not like bubbles. Truth is, this guy needs to read PG's essays.

mojuba 20y ago
That first bubble was a "boom" until it bursted. And what exactly he's supposed to learn from PG's essays?
- nostrademons 20y ago
  I dunno...I think the first bubble was a boom through 95, 96, and 97. Netscape, Yahoo, Amazon, Craigslist, EBay all had real products that made money and changed how people did business. It didn't get to be bubblicious until about 1998, when people started getting venture funding for pet stores. It was right around 97 that people started crying "bubble!", and I think that we're at about the same point in the economic cycle that we were then. Tail end of the midcycle phase. Assuming the housing bust doesn't tank the economy, the real bubble years are still ahead. If we are in the midcycle phase, it's still a good time to start a company. Capital's not flowing freely enough that anyone can get funded, so those who bootstrap good ideas now stand to profit handsomely when money managers seek to pull all their capital out of the failing real estate market and into the stock and startup market. It's not time to worry until VCs start giving out $15M to any idiot with a business plan. You don't want to be that idiot.
- sabat 20y ago
  nostrademons gave a fine answer to this. The fine point here is: there is a difference between a boom and a bubble. Boom == increase in activity supported by actual innovation. Bubble == increase in activity based on marketing fluff or other unsustainable models. PG talks quite a lot about this kind of thing; in particular, he points out that a boom can sustain for as long as there is innovation (wealth-creation) to support it, and that there is no reason the economy can't support, say, 1000 people working on startups rather than 1000 people working for IBM.

danteembermage 20y ago

It was actually very easy to justify the valuations of many (but not all) internet companies in the first "bubble" although not by naive net present value calculations. The key is to model scaling up as a call option on a future larger version of the firm; you only get the larger version if your company ends up valuable. With a standard checking account type model the only ways to increase value are to increase cash flows or decrease your discount rate, but with an option other avenues become available. For example, the lower the strike price of the option, the more valuable. So Webvan = enormous initial outlay = high strike price = bad, whereas two college kids in the Yscraper = low strike = good. Another way to increase the value of an option, and I think this is the critical one for the first boom, is to increase the uncertainty of the value of the underlying asset i.e. the hypothetical larger version of the company. In the boom, that uncertainty was huge, we will all be buying pet food online next year, will Google manage all commerce of any kind in five, etc. So, interestingly, the very fact the wisdom of the crowd had no clue what Web 1.0 business should be worth may have driven the high valuations in the first place. In the second boom it appears 1. Strikes are lower, 2. Uncertainty is lower = affect on valuations is ambiguous. That's basically the conclusion of article as well as I see it, just approached from a different viewpoint.

jsmcgd 20y ago

A bubble can only be identified definitively in hindsight i.e. after it has burst. If it doesn't burst then its just industry growth. I think the lessons were learnt the first time around and because the stakes are now comparatively very low the chances we are now currently in a bubble seem remote.

mojuba 20y ago
Still, there was one clear and strong sign of a bubble before it collapsed: most public dot-coms gave zero dividends, or actually a tiny symbolic sum close to zero. That was a signal for any smart investor to make money now and run away.
- dpapathanasiou 20y ago
  Dividends don't mean that much: Microsoft in its dominant years (when they had 40%-plus margins and were raking in the cash) never paid a dividend (they did so only recently).
  - mojuba 20y ago
    As far as I know Microsoft always did pay dividends, always, and never stopped doing that. Can be checked easily.
    - dpapathanasiou 20y ago
      No, the first time MS ever paid a dividend was in 2003: http://seattlepi.nwsource.com/business/104643_microsoft17.shtml http://seattlepi.nwsource.com/business/104643_microsoft17.shtml Not paying a dividend is not unusual in the tech industry.
      - mojuba 20y ago
        I was wrong, sorry.
- nostrademons 20y ago

Tichy 20y ago

Maybe it wasn't so much the business ideas as the greed that made the bubble burst. Most bubblers just wasted so much money on consultants etc., which really wasn't necessary. Since I heard this time around people are more careful, maybe this bubble won't burst. Besides, I remember Ray Kurzweill writing that the Bubble wasn't really a bubble. The companies that really created value created so much value that they compensated for the ones that went bankrupt.

jey 20y ago
Besides, I remember Ray Kurzweill [sic] writing that the Bubble wasn't really a bubble. The companies that really created value created so much value that they compensated for the ones that went bankrupt. Sure, overall the whole Internet thing has created a lot of actual value, and this is evidenced by the fact that the recession that followed was pretty mild. But it's still true that it was a bubble in the sense that companies were being invested in with little to no actual value, with unworkable business schemes, and a "hemorrhage money now, make billions later!" mentality. The perception was that the Internet was a money-generating panacea, all you had to do was open a site with a .com suffix, like RoofShingles2U.com, and you'd automatically put all roofing salesmen out of business and corner the roof shingle market.

mojuba 20y ago

After all, the guy coined a new term - Bubble 2.0, and I liked it very much.

create_account 20y ago

Shh... don't tell all the people applying to YC!