Hacker News new | past | comments | ask | show | jobs | submit login
The Twilight of Venture Capital (twilightofventurecapital.blogspot.com)
6 points by toby on Jan 9, 2009 | hide | past | favorite | 2 comments



"Smaller new tools emerge around the fringe of the market every day."

Isn't that how all new tools emerge?

I'm not going to disagree with the main premise which is that real innovation on the scale of the invention of the transistor are slowing significantly. Judy Estrin's book details this trend painfully, and lays down good evidence that it started around the 80s.

That said, I think it's useful us for to pin down what revolutionary "new tools" are, and where they come from, yes? I feel that this is an important nitpick because it is hard to say where "new tools" come from. They do certainly seem to loosely and randomly come from academia (see http://en.wikipedia.org/wiki/Transistor, http://en.wikipedia.org/wiki/History_of_molecular_biology) and sometimes from tinkering (http://en.wikipedia.org/wiki/Automobile). However, these advances were always on the fringes.

What will the next jump be? Quantum computing maybe? It seems neglected and pushed to the fringes now so it's quite possible that that will indeed be it. After all, it seems that it's hard for people to seriously innovate in the limelight.


No new tools will emerge that make capital creation possible by small players?

That's a very hubris-tic thinking. I can already think of at least one that emerged recently - unprecedented parallel computation power. "Recently" here meaning after google IPO: virtualization, out of the box distributed computing, out of the box data mining/data analysis tools, on-demand datacenter capacity, cheap multi-core CPUs.

This "tool" just needs a "killer app" that would make this attractive to consumers (all these technologies are already attractive to businesses and are being adopted at enormous rate): I can already think of big players providing desktop virtualization, one YC funded start-up providing a virtualization based "Windows app store".

Here's another "tool" related to this theme: kernels and networking stacks are able to handle increasingly higher and higher number of simultaneous connections (here's a thought: read up on the epoll Linux system call and try and see if Meebo would have been able to start without a significant up-front hardware investment if such a facility was not available).

The only reason I mentioned those "tools" is that it's my specific interest: distributed computing. I see no reason to think there aren't other such tools. For the 2005-2008 "Web 2.0" boom, for example these tools were AJAX, Flash and broadband connections -- allow web applications to not only feel and act like desktop applications, but also do what only desktop applications once could, e.g. edit and play video or edit/mark-up photos in full size and colour. Would youtube or Flickr (both of which brought money to their investors) have taken off if there wasn't wide-spread adoption of cable and DSL?

Also I fail to see how "cleantech" (as much as great deal of it is buzzwords) can't create any new such tools: what about low power, low cost CPUs powering smart-phones and netbooks? Now any one can build rich applications that have almost persistent power and Internet connection (an Internet-connected computer independent of the power or land-line phone/cable grid that's hundreds of times faster than the computers provided computing power to entire universities).




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: