FOWA: The Future of Web Startups (Paul Graham) [en]

[fr] Notes prises à l'occasion de la conférence Future of Web Apps (FOWA) à Londres.

Here are my live notes of this Future of Web Apps (FOWA) session. They are probably incomplete and may contain mistakes, though I do my best to be accurate. Check out Paul’s essay derived from this keynote. The conversation also continues on the YCombinator news site.

steph-note: missed the beginning, very incomplete

FOWA 2007 88

Standardizing things, from funding to acquisitions. Acquisitions are interesting because the buying company knows exactly what “brain power” they are acquiring.

Instead of approaching venture capitalists with a plan, start the company with a few thousands of $$ from your uncle or Y Combinator, and then approach VCs with a company. steph-note: I thought this was the obvious thing to do

We still need startup hubs. You need to make a startup succeed, not just start. Here is the value of startup hubs: face to face meetings. No technology in the world replaces that. Whether you need it is not the question: the important thing is “does it offer an advantage or not”? If it does, then your competitors will have it over you if you don’t do it.

The ability to be able to work face-to-face for three months greatly outweighs the disadvantage of moving.

Seed funding is a national business, contrary to VC funding which is regional. No regional Y Combinator branches. Just like you can’t have a regional “big university”. But maybe seed funding is actually international?

If seed funding is indeed international, then not really possible to create other “Silicon Valleys”, because the people who are really motivated to succeed will move to SV, and those left behind are “less good”.

Acquirers are assholes, even the nicest companies (lawyers, “they’ll make you pay”). Need: Chief Acquisition Officers. Would both identify the opportunity and close the deal. Now it’s two separate steps. Maybe in future, big companies will have both a VP of Technology (in-house) and a CAO (bring good stuff in).

College may change, if hackers start building startups. For the moment it’s warped towards preparing you to have an employer. There’s nothing magical about a degree. Do you need a degree if you’re going to start your own company? The need for degrees is driven entirely by administrative requirements.

Don’t encourage people to start companies in college, though, because that gives them a great excuse to abandon their startup. OTOH, some of their best founders were still in school.

The greatest value of university is not the brand name or maybe the classes, but the other people you meet there. steph-note: not sure this is valid outside the US. Shift from getting good grades to impress employers to actually learning stuff because you’ll need it. steph-note: OMG, is US education that broken?!

Increasing the number of startups would mean you can’t sit on an idea if you have a good idea, because other people have your idea. So if it gets easier to start startups, then they are more likely to actually do it.

If people actually get to work instead of sitting on ideas, technology will evolve faster. Some ideas are too scary! Look at how hard a time Microsoft is having trying to figure out web apps. New ideas implemented increasingly in startups rather than big companies. Big companies are just not a good place to make things happen fast.

Talked with a guy who had his startup recently acquired by a big company. From a “lines of code cranked out”, they were 1/13th as productive after the acquisition. Something about big companies that just sucks the energy out of you.

Y Combinator: there to release energy by making it more easy for hackers to start their startups.

For the moment, the process of starting a company is a whole series of tubes 😉 — lots of kinks in the plumbing.

In future: a big straight pipe. Being measured by performance, fleshing out the arbitrary crap people are measured by nowadays.

Paul talked about exit strategies, not running a company to make money. Startup means exit. If there is no exit, there is not startup. Not all technology companies are startups. Not all new companies are startups.

Hackers actually like to make stuff, they’re not in there for the money. So actually, if you let them make stuff, you can pay them less! Big companies are paranoïd about their brand, they should be less scared about releasing stuff. Companies are judged by their successes, not the crap stuff they might have released (look at Google). Just let developers release stuff to the world.

What can we do to encourage startups? (Question from Ian Forrester, BBC).
A: Make documentaries on people doing startups. Seeing how it goes is usually what convinces people to take the plunge.

If you just want a couple thousand $, don’t raise VC money, just get angel money. What makes Silicon Valley is the angels. Google would have never made it if they hadn’t had angel money.

Y Combinator is going to open source their angel money paperwork, to make it easier for “rich hackers” to invest.