Lessons learned while building Basecamp

The Lessons learned while building Basecamp Session from O’Reilly’s Emerging Technology Conference 2005 with Jason Fried is available at IT Conversations now.

Jason is the founder of everybody’s darling postcontemporary web company 37signals, well known for their smash hit web apps Basecamp, Backpack, and Ta-da List.

4 main themes

• reducing mass
  (it takes more energy to change the direction of a larger object; you have to be able to change – so you need to reduce mass)
  less software
  less software means lower cost of change
  it’s easier to change hardware
  less support is required, the product becomes managable
  encourage human solutions (people come up with their own solutions for their problems; give them enough to solve them their own way)
  tada: no assignments, due date, categories,…
  people put the names in front, or dates; the software is in their hands
  you look at the patterns they produce
  let people use the software in a real environment
  (long term contracts, hardware,…)

• embrace constraints
  (creativity happens within constraints)
  (raise money, hire people, get more software)
  don’t just keep adding
  make sure that the time is spend on the right things
  (use the time-difference)
  IM replaced meetings
  make the money hard to spend, you make better decisions

• getting real
  (process for building things, no specs…, build the interface first)
  (specs are an illusion of agreement)
  the design is the functional specification
  (expert users not necessarily need complex systems; you can build different experiences though)

• managing debt
  (financial, bad code, hacks)

finding the right people – work with the right people is essential
skills are necessary, but

• being positive
• well rounded
• pick up things quickly
• be a good writer (that’s how people communicate nowadays)
  (ok, not everyone will find his David Henemayer Hanson)

trick: version one comes smaller than you want
(rant: beta, public but beta; have confidence in your product)
continue to improve it, always be iterating
say no by default (without being rude)
listen to the product, it will tell you when it is time to add something

a lot of small things that you think do matter, in reality don’t matter
idea every extra minute should be spend on sth new is bad; make what you have great, rock solid.
make your decicions changable (expensive decisions are hard to change, because of all the momentum that went into it)

make decicions just in time
no big inventory, only what you need when you need it
(myth scalability)
it’s totally cool to make mistakes in public, as long as you are able to fix things quickly
be open about it, blog about it
(when they launched basecamp, their were not even able to bill people)

chefs become waiters
switch postitions sometimes