[sugar] OLPC's bizarre behaviors

Martin Langhoff martin.langhoff at gmail.com
Thu May 22 01:38:37 EDT 2008

On Thu, May 22, 2008 at 1:53 PM, Kim Quirk <kim at laptop.org> wrote:
> Lots of things that we do don't meet any normal expectations of a
> 'company'. Most people at OLPC will tell you we are not a 'company'.
> I have been trying to understand it, explain it, live with it ,
> and improve it for a year now. What I think is going on is a unique
> and somewhat chaotic (perfect storm?) intersection of non-profit, open
> source, research lab cultures with the need to ship a real product.

This is excellent analysis. And I'd go a bit further than Kim actually
in saying that I don't feel particularly bad that we are a bit of a

Being a bit of a mess means that we are breaking new ground so quickly
that the ground is changing faster than the org gets used to it.

Which leads to a few observations (which overlap somewhat with what
Kim is saying)

 - IME, people complaining that "we don't know what we are doing" can
be a positive indicator. The scenario outside the car is changing!

 - Learning to organise and handle new situation X is only worthwhile
once we are confident that X is here to stay.

 - Therefore, there will be many situations that are impossible or not
worth to be well prepared for. So "being a constant mess" is a
reasonable approach. We can handle that by saying that "strange new
situations" are common, and we have to keep an open mind and be ready
to work w the team to get new and strange things done.

 - Prioritisation is important. Some things are too much of a
distraction. Letting them go to hell can be less disruptive than an
all-hands effort. This is - IME - the hardest part. When everyone is
ready to take on whatever comes, it's hard to avoid getting the team

Which can also be stated in more blunt terms: We are doing development
of new stuff! If you want it predictable and organised, I hear EDS is
hiring - the processes and procedures manual is 800 pages :-/

All of the above is from my experience in various organisations large
and small, and govt and private. We are radically diffrerent from a
big corp, and even from established non-profits. In this space you can
expect us to be very good at a couple of very specific things, and a
complete mess about a lot of other stuff. We will have to get good at
"some of that other stuff"... in the meantime it'll be frustrating.

There's a good book about this - Waltzing with Bears by either DeMarco
or Yourdon, that says basically: if you are considering a
project that doesn't take you into uncharted territory, *can it*. It's
not worth it if it's not so new that you feel lost and helpless. It's
written for big corps that are frozen in terror ;-) but it applies to
what we are doing @ OLPC.

Uncharted territory. So everytime we spot something in the horizon
there's some fear that the earth might actually be flat.

but I think we should keep sailing no matter what.

 martin.langhoff at gmail.com
 martin at laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff

More information about the Devel mailing list