OLPC: Open Organized Transparent

C. Scott Ananian cscott at cscott.net
Tue May 27 19:38:47 EDT 2008

Well, since I'm apparently the one fingered as "smart", "holier than
thou", and "derisive", let me publicly apologize for being
short-tempered at times.  I do get frustrated when I see the same
issues pop up over and over again: remember there are many many more
of you out there than there are here at 1cc, and in order to be
successful we at OLPC *must* allocate our time wisely.  Sometimes that
means I'm rather short and/or terse.

A related issue is when people loudly insist that "OLPC" solve their
personal problem *right now*.  Again, we have tens of thousands of
machines in the field now, and thousands more every day.  You
personally may care about, say, Java in your browser, but it is not a
priority for "OLPC", by which I mean the 3 people I sit next to.  It
is not part of the software included in our large scale deployments.
By all means work on the problem, and we will certainly help you
publicize the solution you come up with as much as we are able, but
there are not resources to devote to every feature request.  We have
to prioritize.

I have made many efforts in the past to ask for help and publicize
issues for which OLPC would like help but can't dedicate any developer
resources.  Statements such as:

   "Without naming names, though I was excited to help at first, that kind
of insider-outsider issue made me lose interest as a direct contributor
fairly early."

sadden me.  There *are* a few areas where we're actually working on a
particular solution, but in general our feature set far outweighs our
resources: we're always interested in working code, even if it's not
exactly how "we'd" do it.  But I'm afraid the following sentence
illustrates the crux of the problem:

   "In the end my opinion won't really matter, so why waste my

It is true.  Opinions don't really matter for much.  There are less
than ten full-time OLPC developers, it's not like we're some big
company.  We're working flat out to make our Peru and Uruguay
deployments work at the moment; we don't really have time to chew the
fat.  Your opinions matter much more if they are backed up with
working code, or with a community of volunteers to attack some task,
or a well-written report.  We get a lot of opinions.  Many of them
are, frankly, misdirected.   For example:

    "I think the OLPC's decision to sell XO's only in large quantities
and only top down to educational institutions is wrong."

I have zero control over that.  Generally speaking, such discussions
are out-of-scope for devel at laptop.org, which is a developer's list,
not a business-models-and-strategy list.  You should be making these
types of arguments to OLPC's board, say, and they would expect a much
more rigorous argument, backed up with numbers and dollar amounts.  To
be honest with you, it's not an effective use of my time to debate
OLPC's business model.  But if you want the actual answers,
http://wiki.laptop.org/go/Presentations is a good start.

OK, in summary, I'm sorry I can't just mumble some comforting
platitudes here.  But the people I work with *are* passionately
concerned with creating a community around the OLPC, and we are *also*
overworked.  The devel at laptop.org list is likely to remain rigorously
focused on implementation and working code, although there are places
like IAEP which welcome more blue-sky discussion.  In that spirit,
here is a list of development tasks that OLPC doesn't have resources
for, but that we'd love help with:

* Java in the browser.  Sure, trac #6465 and similar.  We also have
people who would like a more convenient way to install Adobe flash and
other plugins.  This can be done by making a slightly modified version
of the Browse activity, but no one has done it yet.

* http://lists.laptop.org/pipermail/devel/2008-May/013696.html

* a long list at the end of

Finally, in response to your numbered points:

1) http://wiki.laptop.org/go/Installing_Debian_as_an_upgrade
 2) http://wiki.laptop.org/go/Developers_program
     (also NN announced a new G1G1 program during
http://wiki.laptop.org/go/Presentations/May_2008_Country_Workshop )
 3) http://wiki.sugarlabs.org/go/Main_Page
     and Microsoft was also discussed during one of the first three talks at

I'm sure I haven't said all the things you'd have liked me to say, but
I've done my best to be open and honest here.  Thank you for starting
this discussion.

 ( http://cscott.net/ )

More information about the Devel mailing list