OLPC - "eaten my homework"

[Jeff Kinz]

On Tue, May 30, 2006 at 02:34:56PM -0400, Ben Scott wrote:
> On 5/30/06, Jeff Kinz <jkinz at kinz.org> wrote:
> >>    Question: Does Cambodia really need to be spending its money on cheap but
> >> durable laptops imported from Taiwan?
> >
> > Its a heck of lot cheaper to make copies of bits than it is to make
> > copies of paper.
> 
>   Neither of the above appear to be correct in their stated assumptions.
> 
>   Most of the cost of a text book (or any book, for that matter) is
> outside the cost of materials and actual printing process.  Royalties,
> promotions, overhead, etc., is where the big cost comes from.  I
> believe I saw a claim that these texts would be "free" in this area.
> Assuming you can get sufficient volume, cost per copy should be a few
> dollars, or even less.

In the even less category we have "zero" from books available in the
Gutenberg project.  There is also work being done on creating free
curriculum texts that would be completely free of licensing costs.

>   Books can rot or be damaged; so can equipment.  I doubt there's any
> real data as to which would be more durable in the target environment.
>  Lacking data, we can't say one way or the other.

Given the number of times I've seen USB keys make it all the way through
a full wash and dry cycle, which a book can't do even once, I'd say we
can rely heavily on the fact that atmospheric heat and humidity will be
far more corrosive and damaging to paper than plastic and silicon.


>   The cost per copy for printed books is fixed for a given print run,
> but will always be non-zero.

The cost of producing a good textbook, even if the licensing/royalties
are zero, is very high.  The paper is a costly, high gloss, heavy weight
and acid-free. The other components (ink, bindings, labor) must be high quality 
as well, otherwise the book will simply fall to pieces within a short
period of time, especially if it is used in hot, humid environments.


>   The cost per copy for an "e-book" is also non-zero.  If nothing
> else, it will consume some of that flash storage, and the power to do
> the copy.  I expect this can be considered insignificant, though.

The virtual storage capability of the OLPC is, in some ways, nearly
infinite.  Each user can have access to their locally stored books, the
books stored on their fellow mesh-members Units, and anything they can
reach across the mesh from the Internet.  The user doesn't have to keep
every book they ever want to use on their laptop.  Since bits travel
faster than physical media, not having the book nearby is no longer a
barrier to having the contents of that book.


>   More significant is the cost of the equipment (the "laptop").  If
> you only have one book, that means your book cost whatever the laptop
> cost -- which is about $130, I guess.  That's an expensive textbook,
> even by US college standards.  It isn't until you start having a few
> dozen texts on a single laptop that you start to see returns.  Once
> you're past 100 or so texts, you should be into the "the laptop method
> is actually cheaper" range.

I'd say that the break even number could possibly be a lot lower than
100 books.  Given the cost of the raw materials and the printing process
itself, including color photo plates,  I think it might be more in the
range of 5 - 10 books.  Perhaps even less

> 
>   Whether or not the 100 texts (printed or electronic) will actually
> make it to the people, or actually be used if they do make it, is
> another question.

Yes, also we have no of knowing how much benefit people will get from
software that helps teach reading, math etc, or better access to other
knowledge and how many "book equivalents" that functionality is worth.

Not to mention the possibility of two way communication with
teachers/tutors in very remote locations.

"Hello, I am Sudar Mahmib. I am the son of the Nigerian minister of
Finance. I am writing to you because I am desperately in need of help.
Rebel insurgents have eaten my homework....."

;-)


-- 
Jeff Kinz, Emergent Research, Hudson, MA.
Speech Recognition Technology was used to create this e-mail