[Olpc-open] How will the OLPC truly help education?

[MBurns]

While Ivan was kind enough to give an initial response, I would like to
expand on some of the point raised. But again, as Ivan said, many of these
things are covered in the Wiki. You are doing yourself a disservice by
jumping right into the mailing list before reading the documentation and
background of the project.

On 11/18/06, Henry Skelton <dimensiondude.oss at gmail.com> wrote:
>
> Most of the answers I have seen talk about books being preloaded. Now
> that is certainly nice, but how is it better than regular books? A
> book costs very little to print (of course, you need material, much
> of which is copyrighted, but that is no different with digital
> copies). Certainly much less than $100.


Ok, lets expand on the idea. Lets say that, in a perfect situation, the a
quality book costs, $2. The covers cost of printing, materials, any of the
administrivia, etc, in producing 10 million copies of a given book. We also
assume the book is in the public domain, or similar. Now, clearly, 1 book
that costs $2, could be one great literary work, some select Shakespeare,
maybe, or something similar, take your pick.

And where are we? We can give 10 million kids 1 book, or 50, if we want to
stick with a $100 price tag per child. And if we were smart, we might be
able to make it different books, and encourage sharing among kids, so that 1
kid's 50 books are different than the next kids. Without doing any math, we
could reasonably get several hundred, maybe 1,000(?) books into a village
and reasonably assume a distributed library would exist. 1,000 books of
whatever metric you choose could be a great thing to give to a child, but it
is quite limited compared to the OLPC.

Taking the other approach, an ebook is something like 50KB to 1000KB(1MB) (a
quick guess from gutenberg.org). Pride and Prejudice is quite large, but
most books are considerably smaller. Each child can get 1,000 books (plus
articles, how-tos, tutorials, etc) on a wider range of subjects and kept
more up to date (internet access and digital files) with only a few dozen
megabytes used on the laptop. And with the nature of a general purpose
computer, those documents can be exchanged, removed and replaced without
trouble. Essentially, we are giving the Library of Alexandria, if you will,
to each child. Hardcopy books don't quite 'scale' that well.

And then there is a very big issue. Cost. $100/person will not be the
> cost, ever. Neither will $150 or even $200. You can't just drop off
> the laptops. You have to give them to each person, set them up, show
> them how to use it, and when it inevitably has problems, someone has
> to fix it. I've seen estimates at about $1000/person, and although I
> haven't investigated them, they seem very reasonable. I'd be
> interested in what others estimate, though.


You're right. There are costs for distribution (covered by the countries),
maintenance (to a small degree), and other overhead. No one says otherwise.
But $100 (or $140) is the price of the laptop, not the price of
administration.

As Ivan mentioned, there are essentially no moving parts on this entire
machine (save, technically, the monitor hinge, wireless antennae and keys).
All of the most common problems that lead to laptops dieing has been
removed. The harddrive and fans simply don't exist, and are the biggest
cause of failure on a machine. The monitor is simplified and rugged. the
connection between monitor and motherboard does not go through the hinge of
the laptop, allowing it to be much more rugged. The case is water resistant
and uses a thicker-than-normal plastic.

Remember, these are very smart people solving, in this case, a technical
problem. They have thought through the use cases and have planned
accordingly.

So you end up with maybe $5 worth of books for $100-$1000 and all of
> the problems associated with maintaining a computer.


This is a ridiculous number. It truth it is closer to $500 from day 1,
depending on how much space is dedicated to such material and might be as
high as $5000 from a Western point of view. It approaches infinity from
there on, as new copies and new files can be retrieved and shared without
additional cost.

<side rant>I don't know, but have you attended college in the US? This is
not meant as an insult, simply to bring up the price of books. $200-500, is
about average *per term* for class textbooks. While these are required to be
the latest edition (or you are 'on your own') the material doesn't
significantly change. The OLPC is a means to eliminate this cost. Several
groups I have read about are working on making Open Source (for lack of a
better term) textbook-like material for various age groups and subject
matters. This essentially gives children the ability to have access to
thousands of dollars of books dedicated to teaching them academic subjects
at their current education level. And we aren't even getting into other
areas of literature, articles or other non-academic documents that will be
included</rant>

Maintaining these seems like it will be a nightmare, as you'll need to be
> able to
> get people out every time someone's computer breaks. And they will.
> Just look at the current state of home computers in Western
> countries. Now, Linux is great, certainly better than what's bundled
> with most PCs, but it isn't perfect. It will have problems, lots of
> them. Books don't. If they are well taken care of, they last decades,
> and they can be reasonably expected to last many years.


Say what you will about Macs and Apple, but the number one reason they get
away with the slogan that Macs "Just Work" (TM) is because they control the
hardware. They get to write drivers that work on ever single Mac-branded
computer that reaches the public and the OLPC is doing the same. They get to
know exactly the type of monitor, NIC, video card, sound card, etc that will
be used on these machines, so there is no costly troubleshooting and no
fumbling to hit a moving target when new devices are released. They define
their hardware and write code to work with it perfectly.

I will go out on a limb and say that because of the nature of how drivers
work, they are the number one reason computers crash and have stability
problems. This is being eclipsed (on the Windows side, but is still a threat
to Linux) by viruses, worms and other malware. Both of these issues are
largely being tackled and solved. So again, the biggest threats to these
machine's ability to function longterm (stable drivers and malware) are no
longer threats. Nothing is a perfect solution, but this is as close as you
can reasonably get.

Now, the main advantage I can see over books is internet access. How
> is that really going to work in rural areas?


The plan is to have each village have Internet access. I don't know if there
have been changes to this plan, but thats what was last said publicly.
Either by dialup, longrange unidirectional wireless antennae, or by
satellite.

Mesh networking depends on a lot of other people keeping up the computer,
> and making sure
> they are on and the networking running.


No, it doesn't, actually. It depends on 1 other person within range having a
laptop with battery power (and minimal power at that, something like .1-.5
watts, if I recall correctly). By design, the 'range' of these is
significantly improved over consumer laptops. How improved is something we
will have to wait and see over the coming weeks.

Other people's computers cannot be lost, stolen, broken, neglected, or sold,
> or your internet
> access will work.


As Nicholas Negroponte has mentioned, the possibility of these laptops being
neglected or lost is almost a non-issue from his extensive experience. These
machines are treated as treasures to the child and the home and are very,
very well taken care of. As for broken, see above. The software is as rugged
as it can be, and component failure should not be a significant problem.

How often will people not have internet access because others are
> irresponsible?


Depends on topology. True Internet access will (again, I am assuming) be
given to the school via the school server, and then laptops within range of
that will share that network connection via a daisy-chaining method.
Depending on the village/town in question, this could be 24/7, or this could
be only during school hours. It is not unreasonable to think Internet access
is available the majority of the time.

How much can it cost to run a school for a year in a third world
> country? I doubt it's $100/child, and am sure it's under $1000. A
> school would actually educate these people, instead of being a very
> problematic tool that could possibly contribute to education.


The schools and communities do not buy these laptops. I'm not sure if that
is what you meant or not, but I wanted to clarify. It is the federal
governments that purchase and distribute them. How much it costs to run a
school, is dependent upon the country and location in question.

I hope this wasn't a rant on my part, and that it answered some of your
questions. You are encourage to join on IRC or continue to mail the lists,
but also, don't ignore the wealth of information myself and many, many
others have gathered on the wiki. ( http://wiki.laptop.org/ )

-- 
Michael Burns * Security Student
NET * Oregon State University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.laptop.org/pipermail/olpc-open/attachments/20061118/83dc72d1/attachment-0001.html