OLPC - "eaten my homework"

[Ben Scott]

On 5/31/06, Jeff Kinz <jkinz at kinz.org> wrote:
>> If it's the later, the costs are the same for both
>> printed and ebooks, as I already pointed out.
>
> That does it. I'm calling my Sister!  :-)

  She can subscribe to this list by sending a message to
<gnhlug-discuss-request at mail.gnhlug.org> with a Subject of
"subscribe".

[re-quoting for clarity]
>> If it's the later, the costs are the same for both
>> printed and ebooks, as I already pointed out.
>
> But you're not correct about that.  The layout and proofing work are
> different for each form of media, and paper media, being far more
> constrictive, requires a great deal more effort.

  Interesting.  I would not expect the required effort to be so
disproportionate.  Fair enough.  I stand informed.

  Now, you later go on to state that, for various electronic media,
one can simply present the material sequentially, without "layout"
done at all.  It stands to reason that one could employ the same
tactics to reduce costs in a printed text.  It would not look anywhere
near as "pretty", of course, but I expect it would be perfectly
serviceable.  Again, you keep implicitly assuming costs for the
printed alternative cannot be lowered by approaching the problem
differently, despite the fact that OLPC depends on that same
technique.

>>   As I said, I've never printed a text book.  But I have been involved
>> in support of efforts to produce low-volume copies of local manuals
>> and brochures, and if you do it in volume, prices can get very low.
>
> I've gotten manuals printed before too.  Its not that low.

  Not in the quantities mere mortals such as myself (or my employers)
have called for, but I've seen price progressions in the past (circa
1997) that were nothing short of incredible.

  We know the economies of scale exist (it costs less per page for
large print jobs vs small).  It is the specifics  which are in dispute
here.

> Especially when you factor in the costs of the time spent by all the people
> who work on getting it done.

  You've already mentioned volunteer efforts and donations of ebook
content to the OLPC project.  Can we assume similar things could
happen for a printing effort?

>>   Let's assume, for the sake of discussion, ten "books" ...
>
> Ben - There is a major error here: the price you cite here : 0.005
> (one-half cent) per page is the cost of the printing and does not
> include the cost of the paper.

  Uhhh... crap.  That's what I get for picking the first thing I find
with Google.

  Tell you what.  Since your family is "in the biz", would you be so
kind as to ask them how low cost-per-page can be for a high-volume,
mass-market printing run?  Not to print a textbook, just a glossy
magazine, book, manual, or similar.  (Again, purpose here is to
illustrate the differences economies of scale make in printing, *NOT
TO CLAIM TEXTBOOKS ARE THE SAME THING*.)

> Also, textbooks are (except for grade 1-3) are typically much larger
> than 200 pages.  On the order of 400-800 is more normal for grades 4-12.

  I know that.  Again I quote my message: "This is probably a big
underestimate, but it gives me some numbers to work with."

[Many paragraphs of fascinating and informative text cut.  Ultimately,
none of that is costing, merely explanation.]

> According to the American Publishers Association one-third of the cost
> of textbooks today is the cost of actually physically producing the
> textbook, including the cost of materials, printing, binding, design,
> layout, and writing.

  Now we're getting somewhere!

  So, right off the bat, we can "save" 66% off the cost of US textbooks.

  Some of that "savings" is distribution, of course.  Does the $130
cost of the OLPC laptop include the cost of actually getting it from
the factory in China to the hands of the kid in Africa?  I can't seem
to find anything on laptop.org that says explicitly, but I get the
feeling it does not.

  Of course, I expect it *would* cost significantly more to distribute
books instead of the laptops.  (Unless someone wants to hijack the
laptops.)

  Now, that 66% is before *any* economies of scale due to larger print
runs are introduced.  It also assumes a text with the same qualities
as a US textbook ("fancy", "pretty", detailed layout work, etc.).

  Do you have any figures on cost of manufacturing (materials, press,
binding) vs cost of editing/layout/etc, vs cost of actual writing?  I
can't find anything definitive on www.publishers.org (APA site).  Best
I could do was

http://www.publishers.org/press/monthly/pdf/JulyAugust%20Monthly%20Report.pdf

which states that "The majority of costs involved in creating
textbooks are in the intellectual capital and labor of the authors and
knowledge workers who create them" (page 2, column 2, first complete
paragraph).

  None of this leads me to believe the possibility of low-cost,
mass-produced textbooks is as unlikely as you imply.

> ... of course with your version they don't get a computer with the books.

  Ahem.  It isn't "my" version.  As I've said several times now, I got
involved in this discussion because I saw bad assumptions made on both
"sides" of the argument.  *My* only real point is that if you're
willing to question assumptions (like OLPC project is doing), you
often find quite a few things aren't the fixed constants you thought
they were, and maybe that concept could be applied to printing, too.

> Some web pages to look at:
> http://www.suhsd.k12.ca.us/elh/library/introduction.html
> http://www.bookspot.com/know/textbook.htm
> http://www.ncu.edu/estimated_cost_of_attendance.asp
> http://www.psusd.us/SARC/2004/DH.pdf
> http://web.bvu.edu/organizations/tack/asp/story.asp?id=1521

  *sigh*  You keep pointing to the costs of US textbooks.

  http://www.dell.com
  http://www.hp.com
  http://www.gateway.com
  http://www.lenovo.com

  Oh my gosh! Laptops cost thousands of dollars!  OLPC will never work!

>>> http://www.green-wifi.org/
>>
>>   Um.... yah.  I notice there doesn't seem to be anything about going
>> beyond a community-sized network.  To get to the net, it assumes
>> there's a nice, big, wealthy university nearby providing you the net
>> feed ("RoofNet")
>
> No - Nothing in their plan requires a university nearby.

  No, just a convenient, free Internet feed.  I'm assuming we don't
have that, for the same reason the laptop is the best source of light
for the target audience.  Am I wrong in my assumption?

> There is more to the web site than the one page.

  Yes, there appear to be a grand total of *four* pages (not including
the biographies and the contact page).

> Each region will require some kind of internet link.
>>   How much of the target audience has enough netfeed to enable even
> the community network model of Internet access?
>
> None - thats not the model, the plan needs one link per region.

  Define "region", please.  The Green Wifi page seems to imply a
municipality -- a city, town, or village.

> In some parts of the world this has been accomplished using dedicated long
> range Wi-Fi links.

  Parts of the world like Bill McGonigle's back yard.  ;-)

> Most important is the ability to email. (IMHO)

  What happened to nearly-infinite storage?  ;)

  I know!  IP over avian carrier!  ;-)

> One way you get a small amount of information, on very heavy paper,
> that you can't read at night (no lights) and nothing else.

  You could always burn the books for light.... um, never mind... ;-)

> The downside?  Having to filter access to pron.

  Does your average African villager even care about filtering porn?
(Serious question, albeit totally off-topic at this point.)

-- Ben