[kwlug-disc] what would you pay for good kernel documentation?

[Robert P. J. Day]

On Sun, 6 Jun 2010, Paul Nijjar wrote:

> On Sat, Jun 05, 2010 at 11:45:20PM -0400, Robert P. J. Day wrote:
> >
> >   i'm thinking that it could be clear from the beginning that,
> > say, the first month's lessons would be free, after which you
> > would subscribe to the rest of the course for some reasonable
> > amount like $39.  if you didn't think the first month was worth
> > it, you walk away and you've lost nothing.  if you were enjoying
> > it, you pay the money and complete the course.  (there's no bait
> > and switch here -- it would be clear that there would be a
> > subscription fee after the first month.)
>
> I know that the word "month" is tentative, but does this refer to
> time or to a fixed number of lessons? I would suggest that the
> latter is more predictable (for the end user) and easier to
> administer (for you).

  not sure yet, that part is still fuzzy but it's easily clarifed once
i decide how much the course will cover.  bit i'm thinking that 20-25
lessons would represent a reasonable *intro* to kernel programming.
if that works out, i can always introduce a more advanced course but
i'm not thinking that far down the road yet.  just wondering if i can
make something like this work and have it worth my time and effort.

> >   i have to admit, given how people in some of the other forums
> > were freaking out over a proposed amount of five or 10 dollars
> > just for documentation, $39 might drive them right into a fainting
> > spell.
>
> If the interactive course is all they get, then you will have
> piracy. One alternative is to include a portion of your *services*
> with the course (which might well raise the price).

  of course one expects piracy.  but only registered users will be
allowed to leave comments so they can ask questions.  that's at least
*one* incentive to actually register.

> For example, you could critique submitted code snippets for
> customers for things like style and common errors. I could see that
> being valuable for somebody who is trying to get code committed to
> the kernel, or to make a kernel module that is going to be
> maintainable over kernel versions.

  that's possible.  but at the very least, having access to each
lesson's comments section would allow registered users to ask
questions to which either i or someone else could supply an answer.
so that represents at least a *first* level of support.  at some
point, i'd have to draw the line.

> That relates to maybe the most important question: what kind of
> target audience would be willing to pay for this? If you really want
> to make some income from this, then those are the people to target,
> not general newbies who are kind of interested in dabbling with
> kernels.
>
> I think one reason people faint over paying money for kernel
> training in forums is that there are a lot of things to learn, and
> unless there is a solid reason to learn about kernels there are
> cheaper learning opportunities out there. (I could make disparaging
> remarks about freeloaders in the Linux world, but I don't want to be
> guilty of self-incrimination.)

  i appreciate all this, and i could hunker down and do copious
amounts of research to determine the target market and how to reach
them and so on, or i could take a chance, put this together, let as
many people as possible know when it's ready to go and see what
happens.

  if people want to chat about this in person and bandy about more
ideas, obviously, i'll be at kwlug monday evening.  and i do
appreciate all the feedback.

rday

-- 

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================