Re: Where rhetoric meets reality (routing)

["Eric J. Ray" <ejray -at- RAYCOMM -dot- COM>]

Ned Bedinger wrote:
> The problem:  I used a book from one of the big computer presses to get
> some ideas on how to write about command line commands on routers.  The
> examples they used were entirely made up using non-internet-routable
> addresses, even where the command syntax called for an internet-routable IP
> address.  These dessicated examples are bad news for me, when I'm learning
> about this stuff!

When you say non-routable IP addresses, do you mean numbers like
435.252.531.4251 or actual IP addresses designated from the pools
of private (not to be used on the Internet) addresses? If the former,
that's bad news. If the latter, then it's probably understandable.
Propagating bad routing commands can wreak havoc on the 'Net (and
has done so), and I'd be a little leery of giving verbatim examples
that would affect my router or someone elses. Given that private
addresses can be routed, just not on the public Net, I'd assume the former.

> To me, these examples look egregiously wrong and misleading. It's like
> kneading a cobblestone to learn how to make bread, you loose the hands-on
> information. I wonder why real examples of network configuration tasks are
> not valued by the publishers/editors/writers/students?  Is it because these
> things are usually taught in a classroom where embellishment is possible?
> For my money, the market for self-paced courseware books and certification
> is robust and publishers should be able to have a few IP addresses set
> aside for examples. A few blocks on a few different networks, a few blocks
> on different subnets and fairly lifelike examples would be easy to create.

As a rule, publishers wouldn't do this--O'Reilly might, but the others
we've worked with wouldn't. Support materials for books are generally
the province of authors, and IP addresses are hard enough to come by
for actual needs, let alone to set aside a block for examples.

> Clearly, non-routable addresses in routable address examples is something
> that professionals can learn to live with (the network engineers I talk to
> fairly sneer at this idea), but as an exercise in problem solving and a
> demonstration of professional demeanor, shouldn't Internic, or a standards
> body, or STC, or some big organization with lots of money/clout/ip
> addresses set up an address reservation for us poor example-starved users
> and writers of internet hardware/software docs?
>
> Is there a better solution that I'm overlooking?

Why not just use private addresses?
RFC 1918 specifies the following addresses for "private internets":
        10.0.0.0        -   10.255.255.255
        172.16.0.0      -   172.31.255.255
        192.168.0.0     -   192.168.255.255

Use those as examples and have fun!

In a more general answer to your question, safe data for examples
is hard to come by in many cases, from names to addresses to almost
anything else under the sun. If anyone has a database or listing that
they'd like to donate to the community, I'm sure that we can find
a home on the TECHWR-L site for it.

Eric

 From ??? -at- ??? Sun Jan 00 00:00:00 0000=