OT: Given what you know now, if you were 21 again...

Michael Dillon wavetossed at googlemail.com
Thu Jul 21 22:33:45 UTC 2011


On 13 July 2011 14:08, Larry Stites <ncnet at sbcglobal.net> wrote:
> Given what you know now, if you were 21 and just starting into networking /
> communications industry which areas of study or specialty would you
> prioritize?

Number 1 - Learn how to learn. If you can't already do what Scott
Young does, then start with this URL and check out some of the other
content on that site
<http://calnewport.com/blog/2011/05/18/anatomy-of-an-a-a-look-inside-the-process-of-one-of-the-worlds-most-efficient-studiers/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+StudyHacks+%28Study+Hacks%29>

Number 2 is to learn how to set up a comprehensive virtualized test
lab, and build complex networks on that. A lot of real-world
router/switch software can run on virtual environments like Dynamips
so you have a chance to dig in fairly deeply for little money.

Number 3 is to leverage Ebay to buy old (and really cheap) routers and
switches. Only buy the really cheap stuff because it will be old and
broken in some way. Don't buy dead boxes, but the really cheap stuff
might not even have IP on it, or it might be a really old and limited
and buggy version. Make it work as best you can. Using version
numbers, find out what bugs exist in a particular box, and see if you
can trigger the bug and/or shield the box from harm. This is not only
a good mental exercise but will give your knowledge some depth of
experience that most existing people have, but which very few
newcomers will ever get.

Fact is that technology is still changing rapidly and you will need to
constantly be learning new stuff so make sure you do the same. That's
why number one above is so important.
Don't specialize. Just follow what interests you and try to avoid
getting over specialized because things change and it is hard to
predict which specialties will be in demand in the future. But people
who can think on their feet, have hands-on experience and are willing
to do whatever needs doing today; those people will always in demand.

P.S. you might also want to look into buying some microcontrollers and
FPGAs, and building your own data comm protocols from scratch. Don't
write a TCP/IP stack at first, but roll your own. Then if you happen
across an old token-ring only router, see if you can build an
interface to bridge it with Ethernet.




More information about the NANOG mailing list