Certification or College degrees?

Sun May 26 06:33:11 UTC 2002

	(this is actually my first NANOG post ever...) 

--On Thursday, May 23, 2002 03:07:55 +0100 cw <security at fidei.co.uk> wrote:

> I am currently studying a BSc degree in merry old England. I have
> just finished my second year (well I'm part way through the exams).
> When I applied to do my degree I found two universities whose course
> were anything related to Networking. Mine is called Computing
> (Networks and Communications).

<snip explanation of curriculum that I'd avoid>

At the Royal Institute of Technology in Stockholm, Sweden, we have a series
of courses that focus on networking. The starting one can be seen as
"getting the programmer to know IP's quirks", but as we progress, we teach
deeper and deeper into the technicalities of routing, including theory of
routing (discussion of Dijkstra, and similar) and practice; we have a
routing lab where we first make them understand that static routes don't
work and then progress into understanding first OSPF, then BGP. 

The entire package runs over a period of half a year. Prereqisites are that
the student is at her/his third year in a Master of Science path aiming for
one of Computer Science, Technical Physics or Electric Engineering; i.e. we
want people to have a solid ground in theory before we teach them the dirty
details of networking. 

The best students are encouraged to write their final paper in the field of
networking. Some of these are later found working at KTHNOC operating the
NREN Sunet and the pan-Nordic REN NorduNet.

Myself, I teach DNS in the introductory classes, including such novelties
as DNSSEC, which we have the students sweat over in the lab. I've been
somewhat depressed by the point-and-click generation, who don't understand
classic Unix, (because the DNS part does border quite a bit on sysadmin
stuff, which we do not teach) but on the whole, it's been successful. 

-- 
Måns Nilsson            Systems Specialist
+46 70 681 7204         KTHNOC  MN1334-RIPE

We're sysadmins. To us, data is a protocol-overhead.