Chuck,
Hope your preparation is going well.
You make some excellent points, that I'd like to take up to the
10,000 foot level and generalize upon.
People often misconstrue that there is one proper user interface
(GUI, menu, CLI) and/or that a tool should give finished results.
The real message, however, is that there really are different user
types with different requirements.
A fairly basic distinction breaks network folk into planners and
operators. Another distinction is on skill level: expert vs.
entry-level.
Tools like ConfigMaker are appropriate for SMB work, and they have
user interfaces appropriate for the entry-level people likely to be
setting up their own networks. If I were configuring a bunch of SMB
networks, I'd be far more prone to use text-based scripts and
templates that emphasize my productivity rather than ease of use.
At another level, Routing Policy Specification Language (RPSL) is an
appropriate tool for describing interprovider routing, although it
isn't a complete solution for defining such problems and doesn't
directly help you understand what policies you should be defining.
RtConfig is a public tool that can generate most of your eBGP
configuration from an RPSL definition, but RtConfig neither is
super-friendly to beginning users, or capable of creating a complete
configuration.
From your description, Enterprise Design Tool should be regarded as
expert-friendly, suited for the problem of doing a first rough design
that MUST be reviewed by a qualified presales engineer. It does
reduce work for that engineer, but doesn't replace her.
Was this the tool that was being used to evaluate CCIE/Design
solutions? Scary if so...network design is sufficiently an art that
I don't think designs can be evaluated by a machine alone. By design,
incidentally, I am not referring to a complete set of configurations
that can be evaluated by a modeling tool such as Netsys, or by a full
Monte Carlo simulator.
Nortel's architect level certification has problems if it will scale
to cover large numbers of people, but has the reality that design
proposals will be evaluated by a panel of human experts.
As many of you know, I like to look at medical education as a good
model for networking. There is no such thing as a "paper MD."
Admittedly, do remember the technical term used for the person that
graduates at the bottom of a medical school class: "doctor."
There is premedical education that simply deals with skills for
understanding specific medical sciences. Traditionally, the first two
years of medical school deal with "preclinical sciences" such as
biochemistry, physiology, histology, pharmacology, etc., although
medical schools increasingly are providing some patient contact in
the first two years.
The next two years of medical school involve some lectures and
reading assignments, but principally closely supervised rotations in
patient care. The student watches more experienced physicians coming
up with care plans and diagnoses, although the student will take
histories and suggest diagnoses and treatments. It is expected the
student will come to the wrong conclusions a reasonable amount of the
time, but learn by the experience.
Moving to the "postgraduate" medical education, one must graduate
medical school and pass some tests to be considered for postgraduate
training ("intern" and "resident" are less popular terms; they tend
to speak of postgraduate year 1, 2, etc.). A PGY-1 physician has an
MD, but are limited in the complexity of what they will touch, and
have relatively close supervision.
It's PGY-3 or -4 before someone is considered fully trained in a
"primary" specialty such as family practice, internal medicine,
OB/GYN, etc. At this point, there are more exams, and one becomes
"board eligible" in a specific field. Typically, one has to practice
and present cases before being "board certified" in a given field.
Board eligibility and certification in subspecialties takes longer
(e.g., 3-4 years of internal medicine, 3 years of cardiology, 1-2
years of interventional cardiology doing angiography). At some
point, paper exams simply are no longer important. It's a matter of
presenting cases, demonstrating you've taken continuing education,
etc.
>This is the Enterprise Design Tool from NetformX. My employer has rolled
>this out to all us sales engineer types, and I use it regularly.
>
>Yes it is fairly decent, and I find it useful.
>
>Yes there are a number of irritating bugs. For example, one cannot place a
>redundant supervisor into a 6509. This is a problem that will be fixed "real
>soon now"
>
>There are devices where available blades do not show up.
>
>But I would say in general this is very useful if you are aware of the
>limits.
>
>Oh yeah - some of the product lines are not well handled in the design tool.
>Aironet, for example. Very high end switches, for example (as if I sell a
>lot of those ;-> )
>
>Also, it can be difficult to find the IOS image you want.
>
>I've sounded negative. Let me assure that I use the tool daily, and in
>general I like it a lot.
>
>Chuck
>
>-----Original Message-----
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
>McCallum, Robert
>Sent: Monday, February 26, 2001 6:03 AM
>To: 'Ccielab' (E-mail); Cisco@Groupstudy. Com (E-mail)
>Subject: Cisco Network Design
>
>Does anyone out there use the Cisco Network Designer tool? If so what are
>your views on it.
>
>Here is the link to view the actual tool.
>
>http://www.cisco.com/partner/cnd/inside.html
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
