From: Dillon Yang (dillony@gmail.com)
Date: Sat Jul 23 2005 - 08:05:53 GMT-3
Standards?
Yes, it is the objective that we have to handle with. Roman was not finished in one day! The IP conquered the internet, why not SNA? Because the blue giant wanted monopoly in network and did some exclusive policy, however, the small vendors allied with each other by Internet Protocol and got success, such as 3com, Compaq. The exclusive DEC or IPX and etc. were gone with winds, IBM almost went into bankrupt.
Why would you like to service as a training partner not a certification authorization? Do you know seven jackals can kill a lion?
Now let us discuss the RFC 5723 ( for today is 2005-7-23) the protocol about certification in IT domain.
1. Introduction
1.1 motivation
Networking is playing an increasingly important role in military, government, and civilian environments. This document focuses its attention primarily on qualification requirement, especially operation of multivendor productions and interoperation between the productions as well. A person with "CSIE" can be considered having capability to do anything in the scope defined, just like the flag "CE" or "FCC" on a electronic device.
1.2 Objective
The qualification can provide a candidate the justicial information about common network concept and versatile capability of operation between multivendor production. Since a single vendor can never provide all devices in a network, the interoperation should be taken into account. GSIE will offer necessary infomation to promote the level of operation.
2. Cooperation
The participants should be responsible in exam, grading and training, updating the database of exam.
(suggestion wanted)
3. Functional Specificaton
3.1 exam
The candidate will take the exam based on the blueprint.
(suggestion wanted)
3.2 grading
The grading should be finished without knowing the information of a candidate by sections. That means the sections of a candiate will be graded by individual proctors from every participant.
(suggestion wanted)
3.3 training plan
The GSIE will provide a failed candidate a detailed suggestion on his weakpoint and advice on training.
(suggestion wanted)
TIA
dillon
----- Original Message -----
From: "Scott Morris" <swm@emanon.com>
To: "'Dillon Yang'" <dillony@gmail.com>; "'Ed Tan'" <tytanx@gmail.com>; "'Group Study'" <ccielab@groupstudy.com>
Sent: Saturday, July 23, 2005 10:46 AM
Subject: RE: GSIE vs ccie program Re: Lee's failed the 2nd
> ;) Not incredibly practical IMHO. Nice concept though.
>
> 1. Who makes the standards? We'll have just as many "this isn't MY
> reality" complaints no matter what level of design or complexity you put
> into it.
> 2. A vendor agnostic cert at that level would make more sense at the major
> ISP level where a predictable set of scenarios exist and complexity varies
> along with the vendors as you have listed below. However, most major ISPs
> really don't put a whole lot of belief into certification the way
> enterprises do. It's about experience. So do we add experience credentials
> similar to what the CISSP does for security management?
> 3. What will a numerical grade accomplish for you? In the end, you still
> set a pass or fail rate. And once you get done, who cares? I know folks
> who graduated college with a degree and a 4.0 who are out of jobs and not
> doing much about it now. On the other hand, I know folks who barely scraped
> through high school who are extremely adept at their profession and doing
> quite well for themselves. So what's in a number anyway? (What did you get
> on your SAT or ACT? (and who cares at this point?))
> 4. Wording is another variable event. Dialects are different. Native
> languages are different. Perhaps what I say doesn't mean the the same to
> everyone else. A review process can help "globalize" things, but there's
> always the chance. (side note: the CCIE program has a review process, and
> yet sometimes shit happens)
> 5. Who gets to decide the level that will "equal to CCIE or excel it"?
> ("exceed" would be the reviewed and replaced word there IMHO) Does each
> group make it harder than the time before them? How do you determine that
> someone is an expert at any technology like Frame Relay? P2P subinterfaces
> can't help.
> 6. Ahhhh.... The benefits of Prometric and Vue. Testking...
> Cheatsheets... Need I say more?
> 7. IELTS or GRE are hardly a comparative methodology. You are talking
> about something where hundreds or thousands of independent questions can be
> asked without any relation to other pieces. In the IELTS the maximum number
> of words in a writing part is 250 words. If we were able to create a set of
> independent scenarios, presumable so innumerate that anyone could discuss
> them openly without giving anything away, and yet let you only use 20 or
> less commands to accomplish the tasks, what would we really be testing?
>
> I really don't mean to rain on the parade here, or completely dismiss the
> idea. I think the concept of a highly technical multi-vendor exam is kind
> of a cool idea. But in the grand scheme of things, how many people really
> achieve a level of insanity where they are truly proficient at multiple
> vendor's stuff? Also presuming that you won't know prior to your particular
> exam which vendors you may or may not see?
>
> Over the years, the CCIE program (and Juniper's JNCIE and Riverstone's RNCIE
> and others) have come a long, long way. There are plenty of people who
> analyze the tests and effects and results and processes along the way.
> While there is always room for improvement, I would not term the exam
> "gaudy" at all. Even with the GRE, you have a huge pool of "potential"
> things (aka the DocCD), but you only get asked a finite amount of them.
> I've never taken the IELTS exam, but from what I can see on their web site,
> I would hazard a guess that there are similar concepts.
>
> Scott
This archive was generated by hypermail 2.1.4 : Sun Sep 04 2005 - 17:00:30 GMT-3