Re: CCIE Lab Price Increase

From: nrf (noglikirf@hotmail.com)
Date: Fri Nov 02 2007 - 09:07:43 ART


----- Original Message -----
From: <M_A_Jones@Dell.com>
To: <noglikirf@hotmail.com>; <smorris@ipexpert.com>; <istong@stong.org>
Cc: <ccielab@groupstudy.com>; <security@groupstudy.com>;
<comserv@groupstudy.com>
Sent: Friday, October 26, 2007 10:57 AM
Subject: RE: CCIE Lab Price Increase

>Now the question is , where would one find an actual emulator that
>mirrors Cisco CLI.

Uh, actually, the real question is, would it really be that hard for Cisco
to build such an emulator? Heck, I very strongly suspect that Cisco
probably already has emulators inhouse, being used by their internal IOS
development/testing team. After all, the truth is, many (probably most)
software projects for embedded devices (and routers/switches are basically
just big embedded devices) are developed on emulators.

I'll give you some examples. A bunch of my friends work as developers for
handheld devices (i.e. Palms Symbian-controlled smartphones, MS Windows
Mobile handhelds, etc.). But they don't directly program a handheld itself.
Rather, their development environment is an EMULATED handheld. Basically,
they're using SDK's that are provided by the vendors that speak to an
emulated handheld that mimics, down to the hardware level, the complete
functionality of that particular handheld. Hence, they're developing from
standard Windows/Linux/Unix workstations that are speaking to a 'virtual
device'. When development and testing is done, they then port their code
to the actual device. Similarly, I know guys who programmed microprocessors
that go into cars (i.e. the processors that control engine fuel injection
systems or the processors that control car electronic user interfaces).
Again, they don't write directly to the processor itself. They write to an
emulated processor.

Now, keep in mind that these are DEVELOPMENT emulators. These emulators are
obviously extremely elaborate because you have to provide a complete
programming environment to the developers. I am not talking about that for
the CCIE lab. I am simply talking about a CONFIGURATION emulator. That's a
far far simpler request. After all, I think we can all agree that, no
matter how hard you think it is to configuring a router, it clearly is
nowhere near as hard as actually developing the software for the router.
{Think of it this way. You think BGP is hard to configure? Just imagine
trying to write the actual BGP code that is in IOS. Now THAT is hard.}.

The point is, Cisco could do it, and in fact probably already has (in some
form). If not, then the open-source community could do it. {After all, if
the open-source community can create entire OS's like Linux, entire RDBMS's
like MySql, entire language compiler sets like gcc, entire GUI's like
GNOME, and the most popular web-server in the world (Apache), then I really
don't think it would be that hard that for the community to create a Cisco
router/switch configuration emulator.} But the point is, it can be done.
It's not THAT hard to do, relative to all of the other far-more-difficult
software projects out there that were nonetheless successfully completed.
Some of you guys seem to be talking as if this is an impossible
undertaking - the "Mt. Everest" of software projects, when I see it as
anything but.

And to re-address the question of why Cisco should want to do this, the
answer to me seems to be simple: you can get rid of all of those labs
around the world (or convert that space to other purposes). After all, does
Cisco really need to maintain a lab in Brazil? Really? People can just take
the lab remotely at some authorized testing centers. They can obtain their
proctor access through decent videoconferencing, a technology that Cisco
also offers. After all, Cisco is SUPPOSED to be enabling remote work,
remote productivity, and remote learning. After all, isn't that the whole
point of even having a network in the first place? Doesn't anybody else
find it ironic that Cisco - a NETWORKING vendor - forces people to
PHYSICALLY travel to sites to test their knowledge about how to configure
that very networking gear that is supposed to enable remote access?



This archive was generated by hypermail 2.1.4 : Sat Dec 01 2007 - 06:37:27 ART