[oclug] DND Sole Source Acan

[Chris Herrnberger]

On Saturday 17 February 2001 12:21, you wrote:
> Richard Lamothe wrote:
> > Chris: Can you provide more details? When was the ACAN published on
> > Merx? What specifically was being purchased, just a desktop  upgrade? If
> > so, what Linux company would offer a functional equivalent? Corel? I
> > rather suspect not. How about Red Hat? The challenge must be undertaken
> > with a firm that can offer a functionally equivalent competitive
> > product.
>
> Yes, I couldn't see much room for a challenge.  You can find it on merx;
> search for "Microsoft and DND".
>
> WINDOWS 95 UPGRADE - ACAN
>  Reference Number:	PW-$$ET-605-6203
>  Source ID:		FD.DP.QC.10077.C2
>  Solicitation Number:	W8474-01BC02/B
>  Published:		06/02/2001
>  Revised:
>  Closing:	16/02/2001  05:00 PM Eastern Standard Time EST
>
> Looks like that's all folks.  That's how our tax dollars are spent.
> ($8M)
>

8M...hmmmm the stations in question are standard issue 166 with 32 megs ram 
(vintage early 90's) soooo how much memory does w2K need to operate at 
reasonable levels and what processors are recommended. Sooo the 8M will 
possibly result in 50k x say 300/unit min upgrade (installed cost) = another 
150M if my math is correct. So lets adjust that to say a 75M extra/add-on 
inclusive of volume discount.

A challenge could then have possibly been based on the following. Leave the 
current units in place, install linux as the OS at the desktop level, deploy 
vmware server solutions for MS compatibility and set the entire system up an 
ltsp server nodes. Current hardware supports this solution. Lots of 
customization but non issue for the right hands. End result would have been 
dual linux/DOS terminal solution that boot in 15 sec or less, full 
compatibility with existing in house MS based apps and linux as the carrier 
OS for further integration with existing NOS. End users would never even see 
a standard linux desktop if so desired. Trivial (again in the right hands) 
but no one wanted to play LOB..... at least the folks that I talked too.  
Support and training components could have been outsourced to the right 
partners......same for integration and server admin. Desktop admin could have 
been eliminated as its a terminal based solution. Setup the help desk off 
site (standard contract) and your rolling. Am I missing anything? Role out 
and testing. On a terminal based solution these can be greatly reduced in 
comparison with desktop deployments. Security consideration? Reduced effort 
as its focussed on the server side. Isolate the mail requirements again linux 
based or freeBSD or variant combined with HP's mail solution and I think you 
could have quite a viable and secure solution. Labour? Possibly a challenge.

In my mind three or four local companies supported by the right distro 
vendor, and it could have been done. Good luck to those who challenged. The 
solution is currently working in other locals in Europe and the States....for 
a lot lot less. Comments on the local viability of the above are welcomed. I 
must admit I got wind of this too late (two week effort is not 
sufficient)....but not the next one..:) Also to be fair in the assessment 
there were probably other non technical reasons that I am not aware of at 
this time why this one went this way.

Best 

Chris