Online extra: Readers respond to NMCI

In your opinion, what can be done to improve the rollout of NMCI?

Too expensive.

Better planning before trying to take on a job of this scope. Reduce the number of subcontractors working on the project.

Stop and fix all of the current issues, upgrade the applications, the baseline is too old.

Change contractors.

NMCI needs more empathy for end user needs. There are many productivity losses by constraining applications and IT support. And, there are still many security issues behind this private network.

Collect true performance data and require actual performance as originally touted. The vast majority of users did not need a new email and document system they needed technical tools.

Cancel NMCI.

Communicate with the end users and care about their work. The Puget Sound Naval Shipyard building I am [at] has been condemned and everyone will be moving to a new building in a couple of weeks and then the demolition begins. NMCI people were running network cables and they didn't stop work when I let them know about our building. They said they didn't care, they were just collecting a pay check. I finally found someone in their organization that cared and the worked stopped.

The horse is already out of the barn thus it is too late to close the door.

There doesn't seem to much of a learning process going on. They seem to have to reinvent the wheel. Incomplete user data slows the building of PCs. Incomplete infrastructure also slows down the deployment of PCs.

Put more effort into solving the numerous problems. Currently all effort has been put into installing seats in order to meet a quota.

Allow R&D folks who require cutting-edge systems too continue purchasing their own systems as required by the projects.

At some point I believe they [the Navy and EDS] are going to have to sit down and rethink what they are trying to achieve. The goal of a secure networked environment spanning the Navy and Marines is good. The implementation has been lousy. Many features of the system add nothing to the security or networking goals and seem to be chosen as the least investment path for EDS versus the best way of accomplishing the Navy's goals.

Spend more time evaluating current Navy/Marine networking issues. Also provide more and clearer guidance to end users regarding NMCI.

I work in CAD-related area. The addition of tools needed to do my job is not there and isn't likely to be in the future.

First, they could find a schedule and stick to it. In typical contractor form they have slipped way to the right, and have held up our purchasing of technology that would have made the last two years a lot easier to deal with. There have been a lot of empty promises, and a lot of mistakes along the way, and the rollout is a constant headache. This continual "NMCI is coming, give us X information!" Chicken Little song and dance has been a waste of time, manpower, and money. I'm sure someone sees the benefits of NMCI, but no one around here does.

Maintain the legacy systems for a period of time (about six months) because the rollover to the new system takes some getting used to, especially remembering all the various passwords needed to log on.

Ensure they understand the needs of the folks who have a job to do.

Only a few NMCI computers should be installed until a command is satisfied that they are ready to make the change from legacy to NMCI. Having a legacy and NMCI computer on everyone's desk for over six months with mostly redundant features is not anyone's best interest except EDS.

NMCI needs to work more with the end user before rollout.

Recognize that the [science & technology] community needs a much more sophisticated set of software than is needed for admin purposes.

As more and more users are cut over, we see a drop in Internet response time.

Nothing.

Provide more human assistance in solving problems.

More input from the end users. Some of the programs I need are not offered through NMCI. It also takes for forever to get a program approved-more lost man-hours.

There are no less than six different types of printers in our working area. We are forced to maintain supplies for each of those printers. If a printer stops working they replace it with another printer and the supplies for the printer that broke are no longer usable and are taken as a loss to our budget. One type of printer would really help.

At this point, nothing. The rollout is so far into the process that most people are trying to make the best of a bad situation.

They need to be able to approve new software a lot faster than they do now. Should have a very flexible contract. (If it isn't in the contract they don't do it.) Government doesn't work well with non-flexibility.

Kill it.

Fix what is not working. Better response time. Better communication when issues have been resolved.

No comment.

Nothing. Abandon it.

Time and time again we have seen government programs and projects fail because of schedules that are moving targets. These schedules move to the right not because of changing government requirements, but because the contractor is unable to meet their own overly optimistic projections.

End contract now before it's too late.

Forget it.

Turn over the rollout to Navy employees.

The rollout for us was first-rate. EDS personnel were sharp and had the answers ' even the Navy people they hired were better than they had been when they were working for us.

Place more emphasis on supporting people trying to accomplish their mission.

A competent prime contractor would improve service.

Stop the roll out.

Fire EDS and get another company.

Reduce the cost for individual needs (e.g.: R/W CD players to be able to back up your hard drive should be no additional cost; Microsoft Project [should be] available to all users.)

Go back to the old ways.

Cancel the rollout wherever R&D functions are being executed in DOD. The "business office" structure of NMCI fits poorly in R&D environments where PCs are often treated as research tools.

Thoroughly train the installers.

NMCI will not work because the Navy is not one-size-fits-all.

Give us the hardware and software we need to do our jobs and give it to us in a timely manner and at a reasonable price. AND, listen to the user's inputs as to what he or she needs.

No comment.

New employees sometimes have no computer or e-mail for a month or two-even though seats were ordered 1-2 months before they arrived. This would never be tolerated in the private sector.

More planning and better training would have been beneficial.

Take more time with rollout process instead of just deploy, deploy, deploy-while leaving messes behind.

Get on with it. The course has been set. May as well see what is on the other side of the train wreck.

Getting software/hardware is a major problem at NADEP NI [Naval Air Depot North Island, San Diego] even though your job may depend on it. All requests for this have to be approved by NMCI. This is ridiculous if one is an engineer and performs R&D work. Let's see: NADEP loves diversity and NMCI/NADEP on the other hand is too regulated and lowers computer efficiency. You people making decisions just don't get it!

Establish some consistency in performance. Make personnel follow-up to truly debug each system. Mine has never worked correctly since the [Remote Access Server] push.

Let people know what the rollout and deployment time tables are and what is causing delays.

Allow involvement at the local level rather than dictating configuration by EDS and top brass.

Limit NMCI to the fleet and military squadrons. The corporate groups don't need it.

Nothing.

Continuous communication is a must. Leadership and communication will help to drive this project forward from both government and contractors.

They need to get more people who know how to load and start programs.

Train users. Make sure the trainers show the user everything on the NMCI machine they might need to use. Do something about the help desk menu-it can be confusing.

I don't work with this application so I can't tell you anything.

Put the customer, the Navy, first.

Bandwidth.

Be more organized, fulfill promises and schedules, be responsive to customers.

Completely abandon NMCI!

NMCI end-user support is lacking, to say the least. And I'm only under AOR [Assumption of Responsibility]-have not yet cutover to NMCI. It takes days to get a fix done that used to take minutes. I was told that after cutover, support will be even worse because they have fewer techs available to each command after cutover. You ask what can be done to improve NMCI-the whole concept is not achievable, in my mind. It's too vast, it's supposed to do too much too fast. It also doesn't have security for commands such as mine that have sensitive info that just anyone can't have access to. Many of their contractors who do have direct access to the Navy's data do not have clearances, although the contract says they do. It's just a bad idea any which way I look at it.

Terminate the contract.

The product that EDS is providing the Navy is a very basic service which generally meets the most fundamental IT requirements in the business and administrative functions. Beyond that basic product, EDS is struggling make any progress in providing a substantial service to the RDT&E [research, development, test and evaluation], intel, and C2 requirements. Actually they are still having problems rolling the basic seats. As for AOR, the American public is the big looser here, not just the Navy. AOR makes EDS responsible for the as-is IT assets at a given site. The old government help desk at this site received 3,000 calls per month per AOR. Now the same help desk is still receiving 1,400 calls per month. The local NMCI help desk is logging 800 calls, of which, they turn over one-third to the government to resolve. This is not saving the Navy, DOD, or the American people money.

Do you have any further thoughts or comments on any aspect of NMCI?

Poor planning, poor execution.

Quit forcing the issues down everybody's throats. Fix the problems first, instead of deploying a defective system.

Cancel or greatly modify the scope of NMCI.

Those of us who have locally developed GOTS [government off-the-shelf software] are being treated like ^&%$# by NMCI. You have to pledge your children to get anything done. NMCI has loaded an unbearable bureaucratic burden on us. Isn't that ironic, an outside contractor burdening government employees with red tape!

EDS was very inefficient in its rollout at my installation. Had to be overly costly. NMCI is quickly becoming more and more inflexible with order changes and meeting changing requirements. It seems it will become a very large and inflexible bureaucracy.

This project layers control away from those who need it to get their job done!

Funding and compliance with IT-21 [Information Technology for the 21st Century] was never given this amount of importance. If the decision-makers (who pushed NMCI through) had been able to fund and empower the local ITs, this could have been accomplished with a lot less growing pains.

I believe triage is a useful tool. Put it out of its misery until a definition is found for what is truly needed.

We were not given "industrial quality" printers. They are one step above "homeowner" quality. They are slow. They are falling apart mechanically. Neither the paper trays nor the cartridges have the capacity needed for production printing.

We have been given direction against saying anything negative.

This is a major endeavor, one that could possibly leave the Marine Corps in a bad position if satisfactory service is not provided. God help us all.

Desktop basic support is good in area of Microsoft Office applications. Everything else is very poor.

It would be nice it the whole thing were dropped.

More administrative rights could be given to the help desk technicians that would eliminate time-constraint problems.

It is basically a "no frills" system with absolute unresponsiveness to the slowness of retrieving, storing, and moving materials. Granted, our activity was computer literate, but it does no good to pull down every outstanding activity to the level of the poorest performers.

Better prepare end users for cutover and check if older programs will be compatible with the new system before putting the system on line. Some personnel are still trying to get needed programs to work. Daily, some personnel are in touch with NMCI personnel in different departments about making the system work. There is the issue one department within NMCI saying that is not their problem. NMCI is NMCI regardless which department it is. All have to work together to make it happen.

Like almost all such processes it will fail under its own weight.

They (EDS) do not have enough technical support staff on site. They seem to cut jobs, someone calls them on the lack of local support and new faces are hired.

NMCI punishes activities like mine that had a robust and modern network infrastructure to bring up to date activities that long ignored theirs.

The cost seems very high for any perceived benefits.

A better keyboard needs to be offered to people with carpal tunnel syndrome, like me.

The vast majority of problems encountered in NCMI deployment result from leadership failures on the Navy side. Mid-level managers have adopted a combative attitude rather than cooperating with EDS to get the job done. Navy underestimated the challenges of legacy application reduction and waited too long to establish the FAM [functional-area-manager] process. Feedback from the field level (those who are actually doing the implementation) is not being solicited and is generally ignored when offered. Higher echelon staffers do not understand how their actions (or inaction) are impacting users at the field/operational level.

I am surprised at the lack of sophistication of some of the hardware, such as a two-button mouse.

I get two to three new softwares a year that I need to use right away but the process to get it approved for use on NMCI takes way too long. I can't get a standalone computer because there is no money to buy it because computer money is tied into NMCI.

Kill it.

I don't know how it is better for us if we now maintain two workstations, need approval to load software, and are charged for additional software and hardware.

do not see the savings projected for this effort. Believe NMCI comes at a tremendous cost. Where will these [dollars] come from and what must be sacrificed to absorb the cost?

It's coming-congressionally mandated, no choice. So we're stuck again.

This is a great start for departments operating with overhead funding, not very well for direct funding projects. Process is process.

It is too costly, overrated, and the end user doesn't have all the tools or the flexibility to adequately do his job. The system frequently bogs down. The intent of NMCI was good, but there is too much control down to the working level for employees to accomplish their jobs in a positive manner.

Cultural resistance is more a problem because new customers are aware of the many problems associated with NMCI during rollout.

If the computers were given to sailors that didn't have one, rather than starting at headquarters, there might have been some benefit. But Remote Access Services are so bad that unless digital subscriber line/cable/network access was available (by the way, NMCI just shut off DSL!) the computers are of limited use. FTP Client and Telnet STILL don't work through RAS. I'm glad I'm retiring soon.

EDS didn't do due diligence when they entered their bid and now the government is paying a much higher price. We need to cut our losses and start over with internal resources or another company.

For reference, read the Gartner article, "Outsourcing Enterprise Network Services in the U.S.", dated 12 January 2004. In it they state not to overestimate the cost saving benefits of outsourcing because they may be difficult to quantify, and even harder to prove. Also read the companion article by Gartner entitled "Retain enough resources to manage outsourcing deals" dated 17 June 2002. The Navy is getting rid of 2210's with the knowledge of the legacy systems, not just network types. Who's going to be left to hold down the "fort" (or ship in this case)?

NETWARCOM [Network Warfare Command] is a major problem. They sent messages because they can't write instructions that comply with current instructions. Their messages are treated as instructions and the information within the message does not comply with current DOD or Navy instruction.

Cancel NMCI.

Hire technicians to take care of problems in a timely manner; having to wait 12 months for a hardware upgrade is unacceptable; service is far too expensive for such minimal support.

What a waste of resources. It has been like plowing under 4,000 family farms to create a single 500,000 acre corporate farm. And couldn't the Navy exercise more buying clout?

Question 6 [Are you optimistic that the department ultimately will see the projected benefits of NMCI?]: If the department sees benefits, it will be the benefits they're looking for, which won't necessarily reflect the worker's reality. Question 9 [Do you think that cultural resistance is still a problem for NMCI?]: Cultural resistance is still an issue to some degree but, if NMCI properly supported the worker's needs, this resistance would have melted away by now. Question 11 [How would you rate the quality of MCI End User Operational Readiness Training Sessions?]: I haven't been offered any training sessions, therefore, I can't rate them.

Extremely costly for what we get, not flexible enough (equipment does not support engineering functions, only office and financial functions), takes too long to get legacy and new applications approved (or disapproved), too long between order and rollout, RAS connections are almost unusable (it takes me hours to download e-mail on travel), laptops weigh a ton.

Eliminating legacy systems that work is not always a good idea. My system is consistent and used the many times that the NMCI system has not been working. The speed of the connection is bad and the inability to access the legacy address system because they underestimated what was required is totally unacceptable.

This process is a nightmare at all levels. Hopefully the Navy will see benefits with the program down the line, but up to this point it has been WAY MORE trouble than it is worth. The price tag for this whole deal is also outlandish.

NMCI has been the 600-pound gorilla at our door for well over two years. Its principal effect has been to put a budgetary and political chill on all advancement. We have software nearly two years old that we have not been able to deploy over fears of what NMCI might mean. The Navy is trying to use the legacy apps thing as a way to pay for this white elephant and to buy time. If legacy apps throw away millions of dollars worth of software licenses only to go buy new licenses somewhere else, someone ring up 60 Minutes. Lean, consistent applications are fine but not by shotgun; not all at once.

This would be a great Dilbert book ' "NMCI, My Life With The Invertebrates."

Provide applications such as Microsoft Project as standard software for schedule tracking to support all PMAs.

Allow more control by user.

NMCI is fundamentally flawed and cannot work. There is no possible way NMCI can function and still meet the security requirements necessary for government-level data processing. Also, the use of large server farms puts too many eggs in one basket in the event of a successful hack or contamination. No system can be 100 percent secure, so why put everything under one lock? This makes NMCI a HUGE target for hackers and terrorist. Say goodbye to NMCI before it's too late!

As far as I can see, NMCI is a major waste of taxpayer money. More government employees are assigned to supporting NMCI than EDS has hired to accomplish their contract. Many important/critical tasks are being left undone in order to expend valuable personnel to support a contract that is unnecessary to the majority of users. I would like to have an accounting of the combination of contract cost to EDS compared to the number of government employees being used to support this contract. I think the result would be very interesting!

I think it was a waste of government funding due to the complexity of IT infrastructure and end user's ability to understand the computing paradigm in general.

Although NMCI will NOT provide the cost savings estimated in the business case, the idea is valid. It will help consolidate disparate networks and get a handle on legacy applications. The planning and aggressive schedule, however, have created extreme cultural resistance. Limited communications do not help overcome these barriers. Perhaps the development of a communication plan would help break down the cultural barriers.

I think the Navy started this without thinking about the long-term problems. There are programs that we use that are no longer available because NMCI does not support the programs.

I believe that setting standards is a good idea. I also have found that standards have been hard to enforce. The good thing about contracting out is it costs to change so we will be forced to stay within standards. But the question is, at what cost? I work on data that is sensitive and contractors should not see. With the unclassified data on an NMCI network the contractor that might be part of the competition has one step up because as administrator they can look at all the data on my desktop.

No.

We were told we would always have the latest version of software. We get no upgrades unless we identify them, buy them and pay NMCI a couple of hundred dollars to load each one.

Great concept, but I think some planners fell short or else falsely projected the costs of this project.

It costs more money, is less useful, inflexible, and very bureaucratic.

NASA did the same thing the Navy is trying to do with NMCI a couple of years ago. They finally realized that one shoe did not fit all. The let their contract lapse and then did the right thing by contracting out to three different vendors for three different functions. The Navy needs to do the same. Re-scope the NMCI contract to handle primarily basic operating support seats only. Then the functions for other specific operational and engineering functions could be contracted out for those specific functions.


inside gcn

  • artificial intelligence (ktsdesign/Shutterstock.com)

    Machine learning with limited data

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group