Government Computer News and Association For Federal IRM

Government Computer News and Association For Federal IRM<@VM>Government Computer News and Association For Federal IRM 2000 (cont.)<@VM>Government Computer News and Association For Federal IRM 2000 (cont. 2)<@VM>Government Computer News and Association For Federal IRM 2000 (cont. 3)<@VM>Government Computer News and Association For Federal IRM 2000 (cont. 4)

With the Association for Federal Information Resources Management, GCN recently convened a roundtable of experienced practitioners, vendors and observers to assess long-term trends for federal information technology deployment. Key excerpts appear here.

Participants were Larry Brandt, program manager for digital government at the National Science Foundation; Janet Caldow, director of the IBM Center for Electronic Communities in Washington; G. Edward DeSeve, partner and national industry director for federal services at KPMG Peat Marwick of New York; Harold Gracey Jr., then acting chief information officer at the Veterans Affairs Department; Sallyanne Harper, assistant comptroller general at the General Accounting Office; David Lehman, chief technology officer at Mitre Corp. of Bedford, Mass.; Harris Miller, president of the Information Technology Association of America of Arlington, Va.; and Mary Mitchell, deputy associate commissioner of the General Services Administration. GCN editorial director Thomas R. Temin moderated the roundtable.
To read the whole roundtable transcript, go to and click on GCN Interview under Current Issue.

TEMIN: Welcome to the National Press Club. I am Tom Temin, Editorial Director of GCN, as you know. It is nice to see a lot of familiar faces. There's a couple of people I haven't seen in a while, and some I haven't met. So welcome.

As you know, this event is sponsored by GCN and the Association for Federal IRM, kind of an annual event for them.

TEMIN: Having gotten through 2000, the date code thing, and having an impending change in administration, we thought -- "we" being AFIRM and GCN -- we would want to do something where we could take a blue-sky look at what is ahead beyond all the current turmoil to where long-term trends are going, and what that might mean for policy, and then eventually towards agency missions, and getting further down the inverted pyramid, toward the deployment of information technology.

You all have been chosen because, between Mike, Bob and I, we know you to be smart, intelligent, knowledgeable, and articulate on those topics.

So with that, why don't we begin by just going around the room. Who you are, your background, and what you're doing these days. We'll start to my left with Ed.

DeSEVE: Thanks, Tom.

I am Ed DeSeve. I am the National Industry Director at KPMG, which means I worry about assurance services, consulting services, and tax services that affect both civilian and the DOD side of the federal government.

CALDOW: I'm Janet Caldow. I'm Director of the Institute for Electronic Government at IBM.

We're kind of a think tank within IBM, focused on public sector, government, health care, and education, do a number of research projects each year, host a lot of people through our downtown Washington facility, and focusing on many of the issues that I think you're interested in.

GRACEY: Harold Gracey. I'm the acting CIO at the Department of Veterans Affairs, my third federal agency, thirtieth year of federal service. I'm worrying about delivering service in this shrinking government.

LEHMAN: Dave Lehman, Chief Technology Officer at Mitre. Mitre is a company that runs three federally funded research development centers for the government, one for the IRS, one for the FAA, and one for the Department of Defense.

MILLER: I'm Harris Miller, President of the Information Technology Association of America. We are a national trade association representing about 26,000 IT companies.

We also have a program which focuses on government in our so-called Enterprise Solutions Division, both on federal and state and the local marketplace, and many of the biggest companies that sell into the federal marketplace, as well as the smaller companies, belong.

We also have a leadership role in a couple of areas that I think are relevant today. We did the first study on the IT work force shortage globally, and we're about to produce another one on Monday, which I'll talk about later this morning in terms of the preliminary results from that survey, which I think will blow some of your socks off in terms of how bad the shortages are now.

HARPER: I am Sallyanne Harper. I am the Chief Information Officer and Chief Mission Support Officer for the General Accounting Office for all of 90 days.

I had a prior career at the Environmental Protection Agency, and before that with the Naval Air Systems Command. I am responsible for the management administration functions at the GAO.

BRANDT: I'm Larry Brandt. I run an $8-million-a-year program at the National Science Foundation, called Digital Government. Our aim is to facilitate and support collaborative projects between the university research community and government agencies. And if you don't think that is a culture gap, then -- I've been at NSF since 1976. It's a wonderful place to work.

MS. MITCHELL: I'm Mary Mitchell. I'm the Deputy Associate Administrator for the Office of Electronic Commerce inside of governmentwide policy. It's our job to work across government on a number of things.

We have been actively involved in the Access America programs, particularly for students, as well as things like grants and streamlining the grants process, and other sorts of more foundation work, like the work in public key infrastructure and digital signatures.

Prior to being at GSA, I came from the National Institute of Standards and Technology, primarily working on enabling technology. One of my areas was supply chains, worked with the private sector for a number of years on that topic, and then with the advanced technology program on really emerging R&D, IT solutions.

TEMIN: My opening question is: How deeply can we go or should we go, or maybe is the whole direction wrong toward outsourcing so many of the technical or even managerial functions that career civil servants have done traditionally?

DeSEVE: I think outsourcing is yesterday's news. I think it's an old question. What we're seeing with the Internet and other enabling technologies is that customers are doing the work that used to be done by government clerks, and will continue to do so.

The trend is, and I could use VA as an example, but Harold can do that better than I can. Let's use Social Security as an example. The trend will be to have people interact directly with systems, not go through a set of intermediaries.

So it will not be an outsourcing in the sense of having new clerks replace old clerks, but it will have empowered customers getting information, doing the transactions on the Internet, and in a variety of other ways.

I really believe that that will be the same kind of outsourcing as credit card processing has been to travel payments.

One of the great outsourcing devices was the IMPAC card. The IMPAC card eliminated the need for lots of people who used to do payment processing along the way. So I think we are asking the wrong question when we look at outsourcing, by and large.

Having said that, and I'll shut up in a minute, but with the idea of partnering rather than outsourcing, where the federal government will find ways to work with industry, we have allowed some new procurement models out there.

There's a company that works with the Postal Service. And they bid the Postal Service the free process of change of address. They do the change of address process for the Postal Service for nothing.

Now, is that outsourcing? Well, I think so, but it's not traditional outsourcing. It's not replacing this set of people with this set of people. So I think there are new models for partnering, as well as for individuals acting directly with the federal government, that will make the traditional outsourcing competing work go away.

CALDOW: I would agree. Partnership is really the -- going to be a major trend.

I guess I interpreted outsourcing a little differently. I was recently out in California, and I think I saw five counties, including the one that is right in San Jose and the Silicon Valley, Santa Clara. And a couple of smaller counties around that area, and one of them has an eleventh grader running their network. I mean it gets back to the skill shortage.

There are just not enough skills to go around. It is going to force a lot of rethinking in partnership, I think, between public and private sector if we are really going to move this country forward at the state and federal level.

MILLER: I agree with what Ed said to some extent about redefining the debate, but I think it depends on the type of activity.

If you analogize to some of the jargon that there is in the private sector, you have in the front end the CRMs, those people calling about Customer Relations Management. How does an organization deal with its customers? And those may be literally day-to-day consumers in the business model, or it may be other businesses, but that's under your customer relations.

I think Ed is right on target there. I mean, there is no doubt that customers, including the government, are going to insist that they don't want to go stand in line. They don't want to fill out forms. They don't want to spend time punching through 15 different telephone codes to get to someone to talk to. They are going to want to do everything online.

We are getting closer and closer, to an application service provider you can use for government, so that you can sit down at your PC at home and you could say, "I want to renew my driver's license," you'd punch in your zip code, and you'd say, okay, you're from the state of Maryland, and you'd go to the Maryland Web site, and you'd renew your driver's license.

How do I get my dog's license renewed? Well, say you live in Montgomery County, and you'd go to the Montgomery County Web site. How do I check on my Social Security? It would take you to the Social Security Web site. How do I check on my veterans' benefits? It would take you to a Web site that Harold and his people manage.

On the supply chain management, which is the back side, how do I get my supplies into my business, how do I deal with my supply chain? I think again, it's going to be an e-based approach, because no one is going to want to fill out forms, no one is going to want to accept bids in large documents. I think that is also going to change.

The third part of the equation, the ERP, Enterprise Resource Planning and Enterprise Resource Management, I think that's where there is going to be a slower transition, even though I agree ultimately it is going to be outsourced or partnered, as Ed said, but that's going to be a little slower, because people still want to control the family jewels, whether you are talking about the financial records, or you're talking about the personnel records, whether you're talking about internal sensitive documents.

But now people are turning to the so-called ASP model, the Application Service Provider model, which is very exciting, from the industry's perspective, but also is raising a lot of questions about the willingness of any organization to turn to another company totally outside the organization and say, that's where our financial records are, that's where our personnel records are, that's where all the patent research we're doing, the records are kept.

It's a challenge, and it's a psychological challenge, and I think it's going to be even bigger in the public sector, because there are going to be a lot of agency officials and a lot of members of Congress who are going to say, are we really comfortable saying that medical records of veterans are being handled totally by someone who is not a government employee?

If you did it in a business environment, you'd say, okay, well, once we decide to outsource, we really don't care whether they store the data in Bangalor, or whether they store the data in Northern Ireland, or where they store it, whereas in the government, even if you could convince the government to outsource some of these sensitive data, I think there would be even more sensitivity if you said, "Oh, by the way, our storage records happen to be in Venezuela." People would say, you can't do that. So it is going to be much trickier, I think.

HARPER: But there is limited precedent for that already in the federal government. In terms of the financials, many federal agencies have their financials resident at some other federal agency. They cross-service each other, and they're moving more in that direction, because they can't afford not to.

MILLER: But it is a federal agency. It's not --

HARPER: That's exactly right, but I think where it comes to sensitive records, like veterans' medical records, your tax records being held in Venezuela, clearly, the initial push will be absolutely not under any circumstances, but I think inroads will be made, because the agencies can't afford to maintain their own systems.

DeSEVE: Harris, I have to disagree with you, because right now where are HCFA records?

MILLER: Well, they have been historically held by the private sector for a long time.

DeSEVE: Right. My guess is, and Harold can speak to this, that a lot of veterans' records are in the hands of third-party contractors currently. I don't know that for a fact.

GRACEY: No, not true.

DeSEVE: Social Security would say that many of their contractors hold their records. At HUD we had contractors who held FHA records. So I'm not sure we haven't already, by moving to data center --

GRACEY: Crossed the line, you think?

HARPER: Right.

GRACEY: Well, we have done it in different programs, because the housing program is administered essentially through mortgage bankers, but back to the larger outsourcing point. I think Harris is right on, and when you get to the ERP or ASP issue, you get to where I think the big crisis is coming, and where the line is coming, which is the management capability of the government. We talk all the time about the IT work force shortage, or our inability to recruit people, and how we are going to have to outsource the technical part of the business. The federal government looks like me right now. People are 50, 51, 52 years old. I think it's a great age, by the way, but there are no young people. There are a handful here and there.

We happen to have some that we have recruited into our medical system, for a whole other reason, practitioners, but people who are going to run the departments of the federal government are not in the federal government today. The people who will be managing the federal government ten years from now are not there.

MITCHELL: Well, the set of skills that you need with this outsourcing model, you can't abdicate your responsibility for managing this. That's the problem. That means you have to have top project management skills. That means you have to have business modeling skills. That means you have to have some real strategic thinking going on about what I'm doing, and why I'm doing it, and who are the alliances that I need to build to make these things work.

So yes, you can outsource a lot of lower-level functions, but if you don't have enough in-house talent to manage it effectively, then you are in deep trouble.

GRACEY: I agree.

HARPER: Let me just add to Mary's list, though. The technical skills to oversee these partnerships, because without all the other skills are not going to count.

HARPER: You need to understand what it is that you are managing.

GRACEY: Technology is not going to replace the management function.

MILLER: What that says is you have to change the model of what your government employee looks like. You need 15 people being paid $200,000 to $300,000 a year, and not 100 people being paid $50,000 a year.

GRACEY: You also have to create some kind of new modern intake mechanism to get people in now, so they can grow with the enterprise, to some extent, before they are charged with the financial management, the program management, the outsourcing management, whatever.

MILLER: They'll never pay them more than the congressional salary anyway, so that is a big hurdle to get over.

TEMIN: Which is something I want to get into, is that whole Civil Service reform possibility.

Can the government get by with that little technical knowledge, even to the point of where something as mission critical as delivering battle data can be done by contractors?

BRANDT: We have kind of an interesting model that we are starting at NSF. It's to fill the gap in terms of people who are technically able in the area of security.

What we are doing is creating a scholarship program for undergraduates, where they will focus on that during the course of their school, and when they're finished, if they come to work for the government, we will relieve them of debts associated with that. I think we will be starting that this year.

TEMIN: What are the kinds of skills, I mean specifically, do you think you need to have as employees?

BRANDT: I think one of the problems is being able to look far enough out. It's not surprising, I come from a research organization, but I think the pace of these technologies is going to continue to roll over. You know, it was the Web, and then it was Java, and it will be something else. It's not going to stop.

For example, you're going to have incredible bandwidth, wireless bandwidth available. I just saw somebody the other day talking about a radio modem they had that works at a gigabyte per second.

How do you know about that stuff? How is the government going to see that coming? I think the same problem is there for companies. I wouldn't speak for Mitre, but I imagine one of their issues is how to look far out enough and see when the tidal wave is coming.

LEHMAN: It is not only the matter of looking far enough out at the technology, but how do you design systems today so that they can incorporate tomorrow's technology?

I think that's where the government has the challenge, to have management that understands those issues, so as they outsource to contractors to build their infrastructure, they don't get captured into a particular architecture, but are able to refresh that architecture rapidly over time and start matching the scale of refresh that the commercial world is seeing, you know, the 18-month refresh.

BRANDT: Can I pose a general question to you? Do you see the government, it's a general question, investing in IT at the level that you see in industry, and should we be?

LEHMAN: Well, I think you have a problem. You are going to have citizenry that is going to expect and demand the kind of services from the government that they see in the commercial sector, that kind of responsiveness. So it can lag, but it can't lag to the extent that it's lagging today.

MILLER: Not to be too myopic about it, but look at the Customs' modernization thing going on right now. I mean for years everyone within Customs and all of its customers have been saying, "The system is outmoded. It needs to be replaced."

Congress has been unwilling to appropriate money. The Clinton Administration keeps insisting that the user community should pay additional fees on top of the $1 billion they already pay annually in fees.

Six weeks ago or so, Commissioner Kelly said, "We are going to pull the plug on this Customs modernization program." It was a political ploy, but it was well done. So now it turns out all the Administration is going to put up $12 million.

We are talking about a program for which over four years it is going to cost about $1.2 billion.

You have four huge, expensive teams that have been working on this, have spent tens of millions of dollars. Everyone in Customs says the number of brownouts is going up dramatically.

The cost statements will be enormous, and yet Congress and the administration can't get together on what, frankly, within a system that generates $22 billion in fees a year, should be a fly speck. People think a billion dollars is a lot of money.

DeSEVE: Harris, do you know what we call that? An opportunity. We are bidding with IBM on the Customs' program, and what Customs is asking at the same time they're asking for money from Congress is innovative, creative solutions from the business community. We don't know how we are going to respond yet. We have some ideas about how to do that.There is a fee base out there already to support that. What the Administration has said is surcharge that fee base for the purpose of this. We haven't figured out exactly how to do it yet, but there is a pony in there somewhere, and the government shouldn't necessarily be fronting all of the money.

The government has said this as well in their draft RFP in October, they shouldn't be fronting all of the money. The question is, should they be fronting any of the money? We don't know, and we're all trying to sort that out.

In the IRS, that wasn't possible. It wasn't really possible to think through a model because the taxpayer is the taxpayer. There's an identity there, a mathematical identity. But the people bringing goods and services in are of a different character.

How do we work with them to provide them enhanced services, perhaps that they would be very willing to pay for, beyond what they're getting now, that goes in some way to support the creation of the system, et cetera, et cetera? I don't want to go too far, because we're kind of giving up our competitive strategy.

MILLER: But again, in a bottom line company, this wouldn't even be a debate. The CTO would go to the CFO and say, look, instead of taking four hours to process the paperwork through traditional Customs' forms, we can now process it, it says, in 15 seconds through a transponder, and the savings is "X" in terms of personnel, costs to the government. No-brainer.

But in Congress, this is a huge debate. It may end up not even being funded. In the private sector, it would be over in ten seconds. The CFO would look at the numbers the CTO gave, and the debate would be done.

No one in the Congress or at Customs, and obviously your teams are looking to enhance this value proposition, but no one is debating that this is going to save money to process a load of steel from Canada in fifteen seconds, as opposed to four hours.

DeSEVE: But Harris, what they might also do, if I can take FedEx as an example, you could have it in three days, or you can have it in two days, or you can have it in one day. If you want the enhanced service, you pay for the enhanced service.


DeSEVE: You use that to put more planes in the air, and so on. All we are suggesting is that there is a business model, the government hasn't thought of yet, that has a variety of services available, and you pay differentially for those services as a way to build the systems. They give you 15 seconds, as opposed to 40 response time.

There is something there that we haven't quite decoded yet, but I think that's the trend. Even Harold could probably think of things in the VA that have that same characteristic.

MILLER: In the private sector, the reason I disagree with you to some extent, Ed, and I had this question come up at the hearing, is, when e-trading comes along, you pay less to do a trade with your broker, not more --

DeSEVE: Right.

MILLER: -- because at the end of the day, it is saving the broker money not to have clerks entering the paper order, making a certain number of errors, having to correct it, et cetera, et cetera, et cetera, store all of those paper files. So your transaction costs, if you do online brokerage, is actually less.

DeSEVE: That's another model.


DeSEVE: You could, and I'm not suggesting that we would even propose this, you could outsource that in major areas, and give the government back the benefit of the lower costs. Someone could build the system, lower the costs, give the government back the benefit of lower cost. Again, that is also in the RFP, that idea. That value proposition is in the RFP as well.

TEMIN: I want to ask a follow-on on outsourcing, and maybe Janet can comment on this. I want to cite two recent examples of outsourcing, one successful, one unsuccessful, and something they had in common.

One was the State of Connecticut, which had at that time the largest outsourcing deal ever at the state and local level, a billion dollars with EDS. Because of union considerations and politics, the thing eventually collapsed.

The other is the Army LOGMOD [logistics modernization] contract. In both cases, the deal was that the companies would hire the employees from the government, and they would keep their jobs.

So my question is: If it's so fat and happy in the private sector, where are the savings if the same number of people are now hired by contractors?

CALDOW: Well, some of the savings are in the data farms, because the price per whatever element of computing power goes way down, because you have these huge farms of servers, and that kind of thing.

I'd like to get back to the earlier point, but there are a couple of things, we were going so fast, I don't want to lose from a little bit earlier, and then maybe come back to this point, but when you look at the problem in government, you have to ask yourself, is it the same proportionate problem as the larger society or is it greater, in terms of the aging work force, in terms of the IT skills? Young people are not coming to government.

If you talk to the Kennedy School of Government at Harvard, our number one graduate school in government, 65 percent of their students are international. Harvard is training the next presidents, premiers, prime ministers, and so forth, all over the world, except for here.

In terms of outsourcing, I think you have to look at voluntary outsourcing, whatever forms that takes, big contracts like the Connecticut deal. But there's also a very growing involuntary kind of thing happening, mostly in state and local government. I don't know how it would affect the federal, where government is actually being disintermediated.

You have a lot of dot com companies loaded with venture capital that are happy to come in and grab the heart right out of government, which is the data, the -- somebody said about the crown jewels a little bit earlier, and the government is just starting to wake up to the fact to what they do have, but it's so poorly, financed isn't the right word, but people haven't totally woken up to the fact of what's happening, that companies can come in, sometimes without even any permission needed, and do the dog tags, and pay the parking fines, and things like that. The government doesn't have any choice.

LEHMAN: Or with the case of the IRS, the companies that come in and do electronic filing, and then when the IRS starts talking about people filing directly electronically with the IRS, then say that you're taking business away from private industry.


MILLER: I think to answer the specific question of both the previous speakers, we were involved in lobbying very heavily for both of those. As you know, we almost lost the Army logistics modernization, because five federal employees complained to their congress people, and it almost got killed. Fortunately, cooler heads prevailed, and it went ahead.

But the soft landing issue you raise is, just a political compromise. And the companies realize that if they were hard-nosed about this and said, "We're going to hire people as we need them," and if they happen to be federal -- federal employees, it becomes a political nonstarter.

But the good news is, and where this has been done in other companies, and the UK has been where it's been done the most, and South Australia, and other places, where they're outsourcing, it turns out the soft landing still makes economic sense, because most of the employees that come along tend to be older employees.

There's an attrition rate. Frequently, they end up moving to somewhere else. These tend to be big teams. You're not talking about a small, little company, so if an EDS, or an IBM, or KMPG win a contract, even if the employee doesn't ultimately stay on that particular contract, the employee frequently moves on.

So I think the companies decided that (a), they had to swallow this soft landing as part of the political compromise, but (b), at the end of the day, it wasn't really that bad a deal for them. There were plenty of other ways to achieve the financial benefits of outsourcing without worrying specifically about head count as being the primary issue.

DeSEVE: You answered the other part of the question earlier when you said that only 12 percent of the cost of government is in head count. If LOGMOD works, it's going to save an enormous amount of money by eliminating warehousing, eliminating inventory, eliminating a lot of other things along the way. That's where the savings are going to be.TEMIN: Sally, do you have any perspective from GAO, after 90 days?

HARPER: After 90 days, no, I don't think I have any perspective from GAO, but I would add just from a general government background, the concern that I have is not with the LOGMOD, because I think that that definitely has good possibilities, and I agree with Harris, you can't go there without the political compromise of the soft landing.

The concern that I have is developing and retaining the expertise within government to manage these [contracts]. I know it is a broken record theme, but we're not growing this expertise, we're not staying up with the latest trends in technology. We are undercapitalized, not just for the technology and the turnover that we need in technology, but we're undercapitalized for the human capital aspects of training, development, recruitment, retention, all of the basics.

As Harold has pointed out, the shape of the work force is tending not to have a lot of entry level people. It tends to be an older work force, and that tends to be a more expensive work force to maintain.

MILLER: I don't think that's a bad thing. Again, I think that the old model was, we have to staff it in-house, and so we are really concerned about the absence of young people coming in. That assumes you are doing all this work in-house.

HARPER: I don't agree. No, I think you've mischaracterized my argument. I'm not at all proposing that we do this work in-house. I'm saying, if you contract out this work, which is probably a good thing and more efficient --

MILLER: Right.

HARPER: -- to do, you still need to have expertise --

MILLER: Obviously.

HARPER: -- internally to oversee those mechanisms, and that is where I raise the question, whether the government has, or can retain or develop, an expertise.

MILLER: Not on $125,000 a year, you can't.

HARPER: That's the issue. I don't think we're in disagreement on that.

MITCHELL: You need an open enough environment to not be locked into some solution that's going to age just like our own.

HARPER: Right.

MITCHELL: The other thing is that, you go ahead and you outsource, what is the guarantee that those people have the appropriate skills and knowledge to do that function for you?

The private sector network security infrastructure is certainly no better than what we have in-house, and you can go too far. Australia is a perfect example, where they outsourced all their contract attorneys. One minute they were on one side, [then] the other. You have to be careful that you figure out what are the objectives of government, what is the role of governance, and how do I not compromise the principles that we're supposed to be maintaining.

MILLER: I think a lot of it goes back down, again, to compensation. I mean people like Ed DeSeve shouldn't be allowed to leave the government, but Ed DeSeve is also entitled to earn money comparable to his capabilities'

The CIO job is about 20 percent technical and about 80 percent business. It's just having a good business sense.

Well, those people are even more valuable in the private sector, so to say we're going to max you out at $120,000 or $130,000, and the private sector is paying three or four times that, plus stock options, et cetera, if we want to try to change things, in terms of compensation, I think that's where the compensation changes will be made.

DeSEVE: There's a problem in the private sector if there aren't good people in the public side. When you write an RFP, if you don't understand what you're writing, we're going to get this tome that's about this thick [indicating], impossible to interpret, and you're going to get the wrong proposal.

If you understand what you're doing, it's going to be slick. It's going to be based on best practices. And you're going to understand it or evaluate it easily.

So we really want counterparts who are really, really smart and thoughtful, because it makes our life a whole lot easier if they are.

TEMIN: An observation on your comment: First of all, we know that it's possible to grow tremendous talent in the government - the intellectual and managerial equal of anybody in the private sector.

On the other hand, at IRS, at Commerce, and DOT come to mind immediately, there are CIOs who have come in at that CIO level from really good private sector jobs. Clearly, they're not going to do that the rest of their lives.

Isn't there some kind of model where people come in temporarily? Why can't we build on that more?

GRACEY: As good and exciting as the government is, and the government that we're in today is much better than the government than I came into in 1970, the government is still not a model for fast movement, nimbleness, flexibility and creativity, and dealing with strategic issues in a real-time fashion, which is how we started this conversation.

I agree with almost everything that has been said so far, except where you get to the rub isn't on any particular one of the things that have been listed in the questions or that we've talked about so far.

It's in the system at large, and the system at large is a stultifying, intimidating, clumsy system to operate in. For all that we were talking before we started, for all the modernizations that have been started and stopped, it wasn't the people, the technology, the partnerships, it was the whole ball of wax. It just wouldn't work. It couldn't keep going.

If you sit where I sit, you're looking now at a political change, as you started the meeting with, in a year, where every big system, every exciting thing going on is going to have to be looked at, reexplained, rejustified, and sold internally in order to keep moving, and while that is happening, things slow down.

We're all going to have to work together to find a way to keep things together to make the government more nimble.

HARPER: That said, what I would premise is, that there is still nowhere else that I am aware of that you can take on as much responsibility as you can handle, regardless of what you look like or where you come from, but based on performance.

The government still can track very quickly, which is why we have been able to grow good people. People still come into the government for a mission. They still come into the Environmental Protection Agency to protect the environment. They still come into the Veterans Affairs to serve the veteran.

There is something that money can't buy, although it truly would help a great deal.

I think that there are places in the government where you can be fast, and you can be nimble, and where you can take best practices that will mirror the best that the private sector has to offer. They're small, but they're there. I would suggest capitalizing on those places and the inherent competitive advantage the government has.

I go out and speak at universities and colleges, and you can sell young people today, on a career in government based on those factors.

Can you keep them without the money? No.

TEMIN: You said there were small opportunities. Do any come to mind that you can think of?

HARPER: I think of the Naval Air System Command, a [Malcom F. Baldridge Award winner of a few years ago, had the same type of philosophy. Matrix-managed, very tight, very lean, but they tended to promote pushing people forward, training people, investing in their human capital, and they tended to hold onto their work force.

BRANDT: I work at one of those agencies. We're the same size we were in 1972 in terms of staff. We have a budget that's probably eight or ten times as big.

Management, honest to goodness, is willing for employees to be entrepreneurial, to come forward with good ideas, because that's the whole culture there. You need to stay at the leading edge with what you're doing, and so keep rolling in new ideas.

MILLER: ITAA just completed its annual survey of CIOs. We interviewed 34 CIOs and IRMs in major federal agencies. Several reported that 50 percent of their IT work force would be eligible to retire in the next three years. So I'm not disagreeing that there are still plenty of good people in the government and still plenty of young people who want to join the government, but 50 percent in the next three years.

At the end of the day, you get to the point where you have to pay the college tuition for your kids, or you have some kind of expenses that are coming along, and no matter how much you love your government and how much you love public service, if your next-door neighbor with the same or less experience is making two or three times as much in the private sector, it's just not going to work over the long-term.

Congress is going to have to tackle this issue. They can't keep putting it off. They are going to have to look at making compensation to senior-level managers much more reflective of what is out there in the private sector; otherwise, they are going to be faced with people who can't do the jobs.

GRACEY: I didn't want to leave on a sour note that I think I created. I think Sallyanne is right, but I think we need to broaden that funnel to get more of those folks in, because I think the other end, the drain is getting bigger, and the pace of folks leaving is going to accelerate tremendously.

LEHMAN: You have to create an environment where the government manager is allowed to take a risk, not punished when it doesn't work out, but rewarded for attempting to be innovative.

MITCHELL: I think I agree. I mean compensation is part of it, but there are really other incentives that keep people there. I joined the government when the opportunities for women weren't quite as good as they are today, and the government was an employer of choice. You could get training, you could get responsibilities. These changes in the compensation are going to take a while. We have to look at what some other incentives are, and how we can spread that.

I think working cross-functionally really can excite a work force. There are other creative incentives that we could be employing that we aren't to the extent that we should.

LEHMAN: The biggest replacement for salary is having an impact, feeling that you have done something, and getting some of these obstacles out of the way so you can be agile, so that you can have an impact and say, "I did that."

DeSEVE: The trick is that people are managing careers now. They're not coming for 30 years.

What you want to do is, you want to get a young person for three to five years, get them ingrained, give them a good experience, pay them not quite as much as they could make, and tell them you can then go out and get a higher-level job outside. But let's track that group.

Let's make sure that we keep some nexus between them and the agency, know who they are, alumni meetings, and when they've been there for four or five years, let's bring them back at a higher level than they could attain in their current organization. I call it the ratchet effect, keep moving them up.

So think about how to help these individuals, manage their careers, keep them around, keep them stuck to the agency in one way or another, and use them maybe over their career of thirty years, use them ten to twelve years in the government, ten to twelve to fifteen years outside the government.

That's a different model as well. It requires some more work on our part to keep those folks close in, as we know who they are.

MILLER: Let me give you our new numbers, because I think these are going to surprise people.Our firm that we hired went out and interviewed 700 hiring managers in firms of more than 50 employees. This is only the private sector.

It did the whole broad range of IT skills jobs. We were asking the hiring managers what kinds of skills sets they were looking for.

I won't talk about the skills sets right now, but I'll talk about the numbers. This organization, Market Decisions, based in Oregon, says there are 10 million IT jobs in the private sector, not six or seven million, which has been the number people have been using up until now.

The reason they say that the number is 10 million, not six million or seven million, is in the past, people have used old definitions, they have used DLS definitions, which are no longer current, and number two, they've never gone down to interview smaller firms. Most surveys either do 100 or up, or 250 or up. That's why they're way off. So 10 million IT jobs. There are almost 850,000 vacancies.

HARPER: In the U.S.?

MILLER: In the U.S. It's eight and a half percent.

BRANDT: So how big of a problem does this number, 850,000 people, represent?

MILLER: At any one time, most labor economists will tell you that there are probably four to five percent vacancies in any type of [occupation]. So it's about twice as high.

So [employers] they are very much interested in network administration jobs, but there are shortages in all the occupations. There is nothing that is really in balance right now.

I think the good news is that there are very positive trends out there in terms of education, what the universities are doing, et cetera. But once in a while, you hear stories that you just want to rip your hair out.

I met with the Dean of the University of Arizona Business School in November, when we were out there for a conference. And they've doubled the number of computer science students from 250 to 500, but he can't get any more money from the state or from his financial officer with the university to increase it.

He said the first day of registrations he had a line around the block for people wanting to sign up for courses. It looked like a rock concert.

LEHMAN: There's a built-in five-, six-, seven-year lag in what the educational system produces that's demanded by the economy.

MITCHELL: And where they are willing to cut. They have the same bureaucracy the government does. They have programs that perhaps maybe aren't as needed, and they're not willing to starve those programs to feed the ones that really need the talent.

TEMIN: Well, building on that 850,000 person gap, there is a vacuum pulling talented people from a lot of things, including government. I want to just explore a little bit further those two questions we were on.

One is: Can the government culture, aside from salaries, and I don't think anyone would disagree that that can be a problem, but can the government culture be conducive to attracting the type of people that we think we're talking about?

And secondly, Larry was about to say something about the presumed pace of change that is required in government, because of the presumed pace of change outside of government.

The question is: Is that true? Should government move as fast as we glibly say everything else is moving, or what?

BRANDT: Well, there isn't any question that the technology is going to change, and that pace of change is going to continue to go up in things like how much storage you can fit on your PC, and how much bandwidth you can get walking around with your cell phone, or whatever appliance you have.

That's going to happen, and it isn't going to slow down. The universities are pumping this stuff out, and they're productizing it, a lot of them.

If you talk to many of these computer science professors, they are running a business on the side, or they get a sabbatical and go off and create a business. The university wants the intellectual property rights of this stuff, and at least at NSF we give it to them, because we want someone to have a profit incentive, so we can keep our nose out of it.

I just think industry will ride that curve. They'll be surfing that tidal wave, and I'm afraid the government is going to find themselves having it crash down on top of them.

LEHMAN: The government should ride that curve, if you can find the mechanisms to do it. Where the pace of change in government needs to be a bit slower is in policy, but we're not talking about policy here, we're talking about using information technology that delivers services to the citizens faster, better, cheaper.

MILLER: Unfortunately, you can't make a distinction like that. That's the point Harold was making before.

We like to pretend that somehow you get the policy out here, and then separately is the business administration side of government; and, in fact, they are inextricably linked.

MILLER: Also, the thing you said before, David, is what's great again in the private sector, is that people are allowed to fail. But in the public sector, the price of failure is a Congressional hearing, which your agency is splashed all over the front page of The Washington Post --

MITCHELL: Part of it, I think, is we build these big monolithic systems, okay, that are too big. You need to learn incrementally, you need to do smaller learning exercises so you understand what the issues are, what the problems are before you launch into these huge efforts.

This is really a cultural change, and we don't have this problem alone. The private sector is in the middle of this as well, perhaps with better controls over it.

HARPER: You also need a steady funding stream and that is another difficulty that we face. They're talking about biannual budgeting now. That may or may not come.

TEMIN: Yes. The lack of capital planning or capital budget, and so forth, is probably one of several policies that haven't happened, because of the lack of legal underpinning that is required for these things to happen.

MITCHELL: The whole stovepipe budgeting process [is how] programs are funded. The CIOs need a voice. In many cases, they don't have that voice to put in place the strategy that they think they need to achieve success.

TEMIN: I want to throw in another thought. Recently, Paul Light, from the Brookings Institution, a was assessing the National Performance Review.

[He said] it created a core of highly committed people, it got a lot of people excited. You can argue all day, but there are some things that were accomplished at NPR. Maybe some things didn't meet the promise of NPR, but it was still a reform initiative that has a discreet life to it that you can point to, and yet his contention was that it's out of gas, people have gone as far as they can go with doing more with less, and that whole idea.

There was no fundamental legal or policy underpinnings to NPR. It was just what can we do with the existing way, because at the same time, the budget for the things, from ACEs to you name it, have continued to be squeezed.

GRACEY: I didn't read Paul's argument or characterization, so I wouldn't necessarily say I would disagree with it. What I'd say is, look at the results. I think government is better now than it was eight years ago. Maybe not by a hundred percent, but it's better.

Take our own example. We're treating more patients with fewer staff and less real dollars than we were eight years ago.

DeSEVE: Do you have a strategic plan?

GRACEY: We have a strategic plan.

DeSEVE: Do you have an IT capital plan?

GRACEY: We have an IT capital plan.

DeSEVE: Do you have an IT board that does capital budgeting and programming --

GRACEY: Yes, we do.

DeSEVE: Did you have those eight years ago?


DeSEVE: I can say that about virtually every organization. That stems right from the NPR. That's an NPR legacy that exists.

GRACEY: The government is going to continue to change for the better, because the government's customers are going to demand it, and those of us in government are going to have to deliver it, and we'll find a way to do it. We always have. It's an exciting place to be.

HARPER: It is. I would also add that underpinning the NPR is the Government Performance and Results Act. I don't think we've seen the power that act is going to be able to bring over time. It also drives the strategic plan, it drives linking your budgeting to the strategic plan, and linking those to actual measurable results, not just outputs. What are the outputs of the work that you're doing?

MILLER: I think Sallyanne and the others are right, they just gave a list, it has made a dramatic difference, and whether Gore or Bush gets elected in November, they're going to come back to this issue again and say we have to make government more efficient, and whether they call it NPR or something else, they are going to continue to focus on these issues.

CALDOW: We also need to focus on it globally somewhat. I mean you look at Singapore, who has tackled the compensation issues. If you work for the government, you are paid higher than you are in private corporations. It's the elite. That's where you strive to be. They pick only the best graduate, you know, on and on.

I think are places where policy and the operations rub together, and Customs might be a good example of that, but there are other places where they can go along independently, where policy may need a little deliberation and the time, and the democratic process, and so forth, to work those issues through.

At the same time, government has to learn to move at the speed of business..

BRANDT: You have an atmosphere of budget surpluses now, too, and that might be what could lead to new initiatives. Now that year 2000 has proven you could, now you can go forward, and there's an atmosphere that maybe will entertain those kinds of new ideas.

MILLER: If they don't do it now, they're never going to do it. The federal government and the state governments have the money now, and if they don't want to spend it now, what are they going to do when the economy, heaven forbid, turns down?

MITCHELL: We always focus on costs. Reality is, if we focused, I think, on the velocity, the costs will follow. If we can put some of these changes and make them happen, and show that you have improved the speed of doing government business, you'll be able to track.

MILLER: There's also atmospherics, and that's why I think this government portal is so important, because a lot of what we are talking about in terms of efficiencies in government are issues only people inside the Beltway understand.

The citizens will see the benefits in information technology when they really can go online and do all these things and not have to stand in line.

And going back to the point that Dave was making before about making jobs in the government exciting, to tell a young person, come, put something online that is going to affect millions of citizens is a lot more exciting than saying come help us fix our database operation for the back office.

LEHMAN: I think the issue is that it's an end-to-end system that has to be put into place. It's not just a portal. A portal does not give anybody any agility at all. It has to be tied into the databases and that is going to require a government-wide security structure.

CALDOW: The states are already there. I mean I do probably more work in the state and local area, and once Y2K was out of everybody's hair, I mean you saw state after state, city after city, RFPs hitting even now, doing their enterprise portals.

It's not just a graphic link, but it is something that will integrate into the back end, which requires real transformation in the back end.

MITCHELL: They've also been much more receptive to these flexible financing arrangements.

CALDOW: Yes. They're looking at partnerships. Governors are getting pretty savvy about this.

DeSEVE: I think it's possible, and Mary certainly said this before, I think that people would offer a free government-wide portal to the federal government, if it could accept it, and if they could have the ability then to get the click-through and the ad revenue that would go along with that portal.

You've heard me say that before. I say that generically, as opposed to specifically, but there's no reason the federal government has to build its own portal.

What then happens is, if you build it, they will come, is my theory. If you build a portal, and you have one agency that is capable of doing certain transactions on that portal, other agencies will say, "Well, that's an interesting model. How do they put middleware on top of your existing Legacy database? And then how do they get communications protocols in place to get through that portal?"

"Oh. Well, we could do that. We know how to do that."

So if you build the big portal, and show one agency capable of doing things, the others will migrate in exactly the same fashion, and Internet speed will begin to take over as they do that. That's at least my theory.

The government can do it for free, if they take advantage of the idea that Singapore has, the life events. When you click onto the Singapore model, what you see is travel, and health care, and so on; and embedded in each of those is a set of life transactions. They're not all with the government.

If I click on the passport section of the life events calendar, I ought to also have Expedia right there at the same time. So I can get my passport. I can get my reservations. I can get my rent-a-car. I can get my hotel reservations simultaneously.

Would Expedia agree to host the passport service portion of a government-wide Website? I think so.

HARPER: Oh, I think so.

DeSEVE: Now, Australia is already allowing travel agents to issue visas to Australia. So there's a rethink here. That's the partnership I was talking about earlier. There's a rethink here about how the big portal works together with commercial partners.

Use communities of users as your analogy and build against the community of users rather than against the government processes.

Would Fidelity and Vanguard like to host that? Would Merrill Lynch like to host [Social Security sites?] So that you could come to the SSA site, can get Social Security-authorized information, with your PIN there, and get back the information about your account. But within a site that you could also click on Merrill Lynch.

Would Merrill Lynch allow me to take the data I got off Social Security, put it in a calculator, along with my 401(k) data, to see alternative investment vehicles?

And would Merrill Lynch be willing to pay a fee to have the ability to do that? I think so. I think that's the model.

DeSEVE: Let's say that company "X" and company "Y" built the site for the federal government.

They then go out and they find the private sector company that with banners and click-throughs, have companion [information] to the lifestyle of the individual, have companion services, where they pay revenue to the company who built the Website, which perhaps rebates to the government. In other words, it may not be free; it may be negative costs to the government.

If the government is willing, with appropriate security, to allow this community or a life situation of their customer to be exploited by the private sector -- and I know that's a policy question, so I'll put quotes around that -- I think there's a way to build these things, to build the big portal at no cost or at a negative cost to the federal government.

Now, we may not be able to get over the hurdle of, oh, we can't sell Social Security data to Merrill Lynch. Well, I didn't say that. I said let the individual be exposed to these people who will be willing to pay revenue for that circumstance. I think there's a model there.

BRANDT: Ed was at this electronic government conference the other day, as I was, and we were talking about the problem with the government of trying to impose anything from the top, when you have all of these separate appropriation processes. And there's an encouragement to have stovepipes, and there's a discouragement.

There's very little incentive to work across agencies. NPR tried to do that, and they used their bully pulpit, but they really didn't have much money behind it.

MITCHELL: Well, but you still -- I mean the issue is, you still have to work with the different stakeholder groups, and figure out what's the push, what's the incentive for them to join that partnership and stay in that partnership.

We've made some mistakes in this whole arena of trying to move electronic government services, because we didn't perhaps understand the business drivers of all the stakeholders. That's a hard thing to do, and sometimes it's a very slow-moving policy kind of thing to do, the privacy issue, even if we put the controls in place today to protect and segment.

That's going to take a long time of the citizens getting used to it, figuring out what the benefit is, what's the stake in it for them to do these things. It's not going to happen overnight.

MILLER: No, but it's going to happen very quickly. We all remember the fiasco of the Social Security Administration when they put --

MITCHELL: That was actually a positive move.

MILLER: I know, but the first day, talk about bad luck. A Washington Post reporter goes online, puts in his Social Security number, and gets somebody else's record. I mean you couldn't imagine a worse-case scenario.

I mean there's supposedly better technology and better implementation, but talk about really bad luck. Now, if that had worked, well, it would be a lot different story.

You're asking -- your initial request, you're talking about two or ten years down the road.TEMIN: Yes.

MILLER: Ten years down the road, I hope the census is all done online. I mean it's absurd, all this falafel about whether people are filling out long forms on how many toilets they have.

The real question is: Why does it cost so much to do the census in the year 2000 when 40 percent of homes have PCS that are wired to the Internet? Why aren't people able to do that online? You could probably cut the cost of doing the census by 25 percent to 30 percent.

TEMIN: Moving on, the Presidential memo talked about a single interface or a single government portal. Is there any chance of that ever happening in this decade or the next decade?

BRANDT: GSA has tried it a couple of times. I was talking with the manager of that and e both agreed that it was a manual process right now, and other efforts had been manual processes, and they would never, ever keep up with trying to be current all the time.

So it has to be an automated process, which implies standards, and open interfaces, and an incentive for agencies to live by those standards across the government.

MILLER: I think you're being much too pessimistic. This is going to happen, citizens are going to absolutely demand it. There will be people marching in the streets, and I think this is a two-year, not a ten-year phenomena.

MITCHELL: I don't think it's a ten-year effort either, but yes, I mean there are an emerging set of tools to help us do that.

DeSEVE: It's a very hard technical problem.

MILLER: That goes back to Mary's point about there being small-scale successes at the state level [which] will be easily replicable at the federal level.

CALDOW: I think the market is going to force this all to happen, I think that's true, too, so you sort of get over the mess in the middle, as it's happening, and look and see what's out there in the five years hence. I think for the federal government, the intra-governmental is still going to be very key.

You've already had the states' CIOs setting up their standards for the portals and so forth, and they're not going to go out and bid every application that gets hooked into it by their 70-some departments. They'll let their departments do that, but it will be very easy to plug and play in that environment.

They're already building in the connections down to the local level. So if you're in the city of Raleigh, and you need a certain license, well, is that the city? is it the county? is it the state? or [maybe its] EPA.

If we don't at least begin to think how these things can act now, you're going to have train tracks that don't meet in the three to five years down the road, between the state and whatever state agencies interface with.

MITCHELL: Well, the federal government relies on a huge number of states to administer their programs.

HARPER: Absolutely. But I think the point that Janet is taking us to is, if the federal government doesn't move quickly enough, we will be at the mercy of trying to deal with the state portals, and how they are structured --


HARPER: -- because they aren't going to change.

CALDOW: Yes. You have the government to citizen. That's almost the easiest part. Those are paying the fines, and renewing your real estate license, and all of that kind of thing. That's fairly easy to do.

But when you start getting into the intragovernmental even at the state level, where things go across the state agency, and then up and down, state to federal, and so forth, it gets fairly complicated, and you need the kinds of designs that are going to sustain that.

TEMIN: The generation that is behind us is the so-called echo boom, which is turning out to be a larger group of people than the baby boom was. So you have one reaching its earning power just as one will reach the leading edge of retirement.

What do you think is happening that is driving people's expectations of services generally, getting beyond systems and policy right now, in society?

GRACEY: We just redesigned our Website, and before we did it, we had a contractor --

MILLER: With a larger font probably, right?

GRACEY: And it speaks. We only get one letter per screen. We did focus groups. Because we were concerned that we had a population that may not want to deal with this over the Internet. And we wanted to know if they did, what did they want, and we found a surprise on both counts, or a surprise on the first count, which was overwhelmingly, regardless of era, World War II through current, they want electronic access, they want to use the Web, and they were very consistent about what they wanted, and they want it to be organized around them, not around us.

So where we used to have, click here for Veterans Health Administration, click here for information about veterans benefits, click here for whatever else, we've organized, based on their input, to their kinds of questions, including things outside of VA.

That was a huge surprise for us, that even the World War II veterans, who, by the way, are now outnumbered by Vietnam era veterans, because World War II veterans are dying at about 1,500 a day want to come to us electronically, because it's the way they do their other business. So I just see that driving into the future.

So the conversation we had earlier, you know, the big business driver, once people get used to doing something a good way and love it, they want to do it everywhere, and we better keep up.

MILLER: I think the short answer is instant gratification. People just expect the best of what technology offers them, "Don't put me on hold. Don't tell me I have to fill out a form. Don't tell me I have to stand in line." This is true in the commercial world. People will not wait.

How long does it take the average elevator door to close? Eight seconds. Seven seconds, all of us are pushing that close door button, gotta get that thing closed in seven seconds. That's just the way people are today, and our kids are going to be even worse, and even more demanding.

CALDOW: The bandwidth, I think, is the biggest thing looming ahead. As we move from first-generation sites, which were content only, to transaction ones, where you're actually doing things, renewing licenses and so forth, the next step is going to be truly interactive video.

CALDOW: Think of your own lives. You may do 90 percent of your banking with the ATM machine, through the phone, through the Internet, or whatever. But occasionally you need to talk to a banker. You need to get in and talk face-to-face, and you'll be able to do that on the Internet.

It's not going to be any different than watching TV or cable, seeing these things on Websites.

If you think about it, I mean how many zillions of Web sites are there now in the world? Each one of them has the capability of becoming a TV station.

You're seeing more and more, that's the way that you're going to interact with people, is very rich multimedia, broadband --

HARPER: It raises interesting policy questions, and security questions and privacy questions. It may be one thing for SAS Institute to work with Wal-Mart to take down somebody's click history when they go to Auto Trade.

It's an entirely different issue if it's the federal government, as you're cruising through the federal government, because it raises all kinds of concerns that go to the heart of our culture about, is there a Big Brother?

I don't think that we are positioned to deal with them right now. I just don't think we have people thinking those issues through.

MILLER: I don't know. I think the commercial world, because they are going to be doing the outsourcing, is actually thinking about these privacy issues a lot, and how they are going to handle that.

I think that's one of the reasons information security has become so important to the private sector, not just because they have to serve as the government sector, but because they have to make sure that the customers are confident in these transactions.

LEHMAN: At the same time, the commercial companies are collecting lots of information of where you go, and what you're looking at, for direct marketing, so they can deliver better advertising.

CALDOW: I heard the other day insurance companies now, with satellite, and global positioning, and things, you will get kind of a real-time insurance coverage.

So if you're driving around the back streets of Vienna or Fairfax, it would be a certain rate, but if you're out on the Beltway, it's, boom, up for that period of time, you're on the Beltway. So I mean, think about that. First of all, they know everywhere you're going, everything you're buying.

TEMIN: What are the services people may want? What do they want? Harold, you mentioned that the VA wants online access organized around their lives, and things beyond VA. Can you go a little bit --

MITCHELL: It's the community service. Yes.

GRACEY: Ed, the expression you used earlier caught me, the life experience kind of thing. [Users] described to us, and then we started thinking what else they might want, whether it's Social Security, whether it's Medicare, whatever.

Right now, that's fairly rudimentary in our world. We point them to their site, but ultimately it could be the portal we were talking about earlier.

It could be a citizen portal for all of us. I can't imagine, or I can only imagine, as these kids that are in junior high school and high school now, who have never known life without devices, get to be citizens, what we might be doing, what they might be asking for, what they might be pushing for, or worse, or better. The invisible, almost transparent existence of technology in their lives, in all kinds of ways, raises all kinds of wonderful possibilities, and scary issues about privacy and security.

MILLER: With all due respect, I think those are exaggerated. People talked like that when telephones came in, too.

I think the same thing is true about the Internet. You're not going to buy a house in five years that doesn't come fully wired, with devices in it, because that's just the nature of where we are, the same way you don't buy a house today, and say, "Oh, yes, electricity. I'd like electricity, too."

It just comes with the house. The water comes with the house, the same way the Internet is going to come with the house.

So I respect those people who are raising these kinds of issues, but, again, I think it's an element who just haven't quite adjusted to the reality of the situation.

A small businessperson said to me the other day, "Well, I want to stay off the Internet."

I said, "Well, basically, you're saying you want to go out of business."

I mean that's like someone 50 years ago, a small businessperson saying, "Well, I don't want to have a telephone, don't want to be in the Yellow Pages."

Well, maybe you could get away with that in the 1910 period, maybe the 1920s. You couldn't get away with it in the forties, and you certainly can't get away with it now.

CALDOW: One of the factors we're looking at, again, because we think that the government service delivery, government to citizen, government to business, government to government is going to happen - what are the larger implications of the citizen to the governance kinds of issues. You get into the whole digital democracy piece, and the communications.

Just thinking of yourselves, you probably can handle at most 100, 200 e-mails a day. When you start getting a couple thousand, maybe 5,000, maybe 80,000, or 2 million, because of a story in the morning's paper, if you're a mayor, or a state legislator, or whomever, how are you going to handle that?

One of the early studies said that 80 percent of people on the Web have never written a letter or called an elected official in their lives, but they are now likely to send an e-mail.

So that's coming. Every state legislator I see, who comes to our facility, says, "What are we going to do?"

BRANDT: I think they did a public comment thing on the Web for bioengineered fruits and vegetables, and got 400,000 comments. I asked somebody who was involved in it, how are you going to draw the meaning out of those comments?

Well, we go through them individually, 400,000 of them. That was a hot topic. But as more and more people get used to the Internet, they will be sending comments on whether or not there should be, you know, an extra manhole installed down there on the corner.

MILLER: The computation and linguistics advancements are going to help, so you can go through those 400,000 and summarize them.

BRANDT: And those algorithms are going to have to be well understood --

MILLER: Right.

BRANDT: -- because you're going to have to be able to say to the Congress, yes, we heard those 400,000, and this is what they said, and we have some confidence --

MILLER: Right.

BRANDT: -- that we have figured out what they really wanted to say.

TEMIN: I mean for a group like Harris's, that has a tremendous implication. People complain about lobbyists. On the other hand, you could say five competing lobbyists can give a congressman, or a policymaker, or a mayor the same range of input as 300,000 e-mails, once you've sorted through them.

BRANDT: It's dissenter mediation.

TEMIN: But this way, Harris speaks for a bunch of people, and his opponent speaks for a bunch of people --

BRANDT: Right.

TEMIN: -- and that's not a bad process, really. It may not be perfect or totally inclusive, but is it --

BRANDT: We all place some trust in some part. I mean I read the newspaper and I believe The Washington Post is going to -- I make the assumption they are going to identify the things that are of interest and treat things objectively.

I don't have the time to go tracking down on the Internet every single one of those things and read the source material. We do the same thing with lobbying. I think the interesting thing --

DeSEVE: So you are talking about a digital intermediary. That's what you just suggested.

BRANDT: That's exactly right.

DeSEVE: I haven't thought about it exactly that way.

BRANDT: You need a -- well, we have a library. You go into the library and --

DeSEVE: Not even a library, but an intermediary who has to make choices, who make value judgments along the way.

BRANDT: And you invest and trust in them, based on however you're going to determine that, and at that point, they're the scout for you.

HARPER: But there are data mining -- I mean depending on how the input is brought in, there are data mining techniques that take out those assumptions, if you will, that strictly look at it based on the statistical analyses, and look at the correlations of those types of things.

If you get raw comments in, that's difficult to work through, but there are ways to get comments in the format that would allow structure --

BRANDT: What you lose is, very soon you're only going to hear what the system thinks you want to hear.

MITCHELL: It homogenizes it.

BRANDT: You're going to hear no one that disagrees with you. There will be no real serendipity in terms of things putting themselves in front of you. It will be a real loss.

MILLER: Well, if you're talking about political officials, that's not how political officials get elected. You know, they get elected and reelected, because, in fact, they do have a pretty good antenna.

MILLER: They understand what an organized campaign is, organized by one particular side or the other, as opposed to something that really reflects the constituents. Otherwise, politicians don't get elected in the first place.

I don't think you should think that they're out there all the time. Yes, polls are important, but they don't respond simply to volume and mass.

LEHMAN: It's another factor, and, in fact, they respond to all of those in one way or another, and I don't think that will change. They will still talk to people to understand the nuances of the issue, and they will use the Internet as a way of sampling the population.

CALDOW: Right.

DeSEVE: Do people do that now?

CALDOW: They're starting to.

DeSEVE: That's interesting.

MILLER: Politicians, some of them encourage it. They give out their e-mail addresses, and they encourage people to e-mail them, and others are just deathly scared of it.

DeSEVE: But do they do kind of the USA Today question-of-the-day kind of thing? Do politicians sample their district? Should Elian go home with his father? That's where they become the digital intermediary to some extent.MILLER: Sure they do. They do it all the time. They've been doing that forever. I mean they send out -- every legislator, at least in early term, sends out questionnaires to their constituents --

DeSEVE: Sure. But on the Web?

MILLER: -- which part of it is bragging about how they made the sun rise today and how they're going to make the moon go down tonight. But then they say, "By the way, what do you think of the great job I've done about" -- I mean, yes, they always do that.

DeSEVE: But I'm really thinking about instantaneous access. Let's say there's a floor vote tomorrow on H1B visas. Do they have a Web site that allows people to interactively talk to them about what the H1B visa is?

MILLER: Not if they have any common sense. It's a lose/lose situation, because what it doesn't weigh is the intensity factor.

TEMIN: How does an agency and Congress, let's say an agency, sort out or get around the fact that if you're talking about irradiated milk, or you're talking about asbestos, or you're talking about cigarette smoke, or guns, most people's emotional opinions aren't qualified opinions?

People have had an opinion about this Cuban boy, and yet somewhere there's a prevailing law that says what's supposed to happen.

So the question is the same for digital voting. The Constitution was consciously and deliberately written in such a way as to not have direct election, and direct legislation created by the people, or even by the President.

You can make fun of the Electoral College, but there's a reason for that. Most people don't even know what it is anymore.

E-mail and that kind of direct discussion, or input, is not just a linear extension of paper. It creates a whole new picture of things.

MILLER: Well, federal regulators will have a particular burden, because when they have a draft regulation out there for comment, they are obligated to respond to comments they receive.

MR. TEMIN: Right.

MR. MILLER: So if they get 400,000 e-mails that say don't allow irradiated milk, they don't have to comment to every single one of them, but they do have to say, we got 400,000 comments against irradiated milk.

Now, they may say we discounted those, because they all seem to be exactly the same wording, and they didn't seem to be based on any scientific evidence.

But they do indicate a public concern about the issue, as opposed to the National Science Foundation, which submitted a 45-page technical analysis or -- as opposed to the Milk Producers Association.

But they can't ignore it, whereas politicians theoretically could just say, you know, "It's just an organized campaign by 'X' group or 'Y' group, and I'm just not going to take that into account."

HARPER: Or it's a measure of public concern, and it could be that once you weigh the National Science Foundation or the National Academy of Sciences against the level of concern evidenced in public comment, it tells you you have to take certain steps if you're going to proceed down a regulatory trail.

You have to take education steps, you have to begin to hold town meetings to address the concerns that are raised. I see that as a positive for our form of government. I think an informed citizenry is essential to a democracy, and that you may have uninformed comments, but nonetheless, it's a pointer that what you're dealing with in a regulatory environment, you have issues that need to be addressed.

LEHMAN: It's change. There's a tremendous amount of change that is going on now, and the business practice of change management is known, and it needs to be applied to these kinds of things, and one of the things it has to be applied to is more volume of input from the citizenry.

TEMIN: Lately, the distributed denial of service attack has been much in the news. Why couldn't you use that same technique to -- you can launch 700,000 e-mails from people who are absolutely convinced that the Air Force has little green men on ice out there.

DeSEVE: During Roswell.

TEMIN: Roswell. We'll give them all Palm-VIIs to phone home. So I'm saying if this is what people want --

DeSEVE: I didn't realize this was a classified meeting.

LEHMAN: You said that so knowingly, Ed.

MILLER: The point you made a minute ago, Tom, I certainly wouldn't advocate saying just because there's now something called the Internet that we reject the republican form of government and go back to basic Athenian democracy. I know some people are talking about that, but I don't think -- that's a much more fundamental political philosophy question.

I mean James Madison would turn over in his grave if you said the elected officials are no longer relevant, they merely report on what Internet poll they took yesterday, and will simply add up the results.

Theoretically, it does become possible, if every single person is online. We will start having a national referendum on every issue. That would be a pretty scary change, in my mind.

It would basically reject the philosophy that was in the Federalist papers, that, as you said, says you don't want the emotion of the moment, you want to have a republican form of government, with checks and balances, and that would clearly be rejecting the idea of checks and balances. It would say whatever the emotion of the moment is, that's what you decide.

MITCHELL: I view it as a positive thing. We've seen a declining interest in voting and a declining interest in participating in the democratic society, and it's kind of all about trust.

This offers a new avenue to continue that dialogue and perhaps build a greater form of trust in the government than there has been in recent years.

DeSEVE: Well, yes and no, Mary. If we can't process what's being sent, then that could increase distrust in the short-term. If people think that they are sending a communication that isn't being received and dealt with, then they could become even more cynical.

MILLER: I think the attack on the Census form is a classic example. Every ten years, when there's a census, somebody complains about it, but because we now have 24-hour news, and CNN has to fill those news slots, and now because we have the Internet, the sort of anti-census people are getting a lot bigger hearing than they've ever gotten before, to the point now where the Speaker of the House has to stand up and say please fill out your Census form.

The President had to say please fill out your Census form. I mean that's the downside of this kind of ability of the Internet, that these people are communicating.

GRACEY: I think over time it's going to be self-limiting. I mean it's like what Larry said about not going out and reading every source news story because you don't have the time.

In the beginning, people might take advantage of the ability to send e-mail, but over time, it's going to kind of fade as it becomes so routine, you would only do it if it were life or death, or a personally compelling situation, I think.

LEHMAN: It depends on how the government manages that, and what kind of impact they think that kind of e-mail is having, and then it's no different than the industry's customer relationship management.

MITCHELL: Exactly. I mean yes, there is a place in government for marketing. Clear communications, perhaps, is a better word. We're adopting more and more commercial practices to manage.

LEHMAN: I think that's where the other conversation ended off just before the break. It's this notion that -- I used to run into people all the time who talked about the government as unique, business practices didn't apply, and I think, in the last ten years, we've seen a tremendous growth in government, recognizing that they are no different from business, and they can run their organizations like a business. They don't have stockholders holding them accountable.

DeSEVE: I only slightly disagree. I think if we can begin using the word, there are management principles, there are fundamental management principles that apply. They don't have to be business principles, or nonprofit principles, or government principles, but managing government properly, there's a way to do that.

TEMIN: Let's bridge this over to how agencies are going to be organized. If we have a population that is largely online and wants direct service, what are agencies going to look like?

Secondly, will we still need clerks and some infrastructure of walk-in, because will there ever not be a so-called digital divide?

GRACEY: I think we should go back to what Harris said about the telephone. I mean depending on whether we all show our true ages, there are different pieces of technology that you remember encountering for the first time, and I think by the time Vietnam veterans are 65, that everyone may have some device, just like literally almost everyone now has a telephone, and expects dial tone kind of service for whenever they pick it up and use it.

What's going to be interesting is whether this is the thing that drives federal horizontal integration. States have it, clearly.

We have these stovepipes in part because of three branches of governments, and those checks and balances, and there are stakeholders all around us on those, but can this drive it? Can this break down some of those barriers? Will it be citizen-oriented government?

LEHMAN: I think that the opportunity with the Internet is to drive these [agencies] to be organized towards the services they provide to the citizenry. I mean the IRS is already reorganized from an internally driven organization to a customer-driven organization.

HARPER: I agree with both David and Harold, and I think it will drive more horizontal organizations, simply because if someone is coming into the EPA and they want information about their zip code, and where they live, or where they're thinking of moving, they don't want to have to go to a separate place to get the water information, the state information, the local town's information, the air information, what industries are located there, what do they discharge. They want to go to one place, one click, and they want it all laid out for them right there.

Well, currently, at the EPA, we're not structured in that way, the databases weren't [so] structured. They've made a lot of progress in moving in that direction. They're going to demand it.

The customers are going to demand it, and that's going to drive it, and that's going to drive the breaking down of those stovepipes.

LEHMAN: But we need more efforts, something like the open GIS consortium --


LEHMAN: -- which is looking with industry at the standards for exchanging information from state and local levels, all the way up through the government, but that's a set of data. It's not all the data that people are interested in.

MITCHELL: You'll say that, and so will Social Security. When they talked to these focus groups across the country, they didn't view it as the federal government and the -- they viewed it as a government.

LEHMAN: Exactly.

GRACEY: I guess my view is, this is new, because we did focus groups years ago and said, what don't we like about the service we provided, and the answer was, you asked me for information that is government information.

HARPER: That you've already got.

GRACEY: You asked me for what I think, as a customer, you already have. We asked you for your service medical records, for your discharge papers from the military. That's government stuff. Even worse, we go to the VA hospital, you ask us for our VA benefits information, why don't you have it?

We're not going to get away with that anymore, and we shouldn't. It's going to become Social Security, Medicare, whatever. People are going to say, I'm 65, I'm retiring, I was in the service, what are the range of things that should be delivered to me as a citizen, and how do I get to them, and they're going to want to go one place and have instant gratification.

BRANDT: Aren't there laws that prevent agencies from sharing certain kinds of data with others, and aren't we going to have to overcome those?

GRACEY: Or define how they share them.

MITCHELL: Well, it's kind of an opt in. Citizens need to be willing to opt in, and they may. You know, you talk about what are people willing to give up from a privacy perspective for other incentives, you know, but that's going to be a very careful line when --

LEHMAN: Will the CIO Council across government be able to drive the commonality infrastructure required, the data structures required to support that?

GRACEY: I personally think that the technology is moving in a way that makes that easier.

LEHMAN: Right. That's true.

GRACEY: So I think it won't be as daunting a task as it was 20 years ago.

TEMIN: Harold, let me ask you a question that's kind of germane. We've been talking a lot about what people really want, or need, is the delivery of the service, and not the clerks and the infrastructure. Is the VA in the business to deliver health care or to operate hospitals?

GRACEY: Deliver health care.

TEMIN: Looking out ten years, do we have 283 VA hospitals?

GRACEY: We've already moved in the direction of a health care system and away from a hospital-based system.

Five years ago, had a million inpatients a year. This year we have 500,000, but we had 35 million outpatient episodes of care, compared to 25 million five years ago, and we're moving to leased space and clinics, contracting.

That makes the handling of the information even more important, because anywhere the person goes to get health care, the provider needs to know, to do the best job, everything that is in the system at large about that patient.

So I just see it becoming -- fuzzing the lines between government private sector and becoming health care overall. That raises the issue of where does the information go. Maybe the patient carries the information with him --

DeSEVE: On a Smart Card.

GRACEY: -- on a Smart Card, or some other device that we haven't thought of yet. Maybe they control who gets to use what pieces of that information.

TEMIN: Doesn't that get also to the issue of the need for managerial expertise? The District of Columbia outsources all these social services, but there is no oversight of any of them, so it becomes a true human disaster.

So yes, it's expensive and inefficient in some ways to operate those hospitals. On the other hand, you do have your own people and your own controls, and whatever issues there might have been with the VA hospitals, it's a different managerial issue than outsourcing.

GRACEY: Yes, and I think that applies across the government. I think that goes back to the very first conversation we had today, which is the creation and maintenance of highly skilled management folks in the government, technical managers, financial managers, program managers, clinical managers. Pick one. We're going to have to keep that core.

CALDOW: Aren't you also kind of leading the private sector in terms of medical records? I mean talk about aging population, what the problems are, compounded with, is it 200 people a day die from medical mistakes, many of which are information-based.

GRACEY: Because we're a big system and we now have electronic entry of prescriptions.

MILLER: Of course, the health care industry is particularly bad in terms of using IT, where, as we all know, they spend less a percentage than any other major market sector, and, of course, they're paying for it, or we're paying for it when we use the medical system, I guess.

The only place I can get a handwritten receipt in this town is when I go to my HMO, when I pay my co-pay, and they handwrite the receipt every time. It's just -- I want to throttle somebody every time that happens.

LEHMAN: And I think I get a bill for it, even though I've paid it.

GRACEY: That's a different issue.

TEMIN: Okay. So ten years from now, there's probably fewer VA hospitals, as we have understood them since the thirties and forties.

DeSEVE: We don't know that. See, you've just made an assumption. I don't know if that's true. In particular markets, the VA hospital may thrive and become the only hospital. How does VA manage things? Well, we let the market manage it, for goodness sakes.

Why does VA have to decide what the where a veteran goes? It doesn't. The veteran has a card. The veteran goes where the veteran wants to go. We may have to set standards, we may have to do quality assurance along the way, but if the veteran's hospital in Philadelphia wants to succeed, it's right next to the University of Pennsylvania Hospital, maybe advises the University of Pennsylvania Hospital. We just don't know. What we're going to have is a much more market-driven health care system, where the individual will make decisions, and the facilities will respond to those decisions.

TEMIN: Let me ask this one. In a shopping plaza in Maryland, there's a dinky storefront that's a Social Security office, and there's, I forget how many of those --

MITCHELL: Thirteen hundred.

TEMIN: Thirteen hundred. That's right. Will that be there in ten years?

MILLER: There's always going to have to be direct government services available for people at the bottom of the socioeconomic ladder, who are getting those economic services. Even if you could get them connected electronically, there's a certain amount of personal hand-holding and delivery of services to people who are economically, socially, physically challenged. So there's always going to be that. Are there going to be 1,300 of them, though?

DeSEVE: Are they going to be multi-service centers, single-purpose centers?

MILLER: Right. Exactly.

DeSEVE: I don't know.

MITCHELL: [I went through the] Social Security experience with a parent. You no longer walk in and sign up for Social Security, you call in, and they've already converted you. Yes, you might walk in, but then you call in an appointment to do that.

That's going to migrate from call-in centers to interactive sessions where you're connected to a call center when you need it, and that's just going to evolve.

HARPER: I doubt it would be just be for Social Security. Whatever your delivery mechanisms are for that entity, why not state and local at the same time?

You may be able to access everything that you need as a citizen's service center, from the dog license site, to the Social Security site, to your passport.

MITCHELL: You are already doing that with some of the DOD and VA sites, and I see that as a natural progression to this more of a [multi-service] center.

DeSEVE: And maybe the Postal Service will run all of them.

MITCHELL: It has some real implications.

GRACEY: They're going to have to bid it out.

GRACEY: Or the school system could do some of it.

DeSEVE: The school systems. The Postal Service theoretically is going to have to be in place as a government agency anyway, so why don't they run the social service information at the same time?

TEMIN: Implicit in this, in many cases, is that government is the provider of last resort. I mean even though we have the most cars and so forth around here, we still want buses, because sometimes there are people who don't have cars. So the same is true of the post office. If we want to send something across town, we use a courier or FedEx, but that's not an economic model for everyone.

MILLER: The thing we have been jumping back and forth on, because we're in a changing time horizon, certainly, in ten years, maybe in four or five years, first of all, the access devices for many people are not going to be PCS.

MILLER: Frequently, they're going to be, as everyone was suggesting before, the television or the telephone, or some other device we haven't figured out. Much of this is going to be voice activated.

MILLER: Yes. So the idea that somehow you have to physically have a PC and sit down at it, and type things, and point and click, everyone realizes is probably the hand crank era for cars in the 1910 era. I mean it's a very important technology right now, but I think very few people think it's going to last much longer.

LEHMAN: In 2003, there will be more wireless connections to the Internet than there will be PC connections. At least, that is the projection.

GRACEY: I think to get at your question, you'd almost have to drop back and look at what -- especially in the ten-year window, what's the whole environment going to look like? How are people going to be working? What are they going to be doing?

What is the transportation going to be like, either because of improvements and technology, or because of the environment and just limitations, physical limitations? What's going to happen with fossil fuel? What are the effects of all this going to be?

DeSEVE: Let's take an example. Why shouldn't we assume that we will be in the office two days a week for four hours to conduct face-to-face meetings, and the rest of the time we don't want you there, because we don't have enough space for you, because our hotel is smaller. We want you working at home.

The expectation for knowledge workers who can network, in any event, maybe only periodically must they come and touch each other. The rest of the time, they are expected to use the network because it's a much more efficient way.

They can use their home and the Internet, and the PC connection, whatever it is, or the Web connection, whatever it is. I think that's what's going to happen; I really do. It's going to be telecommuting more.

LEHMAN: Well, I think there is a whole area of research there, and the kinds of relationships that you can build using collaboration technology over the Internet, there have been many successes there, but there have also been areas where it just doesn't work.

The face-to-face, the power of presence, the power of personality does not come across. It is filtered out through those techniques.

TEMIN: Well, here's another question, too. Again, maybe it relates to us more as government workers than as service providers, but what are the implications of that? Doesn't that assume, like other scenarios make assumptions, that everyone has a home environment suitable to work in? I think that's a huge assumption.

HARPER: I think that there's another assumption that we don't want to make, which is that it's one size fits all. I think there are certain things that don't lend themselves to working in a home environment, and I think that certain touch services are going to always be needed.

DeSEVE: Absolutely.

HARPER: [And there are] those who don't have a home environment that they want to work in. There are some people who need boundaries, for whatever reasons, between their home and their work world.

There would have to be adjustments made for those. There are companies doing that. They cycle people through on certain schedules.

LEHMAN: Not to mention just having young children at home.

HARPER: Right. Exactly.

DeSEVE: Or a refrigerator upstairs.

HARPER: So I think just recognizing that it's not necessarily a one-size-fits-all future, too, is important.

MILLER: But I'm an old-fashioned manager. I want my staff in my office, and I know I'm part of that last generation, but I want to be able to see my staff, and talk to them. I'm not real comfortable with them telecommuting.

DeSEVE: But you're going to retire.

MILLER: Exactly. The people coming in behind me are going to have a much different attitude.

TEMIN: Getting back to service, here's another, again, blue-sky question, ten years out. Today, when I get a fishing license, I go to Sports Authority, or a similar large box store, and they sell the fishing licenses. In West Virginia, you can get them in Sheetz gas stations.

I can't get my driver's license there, can't get Social Security, stamps, et cetera. Ten years, five years'

DeSEVE: Two years.

TEMIN: ' I can go to Wal-Mart a

DeSEVE: Two years. Not Wal-Mart. You're going to be able to get it on the Internet. I mean you can get a fishing license on the Internet now, or on the phone. You can get a Florida fishing license on the phone now.

MITCHELL: You can get your driver's license.

HARPER: You can do your postage over the Net.

DeSEVE: You can get your driver's license on the phone.

TEMIN: But you mentioned, Ed, earlier, that if Social Security data were on a site hosted by someone with ancillary services that could provide information, why wouldn't a stationery store want to sell stamps right next to the vellum paper.

DeSEVE: Well, they do and can. I mean they can do that now.

TEMIN: Or car dealers sell a license, or something?

HARPER: They should be able to. You have to have
your protections in place. You have to have clear understandings of who has the data and what it's to be used for, but --

DeSEVE: You could, if you wanted to, right now, buy a car over the Internet, have the car delivered to your door with the registration sticker in place, and an insurance card in the glove compartment, without ever leaving home. There's no reason you can't do that.

MILLER: Except in some states where the car dealers are trying to fight back and try to prevent that.


DeSEVE: It would take you a little bit to organize those three people to work together, but there's no reason you can't do that.

TEMIN: What I'm asking is, just as we talk about hosted portals that are online, what about hosted real places, where you could get any one of several government services. The other places are open to 10:00 p.m.

MITCHELL: Again, it's back to the trust issue. Do you trust them to be your agent? Are they going to abide by the rules that you've --

DeSEVE: Well, no, you may not trust them to be your agent, but do you trust them to have access to your network? That's all. That's all they have to have.

MITCHELL: That's the same thing.

DeSEVE: I haven't really researched yet, but GovWorks says, We'll get you a dog license, a fishing license, a hunting license, a driver's license. Just call us up, and we'll get that for you, and we'll send it to you.

Now, they can't do it everywhere. They'll also register your complaint on a pothole. So they're a commercial portal to government, as opposed to a government portal to government.

You will be able to get access. The point is, you won't need to go to Wal-Mart to get all those things. If you're at Wal-Mart, either you can take your cell phone out, or Wal-Mart will be able to go through a computer network to get that thing for you if you need it.TEMIN: We have a few minutes remaining. And one question I'd like to ask is what does the data and privacy issue mean for us, going forward?

MILLER: Well, for one thing, I think there are different groups of people. I think there are private advocates who are in the business of making noise about privacy, but let's leave them out of the equation.

Just the average person -- I don't think consumers are stupid. I think when they walk into a mall and it says, "Fill out this form and you win the possibility to win a free vacation to Hawaii," they understand that they are putting that data down for a reason, that they're going to get some kind of solicitation, magazine, et cetera.

Similarly, I think when people go on the Internet and make purchases, just as when they pick up the phone and dial L.L. Bean or some other direct marketing company that's selling, they understand that data is going into a database.

Nevertheless, I think both the business community and government have to do a better job of educating people that the power of information technology, Internet, does mean that what they've experienced before may be ratcheted up, and they have to be even smarter and more judicious consumers.

So I think the argument that somehow consumers, when they go online, don't know what they're doing is a fallacious assumption.

HARPER: I agree, but the indicative fury that sticks in my mind is when one of the state licensing agencies released, without drivers' knowledge, the data associated with that, and all hell broke loose, because there was not a matter of choice there and there was not a matter of notification. People didn't realize that if they went for a driver's license, it would be sold to whomever wanted the data.

MILLER: That's a misuse of data. I agree.

LEHMAN: There is a difference between the information. Do I really care whether somebody knows what I'm shopping for, and in some cases, you can imagine cases where you might.

But for the most part, no, at a department store, but the government does have information which the citizenry would like to have access to online, but it's a private matter, it's a financial matter.

BRANDT: Well, it's different in the government, too, because a lot of this information you have to provide to the government is not voluntary. It's mandatory.

LEHMAN: So there's a privacy issue there, and I think there's even the notion of identity protection, as things go online.

That may be an inherently government function, as protection of identity, as we may move into a infrastructure, where everybody has some kind of identification, but they don't want that stolen.

GRACEY: I think it's hard to judge, I don't know where you draw the line between what you would call privacy advocates and regular folks, because you see pushback, or you did when you were asked to use your Social Security in any number of places that weren't official, or weren't Social Security.

And we struggled with it for years, and used a different number, and then we converted it. It was a big issue, at least of conversation.

So I guess it's the question of bringing our 525-member board of directors, and the folks that they represent, along with the concept that we can be trusted.

Then you have the issue of the general citizenry trust of government. You're going to have to stand up to a higher standard, I think, than Nordstrom does, or American Express does, or Safeway does. So I think in the short run, it's going to be problematic.

In the long run, I think I agree with Harris, that it's going to calm down once it's seen and generally accepted, that this new way of doing business is okay, and it's not massively threatening in any personal way.

CALDOW: I think a lot of it comes down to consent. You can use this data for the purpose for which I gave it to you, not to be combined with six other things that you've done, data mining, and come up with something totally different that I would never have expected. So I think that's going to be a big part.

MILLER: But it contradicts something we were talking about a few minutes ago, which is the need within government, for some percentage of citizens, in effect, don't want to have to tell three different agencies exactly the same information over and over again.

CALDOW: Well, so in that case, when they give their address, it is consent for multiple use. So I think they need to know, be an informed --

MITCHELL: We are sort of a me generation, where -- yes, for convenience we want that to happen, but if they're matching -- if I have an outstanding student loan that I've paid, and I'm a government employee, I don't want them matching for that purpose.

DeSEVE: They're already doing it, Mary.

MITCHELL: I know, I'm just
saying -- I graduated too long ago for it to be a valid example anyway.

BRANDT: So her IRS refund is going to be reduced, because of --

MITCHELL: Right. It is sort of a two-faced -- we want it when it works to our advantage, but we don't want it --

BRANDT: It's okay to get those other bums, but not us.

MILLER: When I was on the state lottery board in Virginia, we had the classic case. This guy didn't want to go in and claim the money because he knew he had a little bit of taxes due, so he got his buddy to do it, and his buddy owed much more.

TEMIN: Another question I want to go around the table on is: Just give me three sentences per person on what will electronic government look like by 2010, sky eye view. Who wants to go first? Then we're to go clockwise from there.

MILLER: Well, electronic government will be everywhere. People will expect all of their government services, say, for things like law enforcement, and even some of that will be done electronically. It will be much easier to do, and there will be a lot of integration among the various levels of government.

HARPER: There will need to be statutory changes to support that, and infrastructure to support that, but I would agree with what Harris said.

BRANDT: I expect that the shrinkage of government will continue, and we will continue to get rid of the intermediary between us and the citizen.

MITCHELL: We are going to pay a lot more attention to what the customers, in this case, our citizens and the businesses who interact with us, want. It is going to be horizontal.

It's going to look across government, not just federal government, but state and local government, and essentially take sort of the start of what we've seen in projects like Access America, and transfer that to deliver services to specific communities.

DeSEVE: It will be transparent. We won't know it's government. We'll just be getting something that we need as part of our life. It will be life integrated. I guess it's kind of the same thing. If we need to go to Australia, we will get our passport, our visa, our tickets, our rent-a-car and our hotel in a package.

It will be packaged in such a way that it's part of our life, and the government part is just another stamp along the way, another thing on my ticket. In fact, my e-ticket may have my visa in it. It may be a machine-readable visa that's part of my e-ticket. It's another set of the bar codes when my e-ticket comes back to me.

So it will be integrated with our life. It will be transparent. It will be ubiquitous. But the other side of the government is that the problems we haven't solved yet, whether they're privacy problems, or data protection problems, or whatever, will become more and more the focus of government.

Government will be more the standard setter, so when we talk about electronic government, it won't be as much service delivery as problem resolution entity. That will become a more major role.

Monitoring the telecom folks, deciding who gets bandwidth, where they get the bandwidth, how we charge for bandwidth, and the whole issue of how we sold bandwidth five years ago may have to be revisited.

CALDOW: I think that ten years from today we'll probably smile at how we tried to describe it today.

DeSEVE: Well put.

TEMIN: I'll have the paper transcript somewhere in the file.

CALDOW: But it's almost like trying to describe ten years in advance what an industrial age government going to look like. With the kind of scope that we're dealing with, and it's important that people take a broad perspective right now, don't limit it to one aspect. So that's how I'd sum it up.

GRACEY: I think it will be personal, that one size won't fit all, so some folks will choose one route to get their service, some will chose another, but I want to build on what Ed said, which is it will be transparent and part of the larger transaction, and maybe in some cases, automatic, that when you conduct a transaction, whether it's getting paid, or whatever, your tax and Social Security things will happen automatically, as will your tax return, et cetera.

LEHMAN: I agree with what's been said. That's the trouble with being last, but I think that the changes required in the capitalization of the systems and the forces to be brought to bear on the different agencies to cooperate, and the acquisition techniques to acquire this system and keep it up-to-date, and incrementally improving it, as industry does, are being invented and yet to be invented.

TEMIN: Okay. And our final question, is, what keeps you up at night in this regard, in this milieu?

LEHMAN: Well, in the near term, I think it's the skill base to keep this moving, and hiring the people. I think that the major issue, as we've talked about these integrated services, is the motivations for bringing the agencies together so that they do cooperate on common standards, common infrastructures, where it's going to be necessary, such as insecurity, agreeing on the semantic meanings of the data elements of the databases, which is a tremendous problem, there are technologies out there that help solve.

GRACEY: Number one is having a 16-year-old daughter who just started driving. That doesn't count, I don't think.

In short-term, security. I have some significant concerns that we worked so hard to network things together that weren't intended to be networked together, that we have some serious vulnerability.

CALDOW: We haven't gotten to the point where we're world citizens yet, and because of that, and because of the distribution of wealth, and how your market portfolio is going to do, and what's the future of your kids, and who is going to have a job, and what kind of job, it's very critical the government do this at federal, state, and local levels all over the world, because it will be the fundamental foundation of the distribution of wealth for the next four or five decades, to whatever is after the digital age.

DeSEVE: [I'm worried about] the gap between expectations and the ability to deliver.

We talk a lot about how big systems projects are supposed to work, we still haven't done well delivering those big systems projects, whichever ones they are, as a group.

We have some great ideas about enterprise resource planning this, and customer relations management that, and so on, but I'm not sure, until we start delivering that stuff, whether it's all going to work the way we intend it to work, and I think that can create a continuing distrust of government.

MITCHELL: A couple of things. I think [about] the ability to come up with the investment capital to make these changes, and how to turn off some of the old stuff so we have that investment capital coming together to solve common problems across agencies. We're only, what, some percentage of the marketplace, so leveraging what's going on in the private sector, and not forgetting the standards and those sorts of things are incredibly important.

That is the route to improve these things. It's not by working amongst just federal agencies. You have to leverage what's going on out there in the private sector.

BRANDT: I have sort of, I think, two [worries]. One is, I have a job that requires me to have some understanding of all of the kinds of IT research that's going on.

That's very hard to do, without just dabbling, so I spend a lot of time trying to make sure that, my b.s. factor is good, so that if someone runs an idea in front of me, I can get a sense of whether it's possible or not. So that's a fairly narrow programmatic management function.

The other thing I'm trying to do is do a good job of proselytizing the notion that agencies need, as a matter of normal business, to do a better job of looking out where the technologies are going, and the university research community can help them do that. That's also a very hard job, and there's no rule book that tells me how to do that.

HARPER: Mine fall into two areas, both of which have already been mentioned. The first would be, short-term, the government's capacity to make intelligent IT investment decisions for the long term.

But the second is more long-term, and I think it's going to be an enduring problem, and that is the tremendous potential for malevolent mischief, given the technologies that we will be basing ourselves on in the future, and whether we're smart enough to deal with those on a real-time basis without trammeling on the very things that we support as a republic.

MILLER: In terms of sort of short-term, specific challenges, I guess it could go back to what David said, I think the work force issue is a very serious challenge, not just for the government, but for the private sector, and the security issue.

The information security industry has accepted -- the IT industry has accepted the right to be just as important as other elements of the infrastructure, such as transportation, telecommunications and financial services.

We have not accepted the responsibility that goes along with that, to make sure that the best stuff is available seven by twenty four, and not every eight-year-old kid or every terrorist can get in and destroy that system. So that's why we are spending so much time on information security issues.

LEHMAN: But the information security issue extends to many of the things you mentioned. It's not just the Internet that's capable of being attacked, but the electric systems, the telephone systems, long-haul sewers and pipelines are all dependent on information technology.

They are being done with RF signals, easy to penetrate, and it's something that we're only becoming aware of now, and trying to drive industry to take steps to protect those themselves. It's an important issue.

MILLER: The good news is we did it [for year 2000]. We can probably do it here, but it's going to take a lot more attention from the senior people in the corporate world and the government.

TEMIN: I want to thank everybody on behalf of AFFIRM, and on behalf of GCN.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.