INSIDER INTERVIEW
Election.com's Mark Strama: No Chads Here
© National Journal Group Inc.
Monday, Nov. 20, 2000

Voters this year have learned more than they ever wanted to about the intricacies of butterfly ballots, chads, recounts and just how long it takes to count 100 million votes. So what comes after punch cards? NationalJournal.com's Julie Samuels talked to Election.com Vice President Mark Strama about the promise and problems with voting online.

Q. Election.com provides online voting technology for both public and private elections. How does an online election work, from the registration through people actually voting?
A. Right now we have two separate systems that we're in the process of integrating on the back end. We facilitate voter registration and absentee ballot requests online, but because the processes still require a physical signature, voters can fill out the voter registration form on our Web site, or fill out a request for an absentee ballot on our Web site, but they still have to sign the physical piece of paper and send it to their county to have their request processed. That's different than the process of voting over the Internet, and at this point in time, those processes are not synched up.

Eventually, we hope that they will be, and we can have an all-electronic process of registering to vote, and requesting to vote absentee, and then even voting absentee over the Internet. The linchpin to being able to do that is the use of digital signatures by consumers, and those just aren't in widespread use at this point in time...

They're doing the best they can do down [in Florida] with technology that was bought on a limited budget and a long time ago.

The voting system is designed to be secure, anonymous and accurate. And, obviously accuracy is a big part of that right now, the idea that electronic voting gives you a lot of precision and eliminates a lot of confusion from the balloting process. The privacy is protected through encryption technology, and the software is incapable of connecting the identity of the voter to the content of the ballot. The ballot is never decrypted until it has been stripped of its identifying characteristics. It is transmitted to us by identifying characteristics that enable us to authenticate that the voter is who he or she says he is.

Q. So when someone actually goes to vote, the technology is set up so they just have enter their PIN number?
A. There's two different scenarios where you would vote over the Internet. One scenario is you would vote on a computer or similar device at a polling place. And in that scenario, there is no challenge to authenticate the voter because they can be authenticated in person by the poll workers. There's no challenge to anonymize the ballot because they don't have to enter any identifying characteristics with the ballot, they don't have to sign the ballot because it will have been authenticated in person. And really, the Internet isn't playing a huge role there, other than for the tabulation of the vote. It's a way of rapidly collecting the vote and counting. But the idea of electronic voting at polling places, on personal computers or some other appliance, is an attractive one now because of the efficiency of counting, because it would be impossible to overvote or to unknowingly undervote, and because it would speed up tabulation a lot.

By the way, I say all that without directing any criticism whatsoever to the Florida or current election process. I really want to be on the record to that effect, because, here's the thing: They're doing the best they can do down there with technology that was bought on a limited budget and a long time ago.... I don't think they've messed up, I just think that it takes patience and time to try to get precision in counting these votes. It's also worth pointing out, 100 million votes were cast and counted in one day, and the election is going to be decided by fewer than 1,000 of those votes. That's like trying to count subatomic particles on a bathroom scale, and it's going to take a little time to get to the bottom of it.

I think that we'll have a lot of time to develop really robust infrastructures before we have 200 million voting online all at once.

Q. How do you think things could have played out differently if certain levels of electronic voting were in place in Florida?
A. You know, there's no telling, because in the near term, you're going to have hybrid elections. Even when you introduce electronic voting, there's still going to be some people voting on paper, and those votes are always going to be tabulated by a different methodology than electronic votes. And neither system is ever going to be perfect. I mean, the user interface on a computer might be more difficult for some voters than the butterfly ballot on a punchcard -- there's just going to be differences.

The electronic vote tabulation will be faster, and it will be accurate, and it will be secure. As jurisdictions review and upgrade their systems, I think they are going to look very seriously at electronic voting.

Q. Why do you think that Internet voting is better than other kinds of electronic voting, touch-screens, etc.?
A. I think that's a simple answer: Because people want to vote on their time and on their terms. If you ask the folks in Oregon, all of whom get to vote by mail, they universally love that system. They really appreciate the convenience of being able to vote on their time and on their terms.... Some people, when they think about Internet voting, and people voting from home, they sort of bemoan the loss of the experience of going to a polling place. But what people in Oregon described to me was sitting around the dinner table talking about elections, sitting together over their ballots and reading through extensive voter guides that are published in Oregon, and a much more deliberative Democratic process there that is permitted by not making people wait in line and hurry through a polling place to vote.

Q. Do you foresee people sitting around the computer, doing the same thing?
A. Yeah, and you know, the great thing about the Internet is that is has content-on-demand. And I really like the idea that people will be able to find out about anything they don't know the answer to while they're thinking about how they want to vote. That doesn't seem so important when you are thinking about presidential elections, but the earliest uses of Internet voting are not going to be in presidential elections, but in small local elections. Those are the elections where we bring the most value because those are the elections that have the lowest turnout, they're the elections that are the most cost-sensitive.

As jurisdictions review and upgrade their systems, I think they are going to look very seriously at electronic voting.

If we can save money in the conduct of a school board election, that's more money that goes to paying teachers and educating kids. They're [also] the elections where information is scarcest for the voters, and so the information on demand that is available on the Internet adds the most value to the voting process. And they're the elections now which in many states are not professionally managed by full-time election administrators, but are often volunteer-managed, where bringing in a trusted third party to help count the votes can add a lot of security to the process.

Q. Will you provide some of the information about the candidates, or will that come from the parties or the campaigns?
A. It's going to depend. For example, in Oregon, and in many states, they send out a voter guide that is established to standards set out by the legislature, in terms of what the content should be. Some of our clients may want us to provide information in our private sector elections; we're often provided biographies of the candidates or descriptions of the issues with pros and cons. It's up to the client, ultimately. It's up to the jurisdiction conducting the election to decide what information they want us to publish in association with the election.

We will never create content ourselves, because there's an editorial bias that goes into the creation of any content, and we can't be a part of that. But, the great thing about the Internet is you can always open a new browser window and start looking for your own content.

Q. You provided online voting services for this year's Arizona primary, and the Voting Integrity Project challenged some of the results...
A. Yes, they challenged it in advance of the election under the Voting Rights Act. And the claim was that if you make it possible for people to vote from home over the Internet, you're diluting the voting strength of people who don't have Internet access.

Q. What's the state of that challenge?
A. It's pending, but the courts and the Justice Department pre-cleared the election. And the primary reason was that in addition to enabling people to vote from home over the Internet, in that election we also enabled anybody, with no excuses, to vote from home by mail. And what we saw in Arizona was a statewide increase in turnout that was as strong in the areas with less Internet access as it was in the areas of more Internet access. And much of that was attributable to the accessibility of vote-by-mail. Vote-by-mail is a very good complement to Internet voting, as long as the digital divide exists.

Q. What about public polling places that provide Internet access?
A. We did have polling places available. What was interesting about that, we had a lot of polling places, over 100 polling places available, on Election Day. Nonetheless, 80 percent of the voters who voted in Arizona chose to vote remotely. Now, of those, half voted over the Internet, half by mail. But, given the choice, the people of Arizona chose to vote on their time and on their terms.

Q. You mentioned the digital divide, and that's something else I wanted to talk about. How do you address that concern?
A. I believe in the near-term, large-scale elections will use the Internet as a supplement to, and not a replacement of, existing voting systems. At least in terms of remote Internet voting.

Q. You foresee Internet voting in the polling places as well?
A. I think that they may also choose to use electronic voting in polling places, and the polling place is accessible to everybody, so that doesn't have any impact on any particular groups. I think that when you introduce remote Internet voting, it should be accompanied by strategies that make voting as accessible to non-Internet users as it is to Internet-users. We did that with vote-by-mail in Arizona. And I also think that a lot of the early adopters of remote Internet voting will be the very small jurisdictions where this is much less of an issue. Because the smaller the jurisdiction, the more homogenous the population.

Q. When you do remote Internet voting, is the time frame still one day?
A. No. That's the great thing about the Internet -- you can extend the voting period at little additional cost. Just like vote by mail occurs over an extended period in advance of Election Day, Internet voting can occur during that time as well.

Q. So you're not worried about logjams and other things like that?

A. Like I say, I think the earliest adopters will be the smaller jurisdictions.... We work extensively with Microsoft to develop very scalable software to very scalable platforms. But, that being said, I think that we'll have a lot of time to develop really robust infrastructures before we have 200 million voting online all at once.

Q. Now if you have extended deadlines, voting that maybe takes a day or two or longer, are you afraid of early leaks?
A. Actually, they found in Oregon that one of the consequences of extending the voting period through vote-by-mail is that it diminishes the effectiveness of last-minute political attacks that are unsubstantiated because you don't know when to time it. So that's one political effect. In terms of it facilitating exit pollsters trying to predict the outcome of the election before the polling is done, I don't think it would be easy for them to do that. I think the best they could do is the same as what they do now, which is try to poll as many people as they can up until the days of the election, but, until the polls close, you don't know how anybody's voted. By the way, we would never decrypt and start tabulating any votes until all the votes are in and the polling is closed.

Q. Another group tried to do an Internet vote in Alaska, and something like 35 people voted. What failed there?
A. I don't know... It's hard to comment on an election you're far removed from. We learned that when a lot of people commented on Arizona that weren't there. But my sense was it was a pretty difficult user interface for the voter. But I wasn't there, so I don't know.

Q. I read that voters in Alaska had to actually download software...
A. Yes, I think they had to load a disc. But I don't know. I'll say this, that was not a binding vote anyway, it was a straw poll.

Q. I've heard some criticism that people are worried about voting becoming too easy, registration being too easy, and you start getting an uniformed electorate. Is there a fear of that at all?
A. Well, there's two separate issues associated with the convenience of the Internet. One is a security concern. If you make it so easy that people can do it multiple times without getting caught, if you make it so easy that it's not secure anymore, that's absolutely a problem. On the other hand, if you make it so easy that suddenly people who don't vote now start voting, that's not a problem. That's a good.

Q. So there's no fear that sometimes you are going to have people that are voting just because they can, not because they know anything about the candidates?
A. No, I don't have any fear of that. You just have got to believe in the people, it's what our country is founded on. And I think that the ability for them to get information when they are voting on the Internet is going to add to the quality of the electorate more than anything.

Q. In Colorado there's a state representative, Lauri Clapp, who's proposing legislation to run Internet voting tests in that state. That's seems to be a baby step start. Are there test runs in other states?
A. There are tests, and committees to study, and all kinds of things going on. We absolutely believe it will evolve incrementally. Like I said, we think the first steps will be absentee voting using the Internet, small jurisdictions using the Internet and electronic voting at polling places.... And I think, like I said, that small jurisdictions that don't run county-wide elections and that aren't professionally administered are the ones that are the most interested and the ones where we have the most value.

Q. We've already touched on things like hackers and computer glitches, but are there any specific things that you are working with now, or do you have any ways to easier explain how to circumvent such problems?
A. The biggest issue that the computer scientists point out about Internet voting is the problem of a virus or a "Trojan horse" on the voter's computer, not so much on the back end of our system, but on the platform from which the voter votes in a remote Internet vote. And we think that concern is a legitimate one that has to be taken very seriously but is much more applicable to large-scale, high-stakes national elections than it is to the size of elections we see using Internet voting in the very near future.

Q. When you talk about 2004, do you see Internet voting playing a role in that election?
A. You know, I think by then it probably will play a role, but despite everybody's desire to look at it in the context of presidential elections, we're just not focused on that. We're focused on smaller elections and things that we can do now. Places where we can add value and where we can start to generate business for the company right now.

Q. What about private organization votes you have done?
A. That's what we have the most experience with, frankly, that's where most of our energy and work is focused right now, is in private sector elections, which is great. It gives us a chance to develop and prove the technology. [And] it gives voters -- many, many voters -- a chance to sample and experience the technology before it's put to use in a large public election.

We did the election for ICAN -- The Internet Corporation for Assigning Names and Numbers -- which is a very large Internet governance group. And that was a neat election because it was all Internet, all electronic voting, five different languages, six continents. And it was a really cool election. We did the Sierra Club election, we do the United Nations Credit Union, we did this very large -- which I am sure you all know about -- we did Youth-E-Vote.... We were the back end to process over 1.3 million votes from K-12 students across the country in elections for president, senator and governor.

Q. How did you get involved with Election.com?
A. I had co-founded a company called NewVoter.com to do the online voter registration that we talked about at the very beginning of this, and that company, before we even really launched, Election.com acquired that company, and that's how I got involved. Before all of this, I was with a nonprofit organization called "Rock the Vote," which works with young people to get them involved with the political process.

Q. How big is Election.com now?
A. About 70 employees, a little over 70 employees.

Q. How about future growth?
A. We have offices in Paris, in U.K.

Q. So this is not just based on American elections?
A. No, in fact, it wouldn't be surprising if the first users of Internet voting in public elections on a large scale were international. We're really active in Western Europe and in Asia Pacific, and we want to be a global company and be a trusted third party to participate in elections all over the world.


[ Insider Interview Archives ]