Bob Delaney's Worst Nightmare

Aug 07, 2013

MPP Bob Delaney indicts online voting in The Star (Online voting: the ultimate hackers’ challenge, 1 August) with a series of vivid criticisms. He prefers a tangible, physical foundation for democracy: voters marking ballots in a dedicated polling place.

Declining voter turnout is immune to appeals to tradition and solemnity. Voting online may soon be the only method that encourages any democracy at all.

Electronic democracy is far more robust than Delaney envisions. Debugging is a standard phase of any web development, and stress testing the server can make the election itself seem like a slow day.

Online voting will most certainly include an audit trail, and probably more than one. It is the anonymity of the ballot that needs special consideration. More on that below.

Human factors, by contrast, still pose risks. Some want to snatch the ballot box when they don’t think the polling place is advantageous. Others are happy to direct selected voters to the wrong polls.

Online democracy can make voting more frequent (the “every four years” trope is infuriating). Routine voting makes for boring voting, which makes for honest voting.

Voting from home would ease the byzantine restrictions on polling places and reporting. Delaney, a political party member, disapproves of e-voting under the influence of mass rallies and propaganda; he should skim through Pot v. Kettle.

Delaney wants a tangible foundation under democracy. The birth certificate, the passport, and the ballot are all artifacts of identity, of “showing up”. He sees electronic voting as a vector of impersonation and misrepresentation.

Those documents aren’t exactly “foundation” any more. Digital computing offers a new type of foundation, one that can originate with legislators and not bureaucrats, one that better aligns the technical with the political.

Delaney shouldn’t worry about hackers and ignore the systems administrators. Sysadmins are masters of the computer’s Control Panel. They hold the digital privileges that control identity data or ballot records (or gas plant email archives for that matter).

Most importantly, sysadmins control the audit measures, like logs of transactions, and even their own online roles. With skill, they can manipulate any data they like and leave no trace.

Legislators authorize the bureaucratic programs that require computers, and grant the funds to purchase the systems. It’s ultimately the legislature that entrusts technical privileges to the professionals. From that point, legislators’ trust exceeds their ability to verify.

What if the legislators retained the highest-level technical privileges of those systems? What if the legislature itself worked the central levers of the Control Panel?

Legislators would administer key system policies through electronic majorities, asking only for expert advice and assistance as they do otherwise. The computer systems would take some instructions directly from lawmakers, not a technologist intermediary.

This architecture discourages misdeeds by any one systems administrator, or public servant, or political operative. No individual could expect to overcome the audit trail.

And no one person could violate the anonymity of the ballot, if the legislature itself supervised Elections Ontario’s online voting systems this way. The ballots could be collected, anonymized, and counted without giving any individuals unverifiable access.

It’s Delaney’s worst nightmare: a continuum of secure voting, from the voters to the legislature to the officials. Fragile trust is reinforced with strong encryption, the new foundation of identity (physical forms of identification used today already depend on encrypted elements).

The appropriate user interfaces and architectural improvements are possible now. All that’s necessary is the vision, and the demand to make it happen.

Delaney should take the hammer away from the loom, and consider online voting as a stronger foundation than paper and polling places.

Kirk Zurell investigates digital technology with the University of Waterloo’s REAP lab (, and is an alumni of Leadership Waterloo Region and the Waterloo Voter Support Committee. His opinions are his own, and he welcomes comments at