With less than 100 days left before the May 2010 elections, the implementation of the automated election system or AES of the Commission on Elections (Comelec) continues to tread in dangerous waters. Volunteers from the Computer Professionals Union (CPU) witnessed the field test of the Precinct Count Optical Scan (PCOS) in Pateros and Taguig on January 29. We are sharing the following excerpts from the CPU volunteers’ report.

___

“At exactly 9:00 am, the Comelec commenced the test proper. We were given ten pre-shaded ballots to feed to the PCOS machines. We did not see the marker that was used in shading the ballots. The ballots themselves are to be the same size that will be used in the elections but the list of names are not the present candidates. The feeding of ballots took 30 seconds each, and 8 copies of local and national returns were printed afterwards.

“By 9:14 am, the Comelec attempted to transmit the returns to the Municipal Board of Canvassers (MBOC) server. They tried to send the data three times but failed. We noticed that all of them, even the representative from Smartmatic, seemed at a loss on how to handle these problems. Four of them were talking to different people on their phones inquiring for solutions.

“A Comelec staff said that the problem is from the MBOC side because the canvassing server is still not ready but other observers from the Center for People’s Empowerment in Governance present at the canvassing servers pointed out otherwise.

“While waiting, we received news that four out of ten ballots from the Aguho Elementary School was rejected by the PCOS machine. They were also suffering from the same transmission problem. The Comelec officer at our test center couldn’t answer what the possible causes are for this.

“After an hour of trying, the transmission of the returns finally succeeded but only to two of the servers: the central server and the KBP parties server. It failed to transmit data to the municipal canvassing server.
Twenty-two copies of national and local returns were printed in over 30 minutes. The final attempt to transmit the data to the MBOC server became successful, though other testing centers continued to experience difficulties in transmitting.”

___

It would have been something else if this happened only in Pateros Elementary School. But it appears that there were other sites that also failed to transmit their data. The 5,000 satellite broadband global area network machines that Smartmatic procured might be inadequate for the 2010 elections since if even right here in Metro Manila, signal transmissions via cell phone signals fail.

In succeeding demonstrations and tests, even a valid ballot that was fed into the PCOS machine was rejected. Comelec officials said that the ballot may have been crumpled but the Commission still has to give clear explanations for the field tests. A 40 percent failure rate translates into at least 15 million disenfranchised voters. That high error rate is unacceptable in any industrial quality control situation.

In addition to the field test failures, the Comelec also noted how the reported cell phone jammers that were imported could be used to derail the elections. If it was really prepared in implementing the automated elections properly, it could do a thorough site survey for each of the polling places before and during elections with frequency scanners to detect possible signal interference. Knowing that this type of possible interference can happen, they should have prepared for contingencies.

A signal jammer is a consumer device readily available in the market. It is being used in churches and other premises. Even a simple UHF (ultra-high frequency) radio when tuned to the same frequency used by mobile phones can already interfere with the transmission. The importation of cell phone jammers should not be a problem if the Commission had prepared for it. They should be the first ones to realize that the powerful elite and existing ruling clique will be using different tricks—including jamming of transmissions—just to keep themselves in power.

We have previously pointed out that the Comelec’s very short timeline to implement the automation will result in an Automated Election System (AES) that lacks adequate and rigorous testing. Machines do break and fail randomly and software will always contain bugs. These could only be minimized if these are subjected to rigorous testing and review before using it in a critical situation. One or two field tests would just not cut it.

The danger signs are there and we have to be vigilant to prevent a massive failure in the elections. The AES in itself will not eliminate cheating and fraud in the 2010 elections. Using technology to speed up voting and canvassing process becomes relevant and useful only if the people behind the technology are credible and will serve to protect the interests of the population in the elections.