THE Supreme Court announced a week ago its decision to void theP300-million contract between Smartmatic and the Comelec to diagnose and repair about 80,000 voting machines that will be used in the 2016 elections. This decision granted the petition of the AES Watch and the Integrated Bar of the Philippines questioning the midnight deal and asking the court to stop the contract.
The Precinct Count Optical Scanner (PCOS) machine and the whole automated election system used in the 2010 and 2013 elections have failed to satisfy basic expectations of a voter with regard to the transparency, reliability and verifiability of elections. I wrote in 2010 and 2013 a series of articles pointing out problems and vulnerabilities of the Comelec-Smartmatic PCOS and the automated election system. All of them were points of failure sometime during actual elections and will still hound us in 2016 should we use the same machines.
Yet it is not only in the national elections that we have had problems in automating polls. In the University of the Philippines, the local University Student Council election is computerized. A local computer group, the UP Linux User’s Group (UnPLUG) have created the first version of a computerized election software named Halalan sometime in the year 2009 and it was used in UP Diliman ever since. This year, the same software was used in the UP Manila and UP Los Banos elections.
Before the use of the Halalan software, waiting for the election results from each college was an event in itself with all the fanfare and even covered by the local radio DZUP and the Philippine Collegian. A large tally board is drawn up at the Vinzon’s hall in UP Diliman and numbers are written down for each elective position as certified returns trickle in. Large colleges usually take longer times to count thus pushing the announcement of winners up to midnight.
The promise of the computerization of the UP elections was to simplify the voting process and reduce the time for the results to be tallied. Since it was implemented, it really did give fast tallies since results are now announced at around 10:00 pm. Yet problems still crop up since the software, as with any other piece of code, would have bugs that were not seen in the past.
More so, there are issues regarding the very speed at which it transmits everything. No one can actually verify that their votes were actually counted unless one forces the administrator of the whole system to dump the whole database for you. Maybe not even with that.
Momentary glitches in the Internet connection can also render someone already logged in to lose their vote if the interruption happens before he or she submits his/her ballot. Brownouts, computers crashing, routers getting fried or cables simply being tripped upon can all cause interruptions in the voting process.
Yet more basic is that no one can certify that the precinct counts are actually the same ones being tallied at the central server since it bypasses the poll clerks at the precincts and the College level electoral board who are the ones supposed to affix their signatures as to the veracity of the results. If they cannot see what numbers they have in their college then what are they certifying to?
If there are discrepancies in the results or formal complaints, can we find out where it happened? With the current program, it is difficult if not impossible. Anyone with access to the database is very powerful and thus vulnerable as well. There exists no clear mechanism for recount since there is no paper trail to go back to. Probably only a new set of polls would satisfy such complaints.
I have pointed this out to the UP Office of Student Affairs (OSA) several years back and repeated some of it to the present one but it seems that the operative guide is still a question of speed rather than making the elections transparent. For example, pilot testing the whole system before election day is part of transparency. Having poll watchers for all political parties and candidates present in every step of the electoral process is also all about transparency even if they will simply watch a button being pressed in a website.
If we are to learn something from this local automated elections, it is that using technology in elections is not just about the speed of the tallies. According to Bruce Scheiner, a computer security expert, an electoral system should be able to ensure accuracy (recording the intent of each individual voter) with all the attendant security involved so that no one can add, change destroy votes that were cast already. The system should also preserve voter anonymity and be scalable to large voter populations.
While all praises should be given to the UnPLUG for its hard work and initiative in making the UP Halalan software, the OSA should consider instituting reforms in the local UP elections. It’s not just about producing fast tallies but in putting transparency, fairness and integrity in every step of the way. We cannot, and should not, put the results of any elections, local or national, on just blind faith to software and technology.
The same is true for the national elections and the AES: technology would not sweep away the problems of guns, goons and gold in our elections. Sometimes, keeping the lemons like the PCOS perpetuates the problem rather than help in overhauling the system.