Friday, December 12, 2014

Implement the Automation Law correctly

In fairness to the previous Congresses, we have had our share of model laws. Laws that in other countries are heralded as trailblazing and innovative. Such is Republic Act No. 8436 or the Automated Election System Law.
The AES Law has five operative principles: free, orderly, honest, peaceful, credible and informed elections. In the scheme of things, we already had two automated exercises, 2010 when the aggregate was beyond the number of total registered voters and SMARTMATIC had to tinker with the servers; and 2013, where we had the phenomenon of lines across ballots being read by the PCOS machines and the 60-30-10. The
60-30-10 totals appearing on the postings on the tabulation board of the authorized citizens’ arm had to be removed from the public’s view because, again, the aggregate was going beyond the number of votes transmitted at that time. In both cycles, SMARTMATIC handled the process from end-to-end in 2010 and in tandem with COMELEC in 2013. Which is better? Campaigners placed huge questions on SMARTMATIC insofar as hardware and software are concerned.
Despite the fact that we are automated, there is not much data you can study in determining voter behavior and trends because no one in COMELEC would dare touch the results and do the studies needed to add another layer of science to campaign management in the country.
When electoral results appear to be manipulated, no study can be made due to the GIGO (Garbage In Garbage Out) principle. It would be interesting to note how many of those who voted for BSA3 voted for his Senate candidates in terms of the demographics: gender, age, SES, locale, etc. How many of BSA3’s support went to UNA, etc. One can do correlation and longitudinal studies when the data sets are stable. The same thing goes for VP Binay and Secretary Roxas, both having been official candidates in 2010 and with the VP’s daughter proving that the Binay base was strong in 2013.
The limited studies that came out are often delimited by one assumption, regularity of results. Even problem areas in 2010 have not been compared with that of 2013, whether at the regional, provincial, city or municipal. The Book of Precincts are not even reviewed in terms of hotspots, time and motion, transmission, etc. Imagine if the data set is managed pretty well, one can do polling using the precinct level data and that changes the ballgame altogether.
A candidate can do microtargeting. One can plan using very simple “aggrupations”: for, against, undecided among actual voters per precinct or even polling centers. That makes a big difference from just sourcing it via locality. And with defined data sets, elections will be much cheaper because ads are not a shotgun proposition.
As a result of R.A. 8436, there should be more information-technology capable staff in COMELEC and yet it seems there are so few and that is why they outsource the same during elections which weakens the system. It is a good sign that those appointed as Commissioners today tend to know systems and ICT but the last appointee brings nothing of that sort to the institution. We hope there are more logistic, systems, risks managers being appointed so our electoral system grows with the technological leaps.
A forensic audit must be done after every election cycle. COMELEC just can’t sleep during non-election years because there are quite a number of things to do to bring the institution to a level of aptitude and performance that cannot be realized in 120 days before the scheduled election.
In 2010, various security features were integrated into the system but were not implemented. I remember the PCOS was programmed to inform the voter that the machine has received the ballot and that he/she is Voter No. on the queue. This was removed.
Then the lamp that will determine the veracity of the signature was removed, and so much more controls. These controls would need to be re-installed for 2016 or let’s have a revised set up, and two ballots: local and national. Local should be manual and national automated.
In 2013, the design of the ballot took a while before voters, candidates and political parties got to see and test the same. Interestingly, a pen that blots could destroy the sanctity of a very long ballot. Then there are lines seen on some electronic copies of the ballot because PCOS machines were kept in dusty and humid stockrooms in Laguna that sacrificed the lanyard of the copier; this could have been avoided if only SMARTMATIC did due diligence post 2010, pre-2013 and post-2013.
The AES Law likewise provided for the period in terms of testing the machines, for dry runs on transmissions, for mock polls, for the production of ballots and the number of ballots per precinct that should be on hand in case of replacement and yet in 2013, there were allegations of excess ballots being printed.
Why can’t we stick to the provisions of the law? Why can’t we learn from the problems encountered in 2010 and 2013 to make our elections conform to the operative principles of the law itself?
Why? Because gatekeepers would like to tinker with the machines to play gods.
And since SMARTMATIC owns PCOS, we should hold accountable SMARTMATIC for the mess we are in right now. Per news reports, “Dominion and Smartmatic have settled their litigation in Delaware” and “a new licensing agreement will provide SMARTMATIC with unlimited and perpetual access to their technology.” If that is the case, automation appears to be merely a hardware proposition. This is not a win-win solution for us since hardware become obsolete almost on a quarterly basis.
Think about it, if automation was to way to go to have fast results, we achieved that. How accurate the results were is another thing. But then again if automation is quite good, why was BSA3 proclaimed President on 9 June 2010, while Joseph Estrada was proclaimed earlier on 29 May 1998, in a completely manual balloting?
There are speedy results if there is a landslide and there is mirroring of results based on last survey.
Is Stalin therefore correct when he said: “The people who cast the votes don’t decide an election, the people who count the votes do?” It looks like it after 2 cycles. Let’s brace up for a messy 2016.

No comments: