on automated polls: how do we watch the show?

Oct 29, 2009 03:00


I'm really not into political subjects; I prefer to talk about politics only in the social or economic context, and mostly while half-drunk. But here's a try on a mostly politically-motivated subject:

As the title denotes, this entry is all about the process of our upcoming elections. I will not talk about politicians and their platforms, nor abhor any kind of politicking that is happening each day in an exponential rate. Many have discussed about it openly and at great lengths, by people who are most credible at the field.

However, I think not everything has been talked about this new process that our democracy will undergo. Sure, I read in the news that voting machines have already been delivered amid the three thypoons ravaging our dear country, and that an IT company has already been commissioned to exhaust the capabilities of these machines. Also, there have been much legal debates on the logistical capabilities to accommodate such a change in election process. Public debate has been about how ready (or not) we are to have an automated election. But now that it seems final
and unavoidable, I think we need to shift public scrutiny one step ahead of what is to come.

It is, I believe, more important in this upcoming elections to talk about the process than the people in its forefront. At the very least, automated elections and electoral candidates should be met with the same level of attention. Why do I think so? Because at the dawn of election day, the candidates are just candidates. They may be trapo, pragmatists, liberal or conservatives, member of one (but loyal to several) of among the myriad of political parties, but nothing more different than the ideologies they adhere to. Of course they become more important as they execute their mandated constitutional responsibilities, but at this coming elections, they are just constants in the scientific process.

What we need to prove is that it is the process, not the people, which is at fault in a failed elections, and here's a very good opportunity to redeem that. I'm not saying that automated elections will answer all our perennial election-related problems. What I'm saying is that if we do good using automation, then by cancellation we can remove it as one of the factors if the same problems arise. If we need to improve, as a people and as a democracy, then we should not fail an automated elections. If we fail during this process, then there's just more for us to blame in our already seemingly hopeless cry for a politically-stable republic.

For me, in this election, my measure of success is to be able to get this one step forward ending up with more faith in the good of our election process. A measure which is incalculable in all respects, but is attainable with a more systematic approach.

One main point I want to raise is education about the automation process. This is a matter which I think is way overlooked. There has been much correlation done between computer illiteracy of the voting population and the success of automated elections.I think the matter merits the doubt, but not the clout. I dare ask that if technology can make people walk, talk, hear, and feel with amazing precision, why can't it make the dumbest exercise their right to vote? This, I think is trivial. With regards to the feel and touch of technology, we can always copy from those who already have learned by experience.

What I worry about, most of all, is the education of our poll watchers. This we cannot copy at all, thanks to the dynamism and creative minds of our people. As new technology opens new frontiers in human experience, it is, unfortunately, also a power struggle for information. Those who have more information gets a bigger share of what it can do. If it can increase efficiency by a multiplier, then it does so in all aspects. So if automation can deliver election results in near-light speeds, it will do as expected, but without discretion whether results were tampered or not. This is because, at the point of delivery of election results, it can already do little to verify the information it is about to transmit.

I do not know how impermeable this technology we are going to use. Maybe it was better for congress to have pushed through with the 100-million peso incentive for anyone who can hack these machines. But what I can say, being someone from a related field, that for this scope and level of control of operations, it is not far from possible to be able to jinx the elections. Misuse of the machine may cause havoc already. Corruption of data can also happen, although maybe in about one in a billion chances. However, the point is that even if all bases are covered and have been thought about, there are still corner holes that have not been filled up, and that is almost for sure.

That is why education of our poll watchers is crucial, almost central to the success of the automation process. Being able to identify the corner holes and safeguard them is an invaluable task. Leave them what they are today and they will be deceived right under their nose. Deception and manipulation in this upcoming elections will not be apparent. If there is one thing digitizing does, it translates anything into a representation of 1's and 0's, and changing one or several of it from the original can make all the differenct. Think of it as a light switch: how many states can it have but two (on and off) and they are already exact opposites? At the very basics, that's all that has to be done to make your way to the elections.  Worse is that corruption is hard to detect if there is no concrete counterpart of the data. Just like not having a printout of your thesis and then suddenly your computer file has been corrupted.

It is a dilemma, in fact, to expose these corner holes or to keep them as secret as possible. By theory, either can do the job. I don't know the government's method of approach, but I'm not so reassured of the latter. Blame it on my prejudice, but there's just too much power and resources our politicians and will-be politicians have, that keeping it a secret from them is almost a delusion.

If there is one more important point, a cliche would put it more appropriately: time is of the essence. Any technology needs time to be mastered. I remember, typing on the keyboard without looking took me almost a year to master, I wonder how long it would take for voting machines? If we are to educate our poll watchers, then let's do it now. We may not have enough time, but any progress is worth it.

Maybe, we, the youth can spare this election. We are more savvy than those of our advance-aged counterparts. Maybe we have the capacity and the stock knowledge to be able to cope up with this new technology even in a short notice. We have Facebook, Twitter, and Multiply to show that we have a sense of communication. Maybe we, too, can share the knowledge faster and innovate on more constructive ways on how to give this election a shot with much more vigilance. To own this election, as with the transition from the old ways to the new age of election process. To grow as a nation of democracy with faith in its most basic political process.

automated polls, politics, philippines, technology, education, elections

Previous post Next post
Up