There was s series of letters published in one of the daily papers here in Singapore about the apparent anomaly that we do lots of stuff electronically, but voting is not one of them. There was a reply from the Singapore elections department (which reports to the Prime Minister, ie, not an independent commission), that there are issues of trust still unresolved.

This thread was triggered by an article on 23rd November that headlined “Faster ballot counting, e-registration at next GE” (the MSM that it appeared in has a habit of locking up the contents so I have the PDF of the article linked below).

That lead to the first letter by a Mr Lee Kwok Weng on 1st December asking why is the voting not done electronically. That prompted a reply by the Elections Department on 4th December “Online voting still not fool-proof”. There was another reply from Mr Cheng Choon Fei on 6th December “Electronic voting open to errors and fraud”.

Understandably, the not entirely pleased Mr Lee, on 8th December, wrote back and the letter was headlined as “If online banking is acceptable, electronic voting should be too”.

Having not seen the previous letters and article, I thought that I could add to the discussion and wrote in to the MSM. Here’s what I wrote on 8th December and sent to STForum@sph.com.sg:

I would like to respond to a letter published in your forum authored by Mr Lee
Kwok Weng about he being surprised about the replies by the Elections Department
on providing online elections in Singapore. [0]

Mr Lee correctly notes that we do banking transactions and many other types of online transfer of value.

There is a simple reason as to why that works: you can verify it. You can check if monies were sent, or emails were delivered, or digitally encrypted and/or signed documents are decrypted/read only by the intended etc. If the electronic transaction was tampered with you will know.

That is a solved problem.

With election systems, however, the part about tampering is not fully solved. Most voting systems have as their corner stone, the secrecy of the vote. You know what you voted for, but you cannot check that the eventual accounting of the votes were not tampered with – because, in the simple case, you would be revealing your vote. I have always told people what I voted for because I feel voting secrecy encourages the possiblities of fraud, but I can also understand the real threat if someone is compelled to vote in a particular way under duress and threat.

Tamper-resistance and anonymous validation are active areas of research and
some of the thought leaders like David Chaum (www.davidchaum.com) have some
practical ideas that could address them, but issues of trust still remain.

I’d invite Mr Lee to consider attending and engaging with the local tech and cyber security community under the auspices of the Singapore Computer Society. The criticality of the source code of the electronic voting systems being open source, verifiable, rebuildable and repeatable, is but a small step in the direction of trust.

The MSM’s forum editor wrote back saying that they will publish my letter with some edits.

Here’s what was published today, 13th December.