While the Egyptian and Mayan scripts were deciphered through perseverance, ingenuity and a bit of serendipity, one of the oldest scripts, the Indus Valley script remains largely undeciphered. Contemporary efforts using computer modeling to crack the code are still in the nascent stage and have already created a lot of furious debate among serious scholars of archaeology, linguistics and history. Asokan-Brahmi is the oldest deciphered script dating to circa 250 BC while the Tamil-Brahmi script follows a couple hundred years later.
A recent discovery of burial urns in Porunthal, Tamil Nadu was dated using accelerator mass-spectrometry (AMS) at a US lab, and found to have an antiquity of 490BC. The burial urns contained skeletons, paddy, numerous beads, and precious gems including steatite. Interestingly the inscriptions on the burial were decoded to read va-y-ra or diamond. If this is correct, then it advances the date of Tamil-Brahmi by 200 years, according to this article in The Hindu.
When Megasthenes visited the court of Chandragupta Maurya around 300 BC, he is said to have observed that they did not have any written script and relied on memory. As mentioned before the earliest decoded script is the Asokan-Brahmi. So this new finding would essentially make Tamil-Brahmi the oldest script. Is Tamil-Brahmi pre-Asokan, asks varnam in a recent post. Not so fast, says Iravatham Mahadevan, a pioneer in the field of Harappan studies who says that we cannot rely on a single data point and that much more such findings are needed to validate this discovery which would imply a paradigm shift in Indian history.
As a side-note, a few years ago I visited the Archaelogical Museum in Puducherry which has a collection of various coins, trinkets and beads from Roman trading around 300 BC in the port city of Arikamedu nearby, including several burial urns. I was looking forward to visiting Arikamedu, but the museum guide said that it was a moot point to go there as all the finds were now housed in the museum!