Lincoln Yeoh <lyeoh(at)pop(dot)jaring(dot)my> writes:
> At 05:00 PM 1/26/03 -0500, Tom Lane wrote:
>> (BTW, the reason 'now' without "timestamp" in front works is that this
>> is not a timestamp literal but a text literal, which will be coerced
>> to timestamp at runtime.)
> Is it known at the moment which of those methods the Postgresql team are
> aiming to continue supporting for the near/medium future?
AFAIK there are no plans to break any of them; though certainly
CURRENT_TIMESTAMP is the most future-proof, being spec-mandated.
regards, tom lane