>>>>Most people born on February 29th celebrate their birthday on February 28th (don't they?)
>>>
>>>That's OK, but we're trying to determine, how old they will be on March 1st the following year. If they be 1 year and 1 day old, it means that the people born on Feb. 28 should be 1 year and 2 days old. Now, it means that they were 1 year old on 28th and are now 1 year and 2 days on 1st.
>>>
>>>This doesn't make sense, so we may assume that both categories will be 1 year and 1 day. But now where is 1 day difference? The 28th born became suddenly one day younger.
>>>
>>>Do you see a paradox and a problem here?
>>>
>>><g>
>>
>>I honestly don't think there is a resolution to this that is truly satisfactory because there is no standard to define how it works, so everything gets left up to interpretation. I wonder how the government does it when they calculate retirement age for people born on the 28th and 29th of February. Of course, if I know the government, in 2009 when a person born on Feb 29th, 1944 shows up, they'll tell him, "Sorry, call back in 2204".
>
>LOL.
>
>On a serious note, can we (programmers) agree on some interpretation and stick to it?

Well, not us, surely, but maybe the IEEE Software standards? Or maybe there is already a standard that of which we are simply unaware.