I only noticed it this morning when I saw the date on a letter but 21st March last week was a palindrome: (if written) 21-3-12. This has got me musing and what was going to be a very short observation and become much more extensive that originally expected.
It seems that we are fast running out of palindromic dates for this decade. At a rough – off the top of my head – estimate, for this year we have left, the 21st of April (21-4-12), May (21-5-12), June (21-6-12), July (21-7-12), August (21-8-12), September (21-9-12) and November (21-11-12) and next year 31st of January (31-1-13), March (31-3-13), May (31-5-13), July (31-7-13) and August (31-8-13). Of these, 21-11-12, 31-1-13 and 31-3-13 seem quite pretty numbers as they use only two different digits.
After 2013, I don't think that the next palindromic date of this written form will not be until 2020: 2nd January 2020 (02-1-20) if 2 is written '02'. However, this seems somewhat artificial and a little too contrived for my liking. The next 'real' palindromic date is, therefore, 12th January 2021: 12-1-21. Following that, the situation is similar to this year with, this time, the 12th of February (12-2-21), March (12-3-21), April (12-4-21), May (12-5-21), June (12-6-21), July (12-7-21), August (12-8-21), September (12-9-21) and November (12-11-21). Of this 12-2-21 and 12-11-21 are perhaps the most appealing for the same reason I gave earlier. However, 12-3-21 has a particular sort of symmetry being the first three digits in ascending and descending order. Of the others, 12-4-21 has the middle digit as the product of all the others and 12-6-21 has the middle digit as the sum of all the others.
After that, we wait until 13-1-31 – when I hope to be 85 (going on 86) years old!
(By the way, 13-1-31 will be a Monday.)