Note sure why this one year old post is particularly relevant today. It at least needs a (2018) in the title.
1. old, but not a repost should be OK and not really need a date (dates are good when it is a repost or notorious. this is as obscure as it can get)
2. not linking to the original source, but to a insightful discussion instead. I often link to slashdot discussions which contain not one, but 3 or more source on the summary with disparate point of views, plus the slashdot discussion, which is extremely relevant (for hardcore network stuff at least). and the post get deleted by not linking to a original source.
Should I submit the periodic table? Every scientific paper ever posted?
It would waste less time if random posts from a year or older appended the year to the title. Seems very common, but not mentioned in the hacker news guidelines.