Vitalik explains "Weak Subjectivity" in this blog here, but it's too difficult to beginners. Can anybody please explain this a bit more easily?
Subjectivity is the idea that there can be multiple potentially correct-looking versions of a particular blockchain ledger, and it's OK if a computer can't automatically choose between them. This is opposed to a blockchain where the correct ledger is knowable by looking at something objective like which chain has the most proof-of-work, or which has the most proof-of-work while conforming to a pre-determined set of validity rules.
Weak subjectivity is the idea that subjectivity is unacceptable for short timespans, but acceptable for long timespans. Specifically, it is proposed that if you have a node that is continuously online, you will be able to automatically determine the correct ledger, but if your node goes offline for many months, you may need to get information from somebody else (eg a friend with a node that was online, or a block explorer) to determine the correct ledger.