The really hard part of date/time is resolving the issues around location -- resolving the where of
05:12:43 seconds, January 12, 2017
The local granularity of date/time creates a problem of validity. The only thing that can be clear is when the data was written to the "scraping" system. With the asynchronous nature of the web (e.g. caching) we can't even really be sure when it was read for important definitions of "read".
Agreed, that can be truly challenging.
It's important to store the timestamp of when the page source was downloaded, together with the timezone if available.
