Recently, I’ve seen people begin to add a digit strings inside the titles of their computer files. These are meant to represent the date on which the document was created, modified, saved, etc. Something like this: awesome_document_20120228.docx. The “20120228” is a numerical representation of February 28, 2012, in YYYY-MM-DD format. In this is great, this is the way that it should be with the year followed by the month followed by the day because that’s how you segment/bucket items if you’re going to organize them by time. Group them all by the year the in which they are created, and within the year you group them by the month. and within the month to group them by the day. If there is an automated program creating these file addendum’s sometimes you’ll even see them get so specific as to get down to the millisecond of when the file was created/modified. 2012-02-28-08-47-34-0897 indicates a file that was modified on February 28, 2012 8:47 AM and 34.0897 seconds.
The thing is you have to follow this convention or else it’s completely useless. For example, if you had a document titled “awesome_document_02272012.docx” and you had versions of this document spanning years of time, if you wanted to arrange your documents by title you would end up grouping all the modifications of the document that you made in any February across all of the years in which you’ve been working on versions of this document. I don’t know about you, but I personally would never want to group all my February’s together. I can’t see a need to do something like that.
So, if you’re going to start inputting eight digit date string, please, please, please, do it with the year first, then follow up with the month, and finally input the date. We don’t writer dates in that manner herein the US, so I can see the tendency not to do that. But grouping things your start with the most general/biggest grouping and proceed down of the smallest.
yyyymmddd or YYYYMMDD or YYYY-MM-DD. Please?