Yeah I was gonna say I'll do the same thing and I usually just end up hitting only hte home row keys, and usually my left hand hits keys before the right hand.
There's also the birthday paradox to consider - each new file increases the number of potential matches for all files that follow, and also is a potential match for all files that precede it. The total number of ways to match is something close to the square of the number of files.
So, I have seen people do this, and I gotta ask... what is the plan?
Like you are saving the file, implying that you plan on using it again... why make it so hard to figure out which one it was? Especially in this case, where he has done it to enough files all in the same place that he eventually hit the same combo again.... like, are there hundreds or thousands of randomly named files in that same folder? Why save them if they will be impossible to find again anyway?
I tend to save things with the name embedded, so "<year>-<month>-<day>_<short-description>" and use the date that's relevant for the file, not necessarily when it was created (e.g. due date for homework, tax return due date, etc). If I don't care (e.g. a downloaded image), I just accept whatever the default is.
some websites will save the file name when you upload a file. it could be done for privacy reasons if you don’t want a website to know how you got a file. this is the best reason i can think of for randomizing file names, but i don’t actually do this in practice. sometimes i’ll just rename the file before uploading if im worried about it. file names are just too useful.