Hi all, Echohawk here
I really, really wish that I'd been more conscientious with notes for those dates when I was compiling that list. Of course, the spreadsheet didn't start out as a comprehensive list of D&D products, it started out -- more than two decades ago --- as an index of all D&D monsters, and the product list grew as a side project as I went along, mostly so I could keep track of which products I'd check for monsters.
I do apologize that the file isn't clearer about the dating methodology. The spreadsheet could really use some explanatory text explaining issues like that. But as is usually the case, it started out as a personal project, and I'd never have thought that anyone else would ever be interested in it (at least until I stumbled onto the Acaeum a few years back
).
To confirm though -- if there is a specific day of the month included in the listed release date, then that date should be the exact day the product was released (barring typos). That's the case for most of entries for the last few years, since
WotC has had a very regular monthly release schedule. If there is a month specified, but the day of the month is listed as "-01", then the month is probably correct (based on the methodology TheHistorian mentioned), but might just be an educated guess, especially for earlier products.
In some cases -- including a large number of the earliest products -- I simply wasn't able to narrow the release down down more than the year, and in those cases the date is listed as "yyyy-12-31". (There are a handful of exceptions to that for products in the same series which were all released the same year, which are then "yyyy-12-28", "yyyy-12-29", "yyyy-12-30", etc. so that they appear in the correct series order when the spreadsheet is sorted by date.)
I completely agree that it would have been far more useful to indicate unknown dates with a "?", and I did try that at one stage, but found that it caused havoc when sorting based on the dates. If I were to start that project from scratch again now, I'd definitely have separate columns for the year, month and day, and only include verifiable information. That would deal with both the date-sorting problem and make it clear what date information I was able to confirm.
Anyway, I'm sincerely flattered by the interest in that file, and happy to answer any questions about the dates to the best of my ability. Also, if anyone finds any specific errors with the dates, I'd love to know about them so I can fix them. I still do regular maintenance on the file, although I haven't released an updated version since the end of 2008.