It's a gregorian calendar. Dates before it was introduced (16th century) don't really make sense. We allow dates between year 1 and the 16th century because it's easier to allow them than disallow them, but negative years would require more support, so it's easier not to support them.

In practice, if you're referring to years that far back, it's much simpler not to use the Gregorian calendar (e.g. use the contemporary calendar of the era), and at that point, you're no longer using the HTML features anyway.

(In reply to comment #1)
> It's a gregorian calendar. Dates before it was introduced (16th century)
> don't really make sense. We allow dates between year 1 and the 16th century
> because it's easier to allow them than disallow them, but negative years
> would require more support, so it's easier not to support them.
> 
> In practice, if you're referring to years that far back, it's much simpler
> not to use the Gregorian calendar (e.g. use the contemporary calendar of the
> era), and at that point, you're no longer using the HTML features anyway.

I disagree. BCE dates are almost always converted to (or interpreted natively as) Gregorian dates, because that is the calendar that was in widespread use when international standardization occurred. In fact, one could even argue that we don't actually use the Gregorian calendar anymore—we use the ISO 8601 calendar, which can represent all dates easily and in the same way.

While I can see why allowing all positive dates would be easier than only allowing dates from the 16th century onwards, I don't see why it would be too big of a burden to also allow negative dates. Surely ISO 8601 is specific enough to describe proper handling?

It doesn't make sense to me to force users of negative dates onto some non-standard (or quasi-standard) system "just because". Allowing negative ISO 8601 dates in the HTML would even allow for easier localized conversion into those contemporary calendars, should the need arise.

[FTR: I was not the anonymous contributor who filed this bug.]