Interesting Links for 16-03-2026

[andrewducker]

1. Royal Mail The Lord of the Rings Stamps

(tags:lotr movies royalmail )

2. UK government to launch £1bn plan to tackle youth unemployment

(tags:unemployment apprentices uk )

3. Excel incorrectly assumes that the year 1900 is a leap year (I've been using it for over 35 years and had no idea)

(tags:excel date history computers )

4. Scientists create first-of-its-kind 'hexagonal diamond' harder than real thing

(tags:diamonds materials )

Comments

Excel incorrectly assumes that the year 1900 is a leap year

[jack]

I vaguely remembered this but hadn't realised it was still true.

What I hadn't realised is that Excel can't represent dates before Jan 1900 at all. It represents dates as number of days since 1900, and apparently it still does. So it's not much different saying it can't represent dates before Mar 1900.

That's why changing the leap year in 1900 would screw up later dates. I'm not sure it's *impossible* to change that. Eg the sheet could have metadata saying which date system it uses, and it could be old-style 1900 (with fake leap year), old-style 1904, or new-style 1900 (with correctly no leap year, and knowing about negative numbers).

Re: Excel incorrectly assumes that the year 1900 is a leap year

[andrewducker]

I didn't realise it didn't support dates before 1900!

Re: Excel incorrectly assumes that the year 1900 is a leap year

[jack]

Now I looked up what google sheets does and I have no idea.

It seems like in Excel you can literally type in a number and change the cell to a date and see what date it thinks it is. I'm not sure I can believe this, but it seems like "1" is "1 Jan 1900", "0" is "0 Jan 1900", and -1 doesn't display.

Google sheets seems to represent dates in a very similar way except that for some unholy reason, it seems like "2" is "1 Jan 1900", 1 is "31 Dec 1899", and 0 is "30 Dec 1899". And it does use negative numbers to represent older dates. It seems to get leap years correct. I assume it uses gregorian dates indefinitely into the past, ignoring any calendar changes etc. But it seems that 2-693,500.00" is "Apr/4/0001". "-693,600.00" is "Dec/25/0000" (which doesn't exist in regular calendars). And "-694,000.00" is "Nov/21/00-1"!

Re: Excel incorrectly assumes that the year 1900 is a leap year

[channelpenguin]

didnt it also at one time not do later than 2035? Or maybe still? I cant recall...

Re: Excel incorrectly assumes that the year 1900 is a leap year

[andrewducker]

Maximum year Excel allows is 10,000, apparently.

[juan_gandhi]

It's not Excel, it's the cultural level of many American C programmers.

[bens_dad]

Ah.
I am transcribing C19 cemetry records and noticed that the dates did not work when I converted the file from Gnumeric to Excel.
My work around is to use a string, which also allows me to preserve the spelling of abbreviations.
However, my colleague says he can enter C19 dates in Excel !

[andrewducker]

Ask him how!

[hilarita]

While formatted as dates not strings? Gosh.