Legacy. We are constantly struggling with it, developing stuff for the future while still stuck supporting old stuff in the past. It's why you still have IBM mainframes and Windows 95 on your corporate network.
BUT YOU HAVE NO IDEA.
BUT YOU HAVE NO IDEA.
Let's talk the Ogham characters, for example. This is a legacy alphabet from the 4rth century that survives today on roughly 400 monuments/grave-stones. An example of this text is below:
᚛ᚋᚐᚊ ᚉᚓᚏᚐᚅᚔ ᚐᚃᚔ ᚐᚈᚆᚓᚉᚓᚈᚐᚔᚋᚔᚅ᚜
᚛ᚋᚐᚊ ᚉᚓᚏᚐᚅᚔ ᚐᚃᚔ ᚐᚈᚆᚓᚉᚓᚈᚐᚔᚋᚔᚅ᚜
Your browser/twitter client probably renders this. I mean, it works everywhere I've tried, though I'm sure people will respond with examples where it wasn't rendered correctly. That's legacy for you -- continues to work, but only mostly.
The reason it (mostly) works is because it's one of the many alphabets included with Unicode. Once you've decided that you are going to have a character-set with millions of characters, then it means you are going to have both Ogham characters ᚏ along with emoji like 💩.
Now let's discuss the classic Unix command-line program 'wc' which counts the number of words in a file. Does it work with Ogham? Yes. Both Linux and macOS have completely different code-bases for 'wc', but both work. 
So I'm writing a document using Microsoft Word, copying/pasting this example in. Then I run spellcheck. Word also recognizes what is a "word" surrounded by spaces in that character set. Like 'wc', it sees 4 words. Here the second word is highlighted as mispelled:
And, apparently PowerShell.
So I've written my own version of the 'wc' program using asynchronous state-machine parsers (because reasons). I have to make sure it, too, works, because legacy
