Joel Spolsky's explanation of Unicode

joelonsoftware.com

11 points·tRAS··6 comments

I have yet to work on a project where neither charsets nor line-endings ever pop-up and cause problems... seeing as it is 2011, this is an absolutely abysmal testimony for IT as a whole.

Microsoft had/has a lot to do with that.

How is it Microsoft's fault that the world couldn't agree on line-termination or charset (especially charset, since there really was no reasonable option before unicode)?

CR/LF. That's why.

That was one of the most popular line ending formats. According to Wikipedia, the following non-MS-related all used CR\LF: DEC TOPS-10, RT-11, CP/M, MP/M, Atari TOS, Symbian OS, Palm OS. Meanwhile Mac OS used just CR, as did the Apple II, Commodore, Acorn BBC, and TRS-80. The only players that used LF were Unix (and derivatives), BeOS, Amiga, and RISC OS.

It just happens that we ended up in a world where basically everything is Unix or Windows, so Windows seems like the odd man out. But history tells a rather different story. The use of CRLF was enshrined in RFC561 (the original email RFC) in 1973, before Unix was popular.

Unicode was originally a 16-bit character set. The only myth is claiming otherwise. There was a 31-bit charset at the same time called ISO 10646. This was not Unicode.

http://en.wikipedia.org/wiki/ISO/IEC_10646#History_of_ISO_10...

Reply on news.ycombinator.com