I remember browsing this site before Unicode was even a vague plan. The march of charsets has not been kind to the site. But if you use that as an excuse to sneer and pass over it, you will be missing out on a true unsung gem. At least, if you like math. This is the sort of site that the web was created for in the first place.
I recommend "Reflections on Relativity" if you need a concrete starting place. Here on HN over the years I have wished it would be published at various times, but I am now happy to say I have a physical copy, despite having read it online at least three times before. It was worth it just to see it typeset properly.
Then perhaps rather I should say before anybody knew about it. Fair enough. The world was not running on Unicode until well after this site was established.
Windows NT's first release was 3.1 from 1993. Windows 2000 (which was really NT 5.0) was released in 2000.
I was fully able to write both Gödel and Poincaré with both 8859-1 and HTML entities and in both WordPerfect and Word back in the 90's (the current version of the page is generated by Microsoft Word 10).
The problem is not unicode but the fact that it was written by an American/someone who is rather sloppy with names. For example, he writes "Paul Von Hindenberg" when he means "Paul von Hindenburg". Things like that are trivial to get right unless one actually doesn't care.
I recommend "Reflections on Relativity" if you need a concrete starting place. Here on HN over the years I have wished it would be published at various times, but I am now happy to say I have a physical copy, despite having read it online at least three times before. It was worth it just to see it typeset properly.