Probably because it's a few static HTML files that are stored with dozens (if not hundreds) of other single purpose sites, whose infrastructure is managed programmatically. (i.e. they have an intern run a script when they need to migrate it to a new cluster, and that's the extent of it)

A client of mine has a site with 4000+ individual pages, and no content management system.

In the corporate world, you just don't touch stuff if you don't need to. There's no time to go and find web pages that haven't changed in 10 years and figure out what to do about them.

4000 files(?) must be crazy to manage. Is there any kind of templating system at all with them?

Now I come to think of it I guess there may also be a need to keep it in place due to links from other parts of the site.

It isn't so crazy. Imagine that 3900 of those 4000 files never change. Every few years there is a redesign on a section of the site, but some sections haven't changed in 7-8 years.

It's easy to manage a large site that only makes a few changes per week.

>any kind of templating system at all with them? //

I'm going to go ahead and guess search-and-replace is their version of templating?

Why would they take them down?

Besides, it's refreshing to find a site structure from the 1990s that hasn't succumbed to link rot.

I guess they don't even know in which server it's hosted anymore.

It's probably way less effort to just leave it here. It's not like it's using a lot of space, or requires some beefy performance anyway.

I'm assuming the CD cover has the URL printed on it