This illegal non compliance becomes much more common in expensive books, digital-only publishers and in data publications which are commercial.

E.g. think of some very expensive report on data centres aimed at the hedge fund trading market, government procurement datasets, geodemographic data, credit risk databases, bloomburg trading data etc.

If it's published, even if it costs a lot, it should all be viewable for free in the library.

Having the thing at all is valuable. The legislation provides a very limited exemption to IP law to allow archiving. Future legislators can work out how to loosen restrictions if its valuable to do so.

They didn't say but they must mean the items in the identifiers list must be ANDed.

Take the inverse of your case: Anyone can put up a site with content about the UK that isn't hosted in the UK nor owned by anyone living or operating in the UK. That meets one of the stated criteria. Does that give them the right to archive it?

That immediatly just exposes that there are bigger questions we have yet to figure out before we can answer that one. The entire premis of copyright broke a long time ago once the definitions of "copy", "redistribute", and "view" became vague, which happened once "view" started taking place via the technological means of sending a copy to you.

You could always resell a paper book. You could always lend a book that you already read, meaning multiple people get to read the same book.

Could you sell access to your private library?

Could you rent out a book? That's what Blockbuster did, but IDK if they maybe had some extra voluntary private agreements with publishers to keep them from trying to shut them down or inhibit their ability to aquire inventory, seperately from what might have been strictly legal obligation.

So with all of that actually still up in the air, it's hard to say if they have the right to archive anything on the internet that simply references the UK, or even stuff that is primarily about the UK. Many people publicly display things that they claim to retain copyright to, so they would argue that merely existing in public view on the internet does not grant anyone else the right to do anything but view it, and only grudgingly allow that most laws in most places also allow people to retain their own personal copy.

But if I view a public site and cache a copy on my laptop and don't re-publish that anywhere, I may still let 10 people come into my house and look at my screen. Though I think charging money for that is covered, like with bars with a cover charge and pay-per-view sports.

If I can cache any site I can see, and even let anyone else look at my cached copy, then how do you disallow the British Library from doing it?

So maybe in fact they didn't mean for the points to all be ANDed, and maybe in fact they do intend to copy anything from anywhere published by anyone if it merely references the UK in any way?

But if they are ANDed, then merely meeting any one condition, or even meeting all but missing only one, means they don't consider the site to qualify.

All in all it sounds like an underspecced undefined mess.

If states want to legislate internet objects, they first need to very clearly explain how they define the objects which they regard as being within their purview.

Or do they provide a wifi and the “borrower” provides the device? Which then allows the borrower to “illegally” take the content with them?

I wonder whether it would be possible to create a website that could only be accessed after being archived, ie in the BLs reading room!