My guess: stolen usernames/passwords. Or actual paid subs.

Does that mean they would have to rewrite each site to remove the username specific links and content out of the mirror to keep the origin sites from banning their accounts?

My thoughts exactly. That would require a lot of HTML parsing, plus the risk that any website might watermark the page.

> That would require a lot of HTML parsing

So, clearly undoable then. Drats.

Identify as google-bot

I wondered the same thing, I just figured they were whitelisted by the paywalled sites but I don't know exactly why. Perhaps archive.today has accounts to these paywalled sites although I am not sure that's the case.

There is another site, 12ft.io, that used to bypass many of these site's paywalls but they got pushback and now longer offer bypass for sites like WSJ and NYT.

I mean you could kind of do the same thing locally if you use archivebox. I think it's just the way it scrapes the web page.

I believe the question here is, what is "the way it scrapes"?

I would guess most paywalled sites will allow Bots to scrape content to improve site discoverability. Whether it is Archive.org, Archive.today, Google, Google news...

* * *

Sometimes the paywall is nothing more than a couple HTML div’s.

You can programmatically remove them and then return the page.

Mostly used agents, sites want to be crawled to appear in search results, but from time to time it does actively login to sites (linkedin was one that stopped working, presumably suspended as it likely violates t&c)

NYT has one of those simple paywalls indeed: https://gitlab.com/magnolia1234/bypass-paywalls-chrome-clean...