Now that junk AI content has polluted the public web, access to pre-LLM content has become far more valuable—that’s why Reddit shut down their public APIs too.
I'd recommend everyone to archive their pages through the Internet Archive instead, as that non-profit seems to be more concerned with ethics than corporations such as Google.
True, but two copies is one, and one is none. Multiple backups are critical, especially as archive.org has been targeted by the last few book publishers, who want it gone. As politicians and news sites quietly modify their content and hope nobody notices, this should really be a service of the Library of Congress, too.
i used to look at cached pages all the time. it was particularly useful if the current version of the page was different to Google's cached version, or if the page was down. then the button to open the cached page disappeared without explanation. one of the many ways that Google search now is worse than it was 20 years ago
Also helpful for pages that try to hide parts they let the search engine index behind a paywall when it’s a human visitor. Like the notorious expertsexchange before it got usurped by stackoverflow.
Or getting around geoblocking. Very annoying. It's also unnecessary to shut down everything because some countries have reasonable privacy protection. Don't snoop on your visitors and you have nothing to worry about. I'm sorry the US policy is written by and for marketing tech companies
The really bad thing is that it's only a matter of time before the Internet Archive is sued into oblivion. People are uploading full copyrighted movies and there's no moderation at all. It's not just cacheing that is at risk here either.
Amongst other things, the Internet Archive is the home of the Prelinger Archives, the largest collection of educational, industrial and other ephemeral films from the silent era on. If the IA goes down, the only place to access those would be commercial outlets like YouTube.
IBM didn't "fall". Just because they don't make phones and pop culture shit anymore doesn't mean that International Business Machines are going anywhere soon.
How many companies do you know that make a multitude of processor architectures and a multitude of super computer architectures?
IBM probably has a black site nuclear bomb research lab aimed straight at their competition too
Google never did make backups of the Internet, why are we pretending like they ever did? Cached webpages were a basic workaround for third-party website downtime; a guarantee that you could reliably see the information you searched for, even if the linked site was down. It was nothing more than a snapshot of the webpage their crawlers saw, where older copies are permanently deleted with every new crawl of the page.
It was never an archival effort, it was a rotating cache. If you were under the impression for all these years that Google was preserving Internet history, I don't know why, because Google never claimed to be doing that. Maybe it's time to reevaluate any other altruistic things you're assuming that mega corporations are up to...
That's fine, I don't trust Google anyways. Any real backing up or archiving of internet culture can't be done for profit or it will be shut down when not profitable or enshitified.
There used to be a little link to the side of all(?) or most(?) search results that would allow you to view a cached ver of the result. If you have used Google historically you have probably overlooked it many times. I can't recall them ever loading for me..
It was great, especially for pages with a specific answer. "How do I cook cordon bleu? Here's a result!" 503 Temporarily unavailable? Fuck, but I need it now!!" Clicks Cached button ... voilà.