So many good points.
I too have been frustrated by abandoned websites and 404 links. If we are taking about the same thing, my strategy is this: if I find a site of truly interesting material that I fear may be disappear for any reason, I personally archive it with the intent of posting the content on my site if it is ever shut down. Obviously that in itself continues to credit the original author. Sometimes when I find a dead link I will try using http://www.archive.org to locate the site and download it from there to preserve it for posterity in the event the web archive ever goes down. Further, I never said not to copy or post materials on other sites. Indeed, such manuals and inserts are in some sense public material. However, my point here is more than referencing articles found in a library, much less the library itself, for an academic paper (though sometimes the location of the documents is germane to the authenticity of the source). This is more akin to posting artifacts from King Tut's tomb without crediting Howard Carter for the discovery. A bit extreme I know, but we in this forum are like Mac archeologists, each discovering a bit of Apple's past, some of which come at great expense to collector's who obtain things which none of us would otherwise have seen to formulate our own opinions over. At a bare minimum, we each deserve some recognition for re-discovering these items and making them available to the rest of the community. This of course varies from individual to individual. But when in doubt, short of seeking actual permission to utilize materials on our own sites, a simple acknowledgment of where we discovered them just seems common courtesy and the right thing to do. It has nothing to do with a claim, but a collegial nod to our fellow archeologists and respect for the work they have done in finding some of these very rare items. Indeed, the one item from another site you may have only found of interest to include on your site, may be part of a collection of items that one of your visitors would find deeply fascinating.
As for trying to correct erroneous sites, Eric Rasmussen and I were just ruminating on this very point when he expressed his frustration over trying to correct errors in Wikipedia, only to have someone less knowledgeable come along 6 months later and screw it up. When it comes to private sites, however, perhaps the best way to correct the errors is on our own sites that then link to the errors we are correcting. Google will eventually sort it all out. After-all, you can lead a horse to water but you can't make him drink.