Piece of Shit? I don't think it's nearly as bad as that. I agree that the message which it posts to talk pages after it has supposedly recovered a copy of a dead-linked page in the Wayback Machine is rather off-putting, but as far as I can see, the bot does about as good a job of actually dealing with dead links as one could expect of a relatively simple piece of software.BrillLyle wrote:I really love the Internet Archive and use constantly and am grateful for the Wayback Machine. Like every time I edit Wikipedia.lonza leggiera wrote:There's already a bot which does this (InternetArchiveBot (T-C-L)), and if it can find a copy of the link's former target on the Wayback Machine, it will even replace the dead link with one to the Wayback Machine's copy.
But I have to say that the InternetArchiveBot is a real PoS / waste of time. On the pages I follow there's a notification. It's bloated, confusing, and annoying. ….
Its user page claims it has been designed "to combat link rot", not to ensure that there is none. The only way Wikipedia could do the latter would be to ban the use of external links altogether.…. I suspect also it doesn't serve its purpose, which is making sure there's no linkrot ….
As off-putting as the talk page message is, it seems to me to give an accurate description of what the bot has done, and it does provide links to information—admittedly inadequate, in my opinion—about dealing with link rot. Its instruction to merely "take a moment to review" its edits, however, seems to me to be somewhat problematic. What it means by this is explained in an FAQ, to which the message provides a link: "It means you should check to see if the archive URLs [the bot] added are in working order and/or if the original URLs are dead in the first place." In my opinion, if one of the original URLs really is dead, then merely checking to see if the corresponding archive URL added by the bot is "in working order" is insufficient.… -- and if there is providing resources or guidance on how to fix it. ….
The problem here is that not only do links die, but the contents of the pages they point to can change drastically over even quite short periods of time. Consequently, the page pointed to by the archive URL added by the bot might well be a far from faithful copy of the one pointed to by the original URL when it was added to the article, and so might not support the assertions for which the original page was cited—even if this latter had done so, which, of course, is far from certain in any case.
I used to try and fix dead links quite regularly whenever I ran across them, regardless of whether they had been tagged or not. I rarely do so these days. As for checking archive links added by the InternetArchiveBot, I did this perhaps the first two or three times I ran across one of its talk page messages, but rapidly got sick of doing so, and haven't bothered since.Do any of you go back and fix the links? ….