Talk:List of dead links
Not sure what remedies to propose
If I'm not sure what remedies to propose, generally speaking, is it still useful for me to add dead links to this list? I am using a link checking tool that turns up lots of dead links, far more than I know what to do with. I can also pass a file of them on to you, David Welbourn, or whoever wants to work on fixing them. -- bg (talk) 23:25, 19 June 2016 (UTC)
- I'm not sure how useful this page is either; perhaps it's too verbose? Like, we probably only have a few limited strategies in dealing with a dead link. I think I had in the back of my mind way back when that this page might be useful for other website authors to tackle linkrot in their pages too, but that's probably wishful thinking beyond the pale. My main concern with dead links now is that there's two types that look identical: (1) normal dead links which might get be fixed someday via Wayback or the content was moved to another site and please someone go look, and (2) very dead links which are really really dead, gone forever, never archived, or robots.txt is blocking it so don't bother trying. Perhaps visually, a very dead link can have a strikethrough markup. Like, do we want a "Template:very dead link" to do that? Is that a reasonable thing to do? -- David Welbourn (talk) 23:58, 19 June 2016 (UTC)
- I wasn't trying to say this page wasn't useful, just that I generally don't know what to propose as a remedy, other than going to the internet archive. The link checker I mentioned automatically looks for internet archive links, and records the url of an archived version, if there is one. I'm happy to share that file with whoever wants it. I could see a "very dead" template being useful. But if it's very dead, do we want a strikethrough, or do we want to make it invisible to the wiki reader? -- bg (talk) 00:06, 20 June 2016 (UTC)
Before we can make good policies about linkrot, we should first list what tools we have to work with. Please list 1) tools that help us discover we have dead links in the first place, 2) our IFWiki templates for marking dead links, 3) special pages on IFWiki that are particularly useful in finding/fixing links, 4) popular external guides and discussion forums for how other wikis and websites deal with linkrot, 5) general search tools including Google and Bing, 6) archive sites including the Wayback Machine at Internet Archive, and 7) any other useful tips not already listed. -- David Welbourn (talk) 07:30, 20 June 2016 (UTC)
- Linkrot finders:
- Link Templates:
- old link
- dead link
- dead ac
- Special Pages:
- Link rot at Wikipedia
- Citing sources: Preventing and repairing dead links at Wikipedia
- Using the Wayback Machine at Wikipedia
- Using WebCite at Wikipedia
- Search Engines:
- Web Archives:
- Other Tips:
- As a preventative measure, explicitly ask the Wayback Machine to make a snapshot of an active site and record the archived page's URL with our link template in case the original page dies later.
- You can add bookmarkets to your browser's bookmarks to easily search for an Internet Archive snapshot of a page or to save a snapshot to the Internet Archive.
- Another place to check for IF resources specifically: the Interactive Fiction Archive
- Pywikibot's weblinkchecker does not list a link as "dead" until it finds it dead at least twice, 7 days apart, to avoid false positives if a website is temporarily down. I don't know if this is standard practice but it seems like a good idea generally to wait and confirm dead links.
- bg has been using the pywikibot tools and has been tweaking the weblinkchecker to skip the links we have already marked "dead" with the dead link template. If anyone else plans to run this script and wants these tweaks please let bg know.
- Automated repair tools:
- Pywikibot's replace.py script does text replacement, and allows for custom fixes that are too complex to do with just regex. Also allows you to view individual fixes and confirm them before editing the page.
- It seems like it should be possible to combine the results of the weblinkchecker tool with custom fixes for the replace.py tool to add archive links in an automated way, if we decide we want to do that.
- InternetArchiveBot is in development at Wikipedia.