Chapter 2, starting at Chapter 2, discusses how to properly cite sources, as well as what sources are acceptable. For each existing source in the article, go through a three-step process to determine if it’s salvageable and, if so, improve it. This section goes through the three options in detail:
If there’s a bad URL, try to fix it.
Determine whether the source is verifiable and reliable (see WP:V and WP:RS). If not, determine if it can be easily replaced. If not, delete it.
If the source is reliable but not formatted properly, convert it into a correctly formatted footnote with full information.
Note
While your goal is to convert all embedded links (the ones that look like this: “[1]”) to footnotes, you might want to avoid the temptation of starting by putting <ref> tags around them to immediately convert them to footnotes. It’s easier to work with sources section by section, and if you create footnotes, the URLs go down to the bottom of the page, where you can’t see them when previewing a section.
Links go bad: A link that worked on the day it was added to an article may not work a month or a year later. That’s why full citations are so critical: If the URL stops working, the citation—to a magazine, newspaper, or other source available offline—is still acceptable, because it’s still verifiable.
Unfortunately, you’re often looking at a source that consists only of a URL. So your challenge is to find where the content moved to, or to get a copy. Here’s a step-by-step process:
If the URL were working, would the source be acceptable? For example, if the link is to a page at BloggersOnFire.com or at ExtremistRant.net, you can probably use it only if the article is about the Web site or the organization behind it, or a notable expert on a particular subject posts about that subject at the site. Most of the time, you have to discard it.
As discussed in the next section, a link to an unacceptable source can sometimes lead you to an acceptable source. But here, if the source is unacceptable, you’re facing a double problem: First you’ve got to figure out a fix for the bad URL, and then you’ve got to get lucky and have that lead you to another, acceptable source. In such a case, you’re justified in only doing a few, not particularly time-consuming things to try to fix the URL. You need not invest a lot of time in something that probably has no payoff.
Can you find a substitute source? For example, Reuters stories are removed from the Web after 30 days, but if the subject was a national story, you’re likely to find it at the New York Times, which provides full access to the last 20 years of its archives at no cost. Select some key words from the sentence just before the URL, and do a Web search, if the facts involve recent events. Or just head for NYTimes.com for a replacement URL and a full citation.
Does the bad link go to a newspaper site? If that’s the case, it typically gets redirected to the front page. The story of interest has probably been archived and now simply has a different URL. Search the archives, even if you know you have to pay to see the full story. You goal is to get a URL that’s the free abstract or free initial paragraph of the news story, to replace the old URL and create a full citation.
Does the bad link go to an existing Web site? If the site doesn’t hide its old content behind an internal search engine, as many newspaper sites do, then the content you’re looking for may still exist at a different URL, and you can find it using a search engine. For example, if you’re working on an article about “Joe Bfystlat” and the site is NukesForPeace.org, try a domain-restricted search like "Joe Bfystlat" site: NukesForPeace.org to turn up any existing pages.
Is a copy of the page stored at the Internet Archive? A large part of the Web has been copied and stored at this site (also known as the Wayback Machine), at www.archive.org. You just copy and paste the bad URL to see if you’re lucky or not.
If entering the full URL doesn’t yield any results, try trimming it. (For example, trim www.example.com/Level1/Level2/detail.htm to be www.example.com/Level1/Level2/; if that doesn’t work, try www.example/.com/Level1/.) If you do find a copy, see the page Wikipedia:Using the Wayback Machine (shortcut: WP:WBM) on how to cite the page.
You’re improving citations to make sure all remaining ones are acceptable per WP:V and WP:RS. So you need to replace each bad source with a good one. If you can’t find a good one, you should still delete the bad source. Here are three approaches to finding a replacement:
If you’ve got a functioning URL to a source isn’t acceptable (a blog, forum, or personal Web page, for example), see if that Web page has a link to an acceptable source. For example, a blog often has a link to the news story the blogger’s writing about, a link you can follow. Then simply replace the unacceptable URL with the better one, and finish fixing the citation.
Blogs often quote part of a news story or document without providing a link. If that story or document is what you’re looking for, then search the Web for part of the quoted text. Pick a group of five or six consecutive words that’s a bit unusual in some way, and search that, putting quotation marks around the words so the search engine looks for them in exactly that sequence.
Finally, as mentioned previously, if the unacceptable source is discussing a news story that got national coverage, look up the story on the New York Times or search Google News (http://news.google.com), and then use that as a replacement.
Note
Remember, an acceptable source doesn’t have to be online. If you have easy access to a microfiche copy of old newspapers in your home town, for example, you can use that information for a citation. Wikipedia prefers online sources when available, but there’s no exclusionary rule.
If you’ve made a good faith effort to find a replacement for an unacceptable source, and weren’t successful, then delete the source, with a brief explanation in the edit summary about what you tried.
Once you have an acceptable source, change whatever was in the article (typically just a URL) to a fully formatted citation—a footnote. Chapter 2 discusses citations in depth, with instructions on formatting footnotes on Creating Footnotes.
Tip
If you cite a source more than once in an article, or if the source looks like a promising place to get information for the article later, then make the footnote for that source into a named footnote. As the leading tag, use, for example, <ref name="AJJones">, where the author is A. J. Jones, rather than just <ref>.