November 28, 2009


With the recent changes to the format of Lionel Deimel’s Web Log, some posts turned ugly. The column containing posts themselves is narrower than before, so some wide graphics no longer fit within that column. I have gone through all the posts on the blog and have replaced too-wide graphics with narrower ones. Note that, in most cases, you can click on a graphic and see it on a page by itself. In a few cases, when you do this, you will see a larger version, generally the original version that has now been replaced in the post itself.

While fixing images that were too big, I took the opportunity to correct a few other formatting glitches. Most of the changes were subtle and won’t be noticed.

The Link Problem

What I did not change were URLs that are no longer correct. It is difficult to know what to do with broken links on my blog. Several situations arise:
  1. A page may simply have disappeared from the Web.
  2. A page may no longer be at its former address.
  3. A page may be completely different from the one originally cited.
  4. Changed circumstances may have made it unclear what page should actually be linked to.
Let me discuss these situations in turn. In each case, it may be necessary or desirable to explain what has been done.

A page is no longer available. This is a difficult situation to deal with. Alternatives are
  • Retain the link with or without comment. More explanation may be needed to tell visitors what they’re missing.
  • Cite another page with the same information, if one exists. In some cases, it may be possible to create a local substitute page.
  • Cite an historical version of the page in the Internet Archive if one is available. (Check out the Archive if you aren’t familiar with it.)
  • Simply delete the link, with or without explanation.
A page has changed location. Often, the URL of a page has changed because its site has been reorganized, because its domain has changed, or both. The fix here is easy if one can discover the new URL and there are no complicating factors. An explanation may be necessary.

A page’s content has changed. Webmasters are sometimes not looking to the future when they select page addresses. A page of recent news may look completely different in July from how it looked in January. I try to link to pages whose content seems likely to be stable, but this is not always possible. If the page’s content has disappeared and the old content is unavailable, removing the link may be the only option.

A changed context may make it difficult to choose a proper link. A trickier situation to deal with is one where the context has changed. For example, a link to the home page of the Episcopal Diocese of Pittsburgh Web site created before the diocese split is problematic. Linking to the Web site of either resulting diocese would be misleading. The current Episcopal diocese is the legal successor to the pre-schism diocese, but the current Anglican diocese maintains the Web site that changed little when the schism occurred, whereas the Episcopal diocese created an entirely new Web site. This is a case where one might want simply to delete the link. Linking to an Archive page might be reasonable, but not without an explanation.

Choosing Links

We all understand references in a book. We know that we can go to a library, find a referenced book, and know exactly what the author was referring to. Hyperlinks on a Web page are neat because the going-to-the-library step is replaced by immediate transportation to the referenced material. Unfortunately, the material may have disappeared entirely, been changed in subtle ways, or otherwise might not tell us exactly what we think it should be communicating. In many circumstances, we would like a Web page and all its links to be frozen in time, so that anyone visiting the page could read it like a book, except that the references would be both stable and immediately accessible. But, sometimes, we want a link to go to a page we expect to change, as when we cite a blog qua blog, rather than as a place to find specific content. (Even a blog can completely disappear, of course, so stability is always an issue.)

As a general rule, Web sites that are somehow journalistic in nature (newspaper sites, opinion sites, blogs) are not regularly reviewed to keep their links up-to-date. As for me, I do get upset when my links, in whatever way, degrade. I try to link defensively, citing explicit or probable permalinks whenever I can. I have been known to copy content to my own site, either citing it there or keeping it in reserve against its possible disappearance from its original Web home. This can raise copyright questions, so I do not do it routinely. Sometimes I simply do not create a link because I think the content will disappear. I might describe the content or copy excerpts as an alternative.

Were it up to me, every Web page ever created and every one of its variants would be archived and time stamped and could be linked to. That will never happen, of course. Although the World Wide Web is a kind of huge library, it seems more like a library created by Lewis Carroll than one created by Melvil Dewey. Somehow, we have to live with that.

If I get ambitious, I may make another pass through my blog posts with an eye to fixing broken or misleading links. Suggestions as to how to go about this are welcome.

No comments:

Post a Comment

Anonymous comments are not allowed. All comments are moderated by the author. Gratuitous profanity, libelous statements, and commercial messages will be not be posted.