Sometimes you could almost believe in gremlins, the evil little creatures Second World War airmen used to blame for errors which cropped up inexplicably from time to time.
Every so often something breaks on a web site like JWC. It happened again “recently”: I can’t be more specific because I only noticed it this morning when I went to use one of the site content tools, the Unicode Converter: type in characters and it converts them to their unicode, hexadecimal or decimal NCR equivalent.
Only it didn’t. Type in something and the only conversion it made was to Unicode UTF-8. There were no error numbers, no warning messages. It just didn’t work. I spent an hour poring over the PHP but couldn’t find the error. Eventually, I went back to square one and copied the code afresh from sceneonthe.net, the web partnership where we originally developed it.
The content tool is back but I’m still none the wiser as to how it went wrong.
Why You Should
Broken pages are a content manager’s nightmare. They may be the result of a programming error or code innocently altered by authorised site admins, or they may be the result of something more sinister. Then there are dead hot links caused by unexpected updates and the rest.
Errors are bad for SEO; they imply lack of attention to detail or even lack of updates and the least updated sites are dead ones. So part of your routine as content manager must be in finding — and fixing — broken stuff, perhaps with a link-check program or browser plug-in or indeed some outside assistance.
Users of TIME’s web site were always pointing out broken links, or worse, broken code (there is a kind of elation when pointing out errors to organisations who should know better).
But don’t shy away from such feedback and encourage it: ask your visitors to report any errors they see. And remember that someone who has never made a mistake has never made anything.