Just a note about a frustrating problem I bumped into this morning, and a question.
We have an intranet application that has been in use for several years that displays various information from our accounting system. While testing a few updates, I discovered some severe script errors on one page that no one had ever reported as a problem. When I viewed the page in the Microsoft Script De######, one of the test strings was in Chinese (maybe) and all the + signs disappeared from the concatenation operations.
Well, it turned out that because I never specified a particular character set in the page, the browser was picking up something from the database for this particular (varchar?) field that caused it to interpret the page as Unicode UTF-7, and that completely wrecked the scripts.
For this application, which only runs in IE because it requires NT credentials to access the server, I added this to each page at the top of the <head> section:
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=Windows-1252">
That's "Western European (Windows)"; more IE-specific CHARSET strings are listed in HTML and Dynamic HTML > Reference > Character Set Recognition on MSDN.
Question: Is there is a standardized cross-browser solution?