Older Web browsers & HTML Entities

Symptom: Older Web browsers - including Cyberdog and 2.x versions of Netscape Navigator and Microsoft Internet Explorer - may not recognize some HTML 3.2 and HTML 4.0 character entities used in the archive display, such as "»" (») and "·" (·). Also, archive contributors may use extended characters in their messages, which may in turn be translated to HTML entities that a particular brower doesn't support - trademark (™) or currency (¥) symbols are good examples. Also, some characters entities that are correctly supported by a browser may not be displayed correctly if the characters aren't available in the user's default display font.

Workaround: We do not intend to address this problem, although we will endeavor to make sure the archive complies with current HTML specifications and common font sets.

Older Web Browsers, JavaScript, & Archive Banners

Symptom: Even though JavaScript is enabled, messages and search results list appearing in the archive frameset are displayed with banners identifying the archive. This information is redundant and takes up a lot of screen space.

Workaround: The archive uses JavaScript to determine if messages and message listings are being displayed in the archive frame set. Currently, the scripts require JavaScript 1.1, which isn't available in some older 2.x browsers - we haven't found a reliable way to write these scripts using JavaScript 1.0. Most 3.x browsers support JavaScript 1.1 or higher.

19-Mar-04 - Database Split

The message database was getting too big for its britches, so I've split it into two chunks. The first chnk carries messages from 01-Jan-2003 to the present, and has new messages added daily. The other chunk carries older messages from (roughly) 01-Feb-1999 to 31-Dec-2002, and is static: not planning on adding any older messages to it. The databases can be searched one at a time; unfortunately, there's no simple and efficient way to combine search results from both databases in one query. (It can be done, but for performance reasons, I'd rather not do it.) I'll keep an eye on usage logs and see if the current solution is causing any problems for folks.

08-Jan-00 - Message Menu Perskicketies

Fixed a puzzling bug in the author pop-up menus that appear with each message; they should now work reliably with or without JavaScript.

02-Jan-00 - Rebuild & Conversion

Did a bunch of work on the archive that I'd been putting off for a while:

• Converted to more recent version of Lasso
• Rebuilt database to allow slighly more efficient use
• Fixed some timing bugs in the POP archiver
• Eliminated links on author names that would bring up every post in the archive by a particular contributor. This works great in small archives, but we're approaching a year of material here, and searching for every message from a frequent contributor is a quick way to bring the database to its knees. You can still search for messages by a particular author, but this way that database won't be sent into a spin cycle as often by a "curious click."

06-Nov-99 - About Page

Added a page briefly describing the HyperCard Mailing List since the HyperCard Resource Page that had previously served as a Web-based explanation and gateway to the list has gone away.

28-Feb-99 - Hello world

Initial implementation.

The HyperCard mailing list archive was designed and implemented by Geoff Duncan, based upon work done for the TidBITS Talk Archive

The archive uses FileMaker Pro from FileMaker, Inc., and is served to the Web via Lasso from BlueWorld Communications. The database lives (with several others) on a Mac OS Web server using WebSTAR from StarNine. Appropriately enough, custom utilities written in HyperCard and AppleScript from Apple Computer integrate new messages into the archive.