My approach to history is pretty fast and loose on these pages. My intent is to steal from history to develop roleplaying content, rather than studying history for its own sake.
The result is that Magic Jar should not be used as a serious history resource. I cannot guarantee that all of the material in it is accurate. I make liberal use of secondary (and even tertiary) sources, I make no effort to check up on the facts I find, etc.
That said, I do care about the historical accuracy of this information. Does that make any sense? I'd prefer if stuff was historically accurate, but I'm not a good enough historian to make it so, given my purposes here. If I have something grossly wrong, I'd like to hear about it. Any comments from people who know more than me about this stuff would be much appreciated.
I make an attempt to list all of my sources, so that the interested historical reader can go to my sources and read more directly the information. Some of these sources are the web, and due to the changing nature of the Internet, may be dead. I'd like to hear about any dead links, so I can fix or remove them.
The contents of Magic Jar are authored in XML (yes, even this page), and HTML is generated using an XSL processor. Currently I'm using LotusXSL as my XSL processor, but this will likely change -- LotusXSL is slow, plus it has a restrictive license.
The pages are authored using Emacs, the extensible, customizable, self-documenting real-time display editor. Emacs is glorious. I use the NT port, because I use Windows NT. I have not made the jump to Linux as my primary operating system, although I do have a Linux server running in my home.
There is a bunch of custom Java code to manage the price database: converting prices of items to differing currencies, etc. Also, bits of the XSL transforms are written in Java, as XSLT extensions (which means my XSLT stylesheets aren't cross-platform).
The pages you are reading are generated on my computer using a GNU Makefile. They are not generated on demand, although this is something I consider occassionally. The files are packaged for upload using the same tool.
The web site itself is hosted at my domain (svincent.com), and my domain is hosted at my employer, Meteko. I really dislike Meteko's web page. It uses Javascript, which I hate, because it doesn't work. Ah well.
The drop cap images are in a font called Ruritania, by Paul J. Lloyd. More of his fonts can be downloaded from Kris's Haven.
The drop caps were implemented by entering all the capital letters into CorelDRAW!, and exported them as GIF images. (I will probably change the images to PNG soon, for the better licensing...) My XSLT stylesheet automatically generates drop caps in specially marked sections. It's a little gross, but if you consider that I am using HTML as a rendering engine, as there are no widely available formatting-object browsers in existence, it works ok.
What does all this technological mumbo-jumbo mean to users of the system? Not much, I suppose, but I figure some people might be interested. :)
I have a curious stance on browser support. There is nothing on these pages that should cause any browser grief: I hardly ever use any HTML tags other than anchors, images, tables, headings, and paragraphs. However, I do make massive use of Cascading Style Sheets (CSS) to style everything, I use tables, which some older browsers don't support, I don't specify image sizes for people with old browsers that are incapable of rerendering the page while the image gets downloaded...
The fact is that I don't particularly care about supporting old browsers. My main concern is one of asthetics and elegance. Using XSL to generate super-thin HTML with cascading style sheets doing all the formatting is elegant, above all else.
Thus, bugs in old or wierd browsers that some people work around, I ignore. I believe that you should browse the Internet with a browser that supports relatively recent standards, and has had most glaring bugs fixed. Commenting out (<-- -->) blocks of script code so that old versions of Netscape Navigator ignore them may work, but it's ugly beyond belief. Users should upgrade their browsers appropriately.
On the other hand, I'm always careful to support Lynx. I don't believe people should be using latest versions of one of the Big Two browsers: I just think that browser developers should support recent standards, and users should ensure that the browsers that they're using are modern and well written. Lynx is a fine idea: the Web is meant to be multi-platform, working well on text-based mediums: CSS has tons of support for that. But Lynx should be upgraded to support the standards it can support: like in-browser XSL transforms, CSS support (where it can), etc.
This policy is not for everybody. Corporations, for example, that want to allow everybody to see their web pages need to bend over backwards to make everybody happy. I have the advantage that if I lose 30% of my visitors because they can't see tables, I don't have to worry about it: this is a personal endeavor.
So there you go. That's my rant. I should go through and clean it up one of these days. I will -- but not right now.