note: all of this is in flux right now. if it doesn't appear to be acting as advertised, set it aside and come back in an hour. :+) the other important caveat is that so far i am just generating the e-book output files. optimizing them is the _next_ step... hello. this is the _online_ version of a tool for converting text into e-books. the tool is "jaguar", and one online version runs as a python script. this online version is basically a _demo_ for the offline applications, which'll be cost-free, and cross-platform for the mac, pc, and linux. as you can see from the screenshot which i've pasted in right below, this tool is very simple... you put your text in the field on the left, and then click the "do it" button, and your e-book output is created. the output file-formats of this web-app are .html, .mobi, and .epub.*** your text must be .zml -- that's short for "zen markup language" -- a set of formatting rules (e.g., put a blank line between paragraphs; put 5 blank lines before a chapter-header and 2 blank lines after, etc.) below, i will point you to documentation on the simple rules of z.m.l. that'll allow you to prepare your own book-text, if you wanna do that. for now, though, you can just fiddle with the text which gets loaded automatically, which is the _test-suite_ demonstrating various stuff. *** the output takes several forms, but -- for the most part -- it's .html. so, in the pane on the right, the created .html gets displayed -- live! this provides immediate feedback on if your .zml is entered correctly. for instance, you'll want to make sure all of your headers look right... if the output doesn't look correct to you, it means the input is wrong; meaning you'll need to change the input until you get the right output. in addition to the _look_ of the output, you'll want to make sure that it _behaves_ correctly. for instance, the items in the table of contents automatically become _links_ to the appropriate chapters, so you will want to test some of them, to ensure they are working as they should. if the displayed .html looks right, and seems to be working correctly, then you can download the e-book file-formats, and check them out. every time anyone clicks the "do it" button, the files are regenerated, and, at present, all the generated files are saved in just one location. so if you want to download those generated files, do it _immediately_ after you click "do it", or else somebody else might over-write yours... here's the folder -- open for viewing -- with the generated files: > http://zenmagiclove.com/zml/suite (when you get there, you should see 2 subdirectories and an .epub.) here are the exact locations of the .epub and the .mobi: > http://zenmagiclove.com/zml/suite/suite.epub> http://zenmagiclove.com/zml/suite/mobi/suite.mobi you might also want to examine the .html files the tool generates. you can unzip the .epub, of course, to view the files that comprise it. or you can view these files (and more) in the .epub directory above. for the .mobi, all the files needed for kindlegen are in its directory. there's a .txt version of the .opf file, for viewing inside the browser. the last time i checked, the .epub file for the test suite _validates_, and the .mobi file is generated without any warnings or errors, but that might or might not be the case as i make changes to the tool... but rest assured that any production version will certainly validate. in addition, it's entirely possible (and perhaps even _probable_) that when you enter your own text, the generated files will fail to validate. it might be because you made errors in the .zml for your input text, or it might be because of some bug i have in the converter program. or maybe someone before you edited the demo text and messed it up. either way, let me know about the problem. even if it's "your fault", because of an error in the .zml, i need to know what kind of mistakes people tend to make... so if you can't fix any .zml problems yourself, don't hesitate to come to me with them. at least not in the beginning. at some time, i'll expect people to become self-reliant. but not at first. (but if it's clear you just didn't do your .zml correctly, and didn't even do the necessary work to learn it, i will kick it back to you very quickly. the rules of z.m.l. are simple; i shouldn't need to be patient with you.) *** so, how do you learn z.m.l.? well, one way will be to look at examples. here's how you can load in the .zml text for "pride and prejudice": > http://zenmagiclove.com/zml/suite/jaguar.py?book=prapr and "the adventures of sherlock holmes": > http://zenmagiclove.com/zml/suite/jaguar.py?book=taosh and here's "alice in wonderland": > http://zenmagiclove.com/zml/suite/jaguar.py?book=alice *** and here's a "skeleton" file that you can use as a "template": > http://zenmagiclove.com/zml/suite/jaguar.py?book=skelt also, last but not least, here's the file with the "rules" for .zml: > http://zenmagiclove.com/zml/suite/jaguar.py?book=rules (i put it here at the bottom, just to see if you read all the way.) this "rules" file is old, so there might be parts that are dated, but i think for the most part that not much has been changed. that being said, though, all these "rules" are subject to change. i'm flexible while i see what kinds of functionality people need, and what seems to be working well and what needs to be fixed. i've worked on a lot of this for years, so i'm fairly sure about it, but until it gets a good workout with a lot of users, i'm flexible. if you need any help, just holler... -bowerbird p.s. here's the straight u.r.l. for the tool: > http://zenmagiclove.com/zml/suite/jaguar.py