<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">I gotta generally agree with Matt.
Occasionally I run into an information modeling project that I can
knock off in an afternoon, but that's pretty rare. Remember that
you will not only need to model "block" content (topics, headings,
paragraphs, lists, etc.), but also tables, cross-references,
images, etc. The latter set can be a bit tricky. Plus, oh, your
metadata.<br>
<br>
With DITA or DocBook, you also get a publishing framework. Also
usually non-trivial to create from scratch, especially if you are
publishing to multiple output formats, using filtering, content
re-use, etc.<br>
<br>
I'll mention with some regret that FrameMaker's DocBook support is
pretty poor. I've never figured out why...the "typical" use cases
for both (books, PDF) line up very well. It may be a
chicken-and-egg issue...I suspect more people would use DocBook if
FrameMaker provided better DocBook support.<br>
<br>
-Alan<br>
<br>
On 7/8/13 6:31 PM, Matt Sullivan wrote:<br>
</div>
<blockquote
cite="mid:AA1E4227-7475-4E72-B3BA-5E4EA1BFAAA3@mattrsullivan.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html;
charset=ISO-8859-1">
<base href="x-msg://105/">A list of what you'll save using DITA or
DocBook rather than creating your own schema:
<div>
<ol class="MailOutline">
<li>Time</li>
<li>Money</li>
</ol>
<div><br>
</div>
<div>(Hey, someone had to say it…)</div>
<div apple-content-edited="true">
<p style="font-family: arial, helvetica; "><br>
-Matt<br>
</p>
<br>
</div>
</div>
</blockquote>
<br>
</body>
</html>