By Claus Rasmussen, with additions by Roberto Mello, Vinod Kurup, and the OpenACS Community
OpenACS™ is a powerful system with incredible possibilities and applications, but this power comes with some complexity and a steep learning curve that is only attenuated by good documentation. Our goal is to write superb documentation, so that users, developers and administrators of OpenACS installations can enjoy the system.
The history of OpenACS documentation: ..began by building on a good documentation base from ArsDigita's ACS in the late 1990's. Some sections of the documentation, however, lacked details and examples; others simply did not exist. The OpenACS community began meeting the challenge by identifying needs and writing documentation on an as needed basis.
By having documentation dependent on volunteers and code developers, documentation updates lagged behind the evolving system software. As significant development changes were made to the system, existing documentation became dated, and its value significantly reduced. The valiant efforts that were made to keep the documentation current proved too difficult as changes to the system sometimes had far-reaching affects to pages throughout the documentation. System integration and optimization quickly rendered documentation obsolete for developers. The code became the substitute and source for documentation.
With thousands of lines of code and few developers tracking changes, features and advances to the OpenACS system went unnoticed or were not well understood except by the code authors. Work was duplicated as a consequence of developers not realizing the significant work completed by others. New developers had to learn the system through experience with working with it and discussion in the forums. Informal sharing of experiential and tacit knowledge has become the OpenACS community's main method of sharing knowledge.
This document attempts to shape ongoing documentation efforts by using principles of continual improvement to re-engineer documentation production.
Documentation production shares many of the challenges of software development, such as managing contributions, revisions and the (editorial) release cycle. This is yet another experiment in improving documentation --this time by using principles of continual improvement to focus the on-going efforts. These processes are outlined as project management phases:
Requirements phase is about setting goals and specifications, and includes exploration of scenarios, use cases etc. As an example, see the OpenACS Documentation Requirements Template which focuses on systems requirements for developers.
Strategy phase is about creating an approach to doing work. It sets behavioral guidelines and boundaries that help keep perspective on how efforts are directed. OpenACS developers discuss strategy when coordinating efforts such as code revisioning and new features.
Planning phase is about explicitly stating the way to implement the strategy as a set of methods. OpenACS system design requires planning. For example, see OpenACS documentation template planning relating to package design.
Implementation phase is about performing the work according to the plan, where decisions on how to handle unforseen circumstances are guided by the strategy and requirements.
Verification phase measures how well the plan was implemented. Success is measured by A) verifying if the project has met the established goals, and B) reviewing for ongoing problem areas etc. OpenACS follows verification through different means on different projects, but in all cases, the OpenACS community verifies the project as a success through feedback including bug reports, user and administrator comments, and code changes.
OpenACS forum discussions on documentation requirements and strategies are summarized in the following sections. Production phases are mainly organized and fulfilled by a designated documentation maintainer. Hopefully the following sections will help spur greater direct participation by the OpenACS community.
By the OpenACS Community. This section is a collection of documentation requirements that have been expressed in the OpenACS forums to 4th July 2003.
OpenACS documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
clarity in presentation. Life with qmail is a recommended example of "rated high" online documentation.
Avoid requirements that significantly increase the labor required to maintain documentation.
Use best practices learned from the print world, web, and other media, about use of gamma, space, writing style etc.
Consistency in publishing -Establishing and adhering to publishing standards
Use standardized language -Use international English (without slang or colloquial terms) for ESL (English as a second language) readers (and making translation easier for those interested in translating the documentation for internationalization efforts).
All jargon used in documentation needs to be defined. Use standardized terms when available, avoiding implicit understanding of specific OpenACS terms.
Document titles (for example on html pages) should include whole document title (as in book title): (chapter title) : (section), so that bookmarks etc. indicate location in a manner similar to pages in books (in print publishing world).
Organize document according to the needs of the reader (which may be different than the wishes of the writers).
Do not make informal exclamations about difficulty/ease for users to complete tasks or understand... for example, "Simply...". Readers come from many different backgrounds --remember that the greater audience is likely as varied as the readers on the internet--- If important, state pre-conditions or knowledge requirements etc. if different than the rest of the context of the document. For example, "requires basic competency with a text-based editor such as vi or emacs via telnet"
Show where to find current information instead of writing about current info that becomes obsolete. If the information is not found elsewhere, then create one place for it, where others can refer to it. This structure of information will significantly reduce obsolescence in writing and labor burden to maintain up-to-date documentation. In other words, state facts in appropriately focused, designated areas only, then refer to them by reference (with links).
Note: Sometimes facts should be stated multiple ways, to accommodate different reading style preferences. The should still be in 1 area, using a common layout of perhaps summary, introduction and discussion requiring increasing expertise, complexity or specificity.
Consistency in link descriptions -When link urls refer to whole documents, make the link (anchor wrapped title) that points to a document with the same title and/or heading of the document.
Consider OpenACS documentation as a set of books (an encyclopedic set organized like an atlas) that contains volumes (books). Each book contains chapters and sections much like how DocBook examples are shown, where each chapter is a web page. This designation could help create an OpenACs book in print, and help new readers visualize how the documentation is organized.
The use licenses between OpenACS and Arsdigita's ACS are not compatible, thereby creating strict limits on how much OpenACS developers should have access to Arsdigita code and resources. The OpenACS documentation has a new legal requirement: to eliminate any dependency on learning about the system from Arsdigita ACS examples to minimize any inference of license noncompliance, while recognizing the important work accomplished by Philip Greenspun, Arsdigita, and the early ACS adopters.
Use a consistent general outline for each book.
Introduction (includes purpose/goal), Glossary of terms, Credits, License, Copyright, Revision History
Table of Contents (TOC)s for each book: the end-users, content and site administrators, marketing, developer tutorial, and developers.
Priorities of order and content vary based on each of the different readers mentioned. The developers guide should be organized to be most useful to the priorities of developers, while being consistent with the general documentation requirements including publishing strategy, style etc.
Use generic DocBook syntax to maximize reader familiarity with the documents.
<book><title><part label="Part 1"><etc...>
By the OpenACS Community. This section is a collection of documentation requirements that have been expressed in the OpenACS forums to 4th July 2003.
OpenACS end-user documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
End-users should not have to read docs to use the system.
Include how to get help. How and where to find answers, contact others, what to do if one gets an AOLserver or other error when using the system. Include types of available support (open-source, private commercial etc.) including references.
Explain/foster understanding of the overall structure of the system. This would be an overview of the system components, how it works, and how to find out more or dig deeper... To promote the system by presenting the history of the system, and writing about some tacit knowledge re: OpenACS.org and the opensource culture.
Introduce and inspire readers about the uses, benefits, and the possibilities this system brings (think customer solution, customer cost, convenience, value). A comprehensive community communications system; How this system is valuable to users; Reasons others use OpenACS (with quotes in their own words) "...the most important thing that the ACS does is manage users, i.e. provide a way to group, view and manipulate members of the web community. -- Talli Somekh, September 19, 2001" using it to communicate, cooperate, collaborate... OpenACS offers directed content functionality with the OpenACS templating system. ... OpenACS is more than a data collection and presentation tool. OpenACS has management facilities that are absent in other portals. ...The beauty of OpenACS is the simplicity (and scalability) of the platform on which it is built and the library of tried and tested community building tools that are waiting to be added. It seems that most portals just add another layer of complexity to the cake. See Slides on OACS features...a set of slides on OACS features that can be used for beginners who want to know OACS is about and what they can do with it. Screen captures that highlight features. Example shows BBoard, calendar, news, file storage, wimpy point, ticket tracking. An OpenACS tour; an abbreviated, interactive set of demo pages.
From a marketing perspective,
differentiate "product" by highlighting features, performance quality, conformance to standards, durability (handling of technological obsolescence), reliability, repairability, style of use, design (strategy in design, specifications, integrated, well-matched systems etc).
differentiate "service" by highlighting software availability (licensing and completeness from mature to early adopters or development versions), community incident support, project collaborative opportunities, and contractor support availability
differentiate price (economic considerations of opensource and features)
Discussion and details should rely on meeting criteria of design, completeness of implementation, and related system strengths and weaknesses. Marketing should not rely on comparing to other technologies. Competitive analysis involves mapping out strengths, weaknesses, opportunities and threats when compared to other systems for a specific purpose, and thus is inappropriate (and becomes stale quickly) for general documentation.
When identifying subsystems, such as tcl, include links to their marketing material if available.
create an example/template comparison table that shows versions of OpenACS and other systems (commonly competing against OpenACS) versus a summary feature list and how well each meets the feature criteria. Each system should be marked with a date to indicate time information was gathered, since information is likely volatile.
To build awareness about OpenACS, consider product differentiation: form, features, performance quality, conformance quality (to standards and requirements), durability, reliability, repairability, style, design: the deliberate planning of these product attributes.
Include jargon definitions, glossary, FAQs, site map/index, including where to find Instructions for using the packages. FAQ should refer like answers to the same place for consistency, brevity and maintainability.
Explain/tutorial on how the UI works (links do more than go to places, they are active), Page flow, descriptions of form elements; browser/interface strengths and limitations (cookies, other)
Discuss criteria used to decide which features are important, and the quality of the implementation from a users perspective. Each project implementation places a different emphasis on the various criteria, which is why providing a framework to help decide is probably more useful than an actual comparison.
Package documentation requirements have additional requirements.
A list of all packages, their names, their purposes, what they can and cannot do (strengths, limitations), what differentiates them from similar packages, minimal description, current version, implementation status, author/maintainers, link(s) to more info. Current version available at the repository.
Include dependencies/requirements, known conflicts, and comments from the real world edited into a longer description to quickly learn if a package is appropriate for specific projects.
Create a long bulleted list of features. Feature list should go deeper than high-level feature lists and look at the quality of the implementations (from the user's perspective, not the programmer's). Example issues an end-user may have questions about: Ticket Tracker and Ticket Tracker Lite, why would I want one of them vs the other? And, before I specify to download and install it, what credit card gateways are supported by the current e-commerce module? There are some packages where the name is clear enough, but what are the limitations of the standard package?
End-user docs should not be duplicative. The package description information and almost everything about a package for administrators and developers is already described in the package itself through two basic development document templates: a Requirements Template and Detailed Design Document.
By the OpenACS Community. This section is a collection of documentation requirements that have been expressed in the OpenACS forums to 4th July 2003.
OpenACS administrators' documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
For each requirement below, include links to developer tutorials and other documentation for more detail.
Describe a structural overview of a working system and how the components work together. "The Layered Cake view" a general network view of system; a table showing system levels versus roles to help with understanding how the subsystems are interconnected.
Provide a comprehensive description of typical administrative processes for operating an OpenACS system responsibly, including reading logs and command line views that describe status of various active processes.
Create a list of administrative tools that are useful to administrating OpenACS, including developer support, schema-browser and api browser. Link to AOLserver's config file documentation.
Resources on high level subjects such as web services, security guidelines
Describe typical skill sets (and perhaps mapped to standardized job titles) for administrating an OpenACS system (human-resources style). For a subsite admin/moderator attributes might include trustworthy, sociable, familiarity with the applications and subsystems, work/group communication skills et cetera
Describe how to set up typical site moderation and administration including parameters, permissions, "Hello World" page
Show directory structure of a typical package, explanation of the various file types in a package (tcl,adp,xql) and how those relate to the previously described subsystems, when they get refreshed etc.
Ways to build a "Hello World" page
Show examples of how the OpenACS templating system is used, including portal sections of pages. For example, create a customised auto-refreshing startpage using lars-blogger, a photo gallery, and latest posts from a forum. This should rely heavily on documentation existing elsewhere to keep current. This would essentially be a heavily annotated list of links.
Show ways of modifying the look and feel across pages of an OpenACS website. Refer to the skins package tutorial.
Describe a methodology for diagnosing problems, finding error statements and interpreting them --for OpenACS and the underlying processes.
FAQs: Administration tasks commonly discussed on boards: admin page flow, how to change the looks of a subsite with a new master.adp, options on "user pages" , a quick introduction to the functions and processes. info about the user variables, file locations
OpenACS installation documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
state installation prerequisites. For example: "You should read through the installation process to familiarize yourself with the installation process, before beginning an installation."
list critical decisions (perhaps as questions) that need to be made before starting: which OS, which DB, which aolserver version, system name, dependencies et cetera. Maybe summarize options as tables or decision-trees. For example, "As you proceed throughout the installation, you will be acting on decisions that have an impact on how the remaining part of the system is installed. Here is a list of questions you should answer before beginning."
list pre-installation assumptions
Show chronological overview of the process of installing a system to full working status: Install operating system with supporting software, configure with preparations for OpenACS, RDBMS(s) install and configure, Webserver install and configure, OpenACS install and configure, post-install work
OpenACS developer tutorial documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
list learning prerequisites to customize, fix, and improve OACS modules, and create new ones. You are expected to have read and understand the information [minimum requirements similar to adept at Using OpenACS Administrating Guide] before reading this guide.
Refer to development documentation instead of duplicating here
List suggestions for installing and setting up a development environment; these can be annotated links to the installation documentation
Provide working examples that highlight the various subsystems, tcl environment, OpenACS protocols, aolserver template and ns_* commands, OpenACS templating, sql queries, db triggers, scheduling protocols, how to use the page contract, how to get the accessing user_id etc
Show how to construct basic SQL queries using the db API,
The life of an http request to a dynamic, templated page
General rules to follow for stability, scalability
Show the step by step customizing of an existing package that meets current recommended coding styles of OpenACS package development, by referring to developer resources.
Use the ArsDigita problem sets and "what Lars produced for ACS Java" as inspiration for a PostgreSQL equivalent tutorial about developing a new OpenACS package including discussion of the significance of the package documentation templates
Include a summary of important links used by developers
Note any deprecated tools and methods by linking to prior versions instead of describing them in current docs
OpenACS developer documentation should meet the following requirements. No significance has been given to the order presented, topic breadth or depth here.
list documentation assumptions, such as familiarity with modifying OpenACS packages. All kernel docs are here etc.
This documentation should be written for ongoing use by developers, not as a tutorial.
List of practical development and diagnostics tools and methodologies.
List of OpenACS development resources, api-doc, schema-browser, developer-support package etc.
Identify each OpenACS subsystem, explain why it is used (instead of other choices). In the case of subsystems that are developed outside of OpenACS such as tcl, include external references to development and reference areas.
Show current engineering standards and indicate where changes to the standards are in the works.
Sections should be dedicated to DotLRN standards as well, if they are not available elsewhere.
Add overview diagrams showing the core parts of the datamodel including an updated summary of Greenspun's Chapter 4: Data Models and the Object System
package design guidelines and development process templates including planning, core functions, testing, usability, and creating case studies
Standard package conventions, where to see "model" code, and guidelines (or where to find them) for:
using the acs-api
package_name and type_extension_table
adding optional services, including search, general comments, attachments, notifications, workflow, CR and the new CR Tcl API
Document kernel coding requirements, strategy and guidelines to help code changers make decisions that meet kernel designers' criteria
OpenACS documentation development is subject to the constraints of the software project development and release methods and cycles (the section called “Using CVS with OpenACS”). Essentially, all phases of work may be active to accommodate the asynchronous nature of multiple subprojects evolving by the efforts of a global base of participants with culturally diverse time references and scheduling idiosyncrasies.
The documentation strategy is to use project methods to involve others by collaborating or obtaining guidance or feedback (peer review) to distribute the workload and increase the overall value of output for the OpenACS project.
OpenACS documentation is taking a dual approach to publishing. Documentation that is subject to rapid change and participation by the OpenACS community is managed through the OpenACS xowiki Documentation Project Formal documents that tend to remain static and require more expressive publishing tools will be marked up to conform to the DocBook XML DTD. The remaining discussion is about publishing using Docbook.
It is open-source.
A growing community surrounds DocBook (has mailing lists)
A number of free and commercial tools are available for editing and publishing DocBook documents.
It enables us to publish in a variety of formats.
XML separates content from presentation: It relieves each contributor of the burden of presentation, freeing each writer to focus on content and sharing knowledge.
It is well tested technology. It has been in development since the early 1990's).
Reasons why we are using Docbook XML instead of Docbook SGML:
Consistency and history. We started with a collection of DocBook XML files that ArsDigita wrote. Trying to re-write them to conform to the SGML DTD would be unnecessary work.
XML does not require extra effort. Writing in XML is almost identical to SGML, with a couple extra rules. More details in the LDP Author Guide.
The tool chain has matured. xsltproc and other XML based tools have improved to the point where they are about as good as the SGML tools. Both can output html and pdf formats.
Albeit, the road to using DocBook has had some trials. In 2002, Docbook still was not fully capable of representing online books as practiced by book publishers and expected from readers with regards to usability on the web. That meant DocBook did not entirely meet OpenACS publishing requirements at that time.
In 2004, Docbook released version 4.4, which complies with all
the OpenACS publishing requirements.
Producing a web friendly book hierarchy arguably remains DocBooks'
weakest point. For example, a dynamically built document
should be able to extract details of a specific reference from
a bibliographic (table) and present a footnote at the
point where referenced. DocBook 4.4 allows for this with
The Definitive Guide is a good start for learning how
to represent paper-based books online.
The following DocBook primer walks you through the basics, and should cover the needs for 95 percent of the documentation we produce. You are welcome to explore DocBook's list of elements and use more exotic features in your documents. The list is made up of SGML-elements but basically the same elements are valid in the XML DTD as long as you remember to:
Always close your tags with corresponding end-tags and to not use other tag minimization
Write all elements and attributes in lowercase
Quote all attributes
You are going to need the following to work with the OpenACS Docbook XML documentation:
Docbook XML DTD - The document type definition for XML. You can find an RPM or DEB package or you can download a zip file from the site linked from here.
XSL Stylesheets (docbook-xsl) - The stylesheets to convert to HTML. We have been using a stylesheet based upon NWalsh's chunk.xsl.
xsltproc - The processor that
will take an XML document and, given a xsl stylesheet, convert
it to HTML. It needs libxml2 and libxslt (available in RPM and
DEB formats or from xmlsoft.org.
After you have the tools mentioned above, you need to define a title for your document. Then start thinking about the possible sections and subsections you will have in your document. Make sure you coordinate with the OpenACS Gatekeepers to make sure you are not writing something that someone else is already writing. Also, if you desire to use the OpenACS CVS repository, please e-mail the gatekeeper in charge of documentation.
You can look at some templates for documents (in Docbook XML) in the sources for acs-core-docs, especially the Detailed Design Documentation Template and the System/Application Requirements Template.
book : Docs for one package - templating | +--chapter : One section - for developers | ---------+------------------------------------------------------ | +--sect1 : Single document - requirements | +--sect2 : Sections - functional requirements | +--sect3 : Subsections - Programmer's API | ... : ...
The actual content is split up into documents that start at a
sect1-level. These are then tied together in a top-level document that
contains all the information above the line. This will be explained in more detail in a later document,
and we will provide a set of templates for documenting an entire package.
For now you can take a look at the sources of these DocBook documents to get an idea of how they are tied together.
Given that your job starts at the
sect1-level, all your documents should open with a
<sect1>-tag and end
with the corresponding
You need to feed every
<sect1> two attributes. The first attribute,
id, is standard and can be used with all elements. It comes in very
handy when interlinking between documents (more about this when talking about links in the section called “Links”).
The value of
id has to be unique
throughout the book you're making since the
id's in your
sect1's will turn into filenames when the book is parsed into HTML.
Right after the opening tag you put the title of the document - this is usually the same as
xreflabel-attribute. E.g. the top level of the document you're
reading right now looks like this:
<sect1 id="docbook-primer" xreflabel="DocBook Primer"> <title>DocBook Primer</title> ... </sect1>
Inside this container your document will be split up into
each with the same requirements -
attributes, and a
<title>-tag inside. Actually, the
xreflabel is never required in sections, but it makes linking to that section a lot easier.
When it comes to naming your
sect2's and below, prefix them with some abbreviation of the
id in the
sect1 such as
For displaying a snippet of code, a filename or anything else you just want to appear as a part of
a sentence, we use
These replace the HTML-tag
depending on whether the tag is describing computer output or
For bigger chunks of code such as SQL-blocks, the tag
<programlisting> is used. Just wrap your code block in it; mono-spacing, indents and all that stuff is taken care of
By having unique
id's you can cross-reference any part of your book
with a simple tag, regardless of where that part is.
Put this in your XML:
- Find information about creating a package in <xref linkend="packages-making-a-package"></xref>.
And the output is:
- Find information about creating a package in Making a Package.
Note that even though this is an empty tag, you have to either:
Provide the end-tag,
Put a slash before the ending-bracket:
If the section you link to hasn't a specified
the link is going to look like this:
Put this in your XML:
-Find information about what a package looks like in <xref linkend="packages-looks"></xref>.
And the output is:
- Find information about what a package looks like in the section called “What a Package Looks Like”.
Note that since I haven't provided an
xreflabel for the subsection,
parser will try its best to explain where the link takes you.
If you're hyper-linking out of the documentation, it works almost the same way as HTML - the tag is just
a little different
<ulink url="http://www.oracle.com/">Oracle Corporation</ulink>
....will create a hyper-link to Oracle in the HTML-version of the documentation.
NOTE: Do NOT use
ampersands in your hyperlinks. These are reserved for
To create an ampersand, use the entity
Note: The graphics guidelines are not written in stone. Use another valid approach if it works better for you.
To insert a graphic we use the elements
Two versions of all graphics are required. One for the Web
(usually a JPEG or GIF), and a brief text description. The
description becomes the ALT text. You can also supply a version for print (EPS).
<mediaobject> <imageobject> <imagedata fileref="images/rp-flow.gif" format="GIF" align="center"/> </imageobject> <imageobject> <imagedata fileref="images/rp-flow.eps" format="EPS" align="center"/> </imageobject> <textobject> <phrase>This is an image of the flow in the Request Processor</phrase> </textobject> </mediaobject>
Put your graphics in a separate directory ("images") and link to them only with relative paths.
Making an unordered list is pretty much like doing the same thing in HTML - if you close your
<li>, that is. The only differences are that each list item has to be wrapped in something more, such as
<para>, and that the tags are called
<itemizedlist> <listitem><para>Stuff goes here</para></listitem> <listitem><para>More stuff goes here</para></listitem> </itemizedlist>
The ordered list is like the preceding, except that you use
<orderedlist> <listitem><para>Stuff goes here</para></listitem> <listitem><para>More stuff goes here</para></listitem> </orderedlist>
<variablelist> <varlistentry> <term>Heading (<dt>) goes here</term> <listitem><para>And stuff (<dd>)goes here</para></listitem> </varlistentry> <varlistentry> <term>Another heading goes here</term> <listitem><para>And more stuff goes here</para></listitem> </varlistentry> </variablelist>
DocBook supports several types of tables, but in most cases, the
<informaltable frame="all"> <tgroup cols="3"> <tbody> <row> <entry>a1</entry> <entry>b1</entry> <entry>c1</entry> </row> <row> <entry>a2</entry> <entry>b2</entry> <entry>c2</entry> </row> <row> <entry>a3</entry> <entry>b3</entry> <entry>c3</entry> </row> </tbody> </tgroup> </informaltable>
With our current XSL-style-sheet, the output of the markup above will be a simple HTML-table:
If you want cells to span more than one row or column, it gets a bit more complicated - check out
for an example.
Our documentation uses two flavors of emphasis - italics and bold type. DocBook uses one -
<emphasis> tag defaults to italics when parsed. If you're looking for
emphasizing with bold type, use
Words that are marked as index-words are referenced in an index in the final, parsed document.
This section is quoted almost verbatim from the LDP Author Guide.
Once you have the Docbook Tools installed, you can convert your xml documents to HTML or other formats.
With the DocBook XSL stylesheets, generation of multiple files is controlled by the stylesheet. If you want to generate a single file, you call one stylesheet. If you want to generate multiple files, you call a different stylesheet.
To generate a single HTML file from your DocBook XML file, use the command:
xsltproc -o outputfilename.xml /usr/share/sgml/docbook/stylesheet/xsl/nwalsh/html/html.xsl filename.xml
This example uses Daniel Veillard's xsltproc command available as part of libxslt from http://www.xmlsoft.org/XSLT/. If you are using other XML processors such as Xalan or Saxon, you will need to change the command line appropriately.
To generate a set of linked HTML pages, with a separate page for each <chapter>, <sect1> or <appendix> tag, use the following command:
xsltproc /usr/share/sgml/docbook/stylesheet/xsl/nwalsh/html/chunk.xsl filename.xml
You could also look at the acs-core-docs Makefile for examples of how these documents are generated.
The LDP Author Guide has a lot of good information, a table of docbook elements and their "look" in HTML and lots of good links for tools.
James Clark wrote nXML Mode, an alternative to PSGML Mode. nXML Mode can validate a file as it is edited.
David Lutterkort wrote an intro to the PSGML Mode in Emacs
James Clark's free Java parser XP. Note that this does not validate XML, only parses it.
DocBook Tool for Linux: Converts docbook documents to a number of formats. NOTE: I only got these to work with Docbook SGML, NOT with Docbook XML. If you are able to make it work with our XML, please let us know.
AptConvert from PIXware is a Java editor that will produce DocBook documents and let you transform them into HTML and PDF for a local preview before you submit.
In the process of transforming your HTML into XML, HTML tidy can be a handy tool to make your HTML "regexp'able". Brandoch Calef has made a Perl script with directions (now via archive.org) that gets you most of the way.