The M040 release of Arbortext Version 5.4 is available for download from the support site. You can view the full Release Notes here (login required) and download the latest release here (login required).

Here’s what the release notes say about Version 5.4 M040:

Arbortext Editor

• Using the Insert Division toolbar button to insert division structures in a document no longer inserts incorrect division elements. (SPR 1935911)
• Double-clicking on the starting tag of an element which contains an ID attribute value now selects the element if no references exist for that ID to link to. (SPR 1941186)
• When finding content with the Find/Replace dialog box, included objects are no longer automatically locked and documents are no longer unexpectedly expanded to show their entire hierarchy. (SPR 1941966)
• Namespaced MathML markup is no longer removed from DITA topics during composition or when generating a resolved document for styling. (SPR 1942309)
• The pending_delete callback is now properly called in response to a replace done using the Find/Replace dialog box. (SPR 1944087)
• Certain repeated searches of tables no longer results in Arbortext Editor terminating unexpectedly. (SPR 1951745)
• Userules no longer fail to fire when more than one userulehook is registered. (SPR 1952704)
• Unknown attributes on Arbortext-specific tags are no longer treated as invalid attributes. Doing so prevents older versions of Arbortext Editor from reporting errors in documents saved by more recent versions of Arbortext Editor that contain new attributes on the Arbortext-specific tags. (SPR 1960250)
• When performing silent installations (using msiexec /quiet /i arbortext.msi) all standard file associations are now properly set. (SPR 1962806)

Composition and Publishing

• Fixed an issue with previewing and publishing documents containing certain compressed TIFF images. (SPR 1935804)
• Publishing DITA documents containing properly-named Astoria namespaced attributes no longer causes errors during composition. (SPR 1955029)
• When publishing using the Arbortext Publishing Engine, graphics now render properly for documents with document types that specify graphics must be referenced using entity references instead of file references. (SPR 1955767)
• Publishing documents to print or PDF using a FOSI generated or exported from Arbortext Styler no longer results in buffer overflow messages when the ACL function fosivar_value() is called from an ACL system-func to get the value of a FOSI variable that has no value. (SPR 1959336)

Arbortext Styler

• The Attribute Model for Indexing (for example, as used in the CALS DTD) now works properly for Arbortext Styler chunked (for example, web) outputs. (SPR 1945446)
• Certain controls in the Arbortext Styler user interface now render CJK text properly. (SPR 1952494)
• In Attribute Tests in Arbortext Styler stylesheets, undeclared elements and attributes are now always processed the same as declared elements and attributes in FOSI-based outputs. Previously, undeclared elements and attributes worked properly only when they were namespaced or when they where used with freeform XML documents. In FOSI stylesheets, specvals and fillvals where attloc contains an undeclared element, or attval contains an undeclared attribute are now processed the same as if the element or attribute were declared. Previously, these specvals and fillvals worked properly only when they were namespaced or when they were used with freeform XML documents. (SPR 1957126)

Import/Export

• Exporting certain documents to RTF no longer produces empty RTF files. (SPR 1943004)
• When importing RTF documents, Word built-in document properties are now properly recognized. When importing FrameMaker MIF files, text inside anchored frames is now always properly imported. (SPR 1952068)

PTC Server and Repository Adapter

• The pending_delete callback now properly handles delete operations depending on the object types in the buffer. (SPR 1944087)
• When multiple users are using the PTC Server connection to work with different Windchill workspaces in offline mode on the same workstation, messages about failing to connect to the server and being unable to download content are no longer displayed. (SPR 1954438)

The M040 release of APP Version 10 is available for download from the support site. You can download the latest release here (login required).

Here’s what the release notes say about Version 10.0 M040 (Build 1189)

• Fixed issue where either <?tpfitbr> or autoleading caused the last line of a para to overflow and then widows and orphans were being ignored. (SPR 1865826)
• Fixed column headers for multiple levels of nesting (SPR 1664778)
• Add support to allow blocks indents to be absolute (relative to margin) using <?blockindentabs>. (SPR 1805054).
• Fixed a number of cases involving column footers and nesting (SPR 1664778)
• Fixed which top margin is applied following a skipped column header or footer. (SPR 1949036)
• Added FOM Accolades.
• Fix crash with contextdown. (SPR 1952453)
• Fixed namespace support in Strip Space control streams. (SPR 1877605)
• Added support to preserve children in Strip Space control streams (preserve 1,”name”). (SPR 1877605)
• Fixed issue where widow lines weren’t being counted properly at the end of a block or table. (SPR 1954407)
• Fixed issue where the orphan priority was ignored for the last paragraph if the keeps were too large (SPR 1954409)
• Fixed handling of <?tbwmode=2> tables splitting horizontally and then overflowing when nested.(SPR 1954968)
• Fixed problem with psequp getting triggered twice after a blank page. (SPR 1938805)
• Improved error messages with invalid parameters to FOM fFormatting.contextAdd() and contextRemove().
• Fixed issue so accolades correctly ignore the value of <?tpxbrk>. (SPR 1907764)
• Added support to combine footnotes by reference, using combine=2 in the footnote control stream.(SPR 1921186)
• Fixed pointer error when using a footer within a table that is nested within a table that also contains a footer. (SPR 1936268)
• Fixed crash in colour dialogue when colour name is longer than 40 characters. (SPR 1913425)
• Fixed cursor positioning in very wide edit bar. (SPR 1917906)
• Fixed problems saving OpenType fonts to a bundle. (SPR 1919464)
• Fixed problem with <?bk> on end of word at end of line. (SPR 1907456)
• Fixed problem with multiple ligature table changes in a word at end of line. (SPR 1954907)
• Fixed problem with fish loops crashing when using old show string syntax. (SPR 1958757)
• Modified APP installer to make each patch version install independently to any other v10 version. (SPR 1901316)
• Added support for fStream.control with frame, layer and layergroup types.
• Fixed graphic state problems when outputting EPS files to PDF. (SPR 1958767)
• Fixed crashes and timing issues with fApplication.loadTemplate(). (SPR 1963511)
• Fixing frame corruption when setting FOM fFrame column widths (SPR 1963512)
• Updated FOM Reference Manual sample to include v10.0 M040 changes.

In a tough market, you need to stay ahead the pack. The best way to do that is to make sure your skills are in demand. One of the highest demanded skills in the publications arena is Arbortext. We are offering this one day session for you to learn what it takes to create DITA structured documents using the leading industry tool: Arbortext Editor.

Reserve your seat now

Current Seminar:

“Introduction to Arbortext Editor and the Arbortext DITA Application”

Details:

What is Arbortext?

Next-century tools for dynamic product information development. The Arbortext product suite offers solutions at every stage of information development from defining, authoring, illustrating, to managing, and delivering content.

Arbortext is an end-to-end product information delivery system that automates the authoring and publishing processes for the delivery of high quality, tailored product information in the form of operator manuals, service documentation, and eLearning courseware.

What is DITA?

DITA is one of the most important innovations in XML publishing in years

Short for “Darwin Information Typing Architecture,” DITA is an IBM invention that the company contributed to the community under the auspices of OASIS. DITA is an architecture based on XML for publishing technical information, and you can think of DITA as having three distinctive characteristics:

• DITA is a data model – In other words, DITA defines a DTD and an equivalent Schema that prescribe which tags you may use. What makes DITA unique is that you create your own tags using a mechanism called “specialization” that requires you to define new tags by basing them on existing tags.
• DITA is a methodology for reuse - Unlike book-oriented approaches to data modeling, which lead authors to continue writing information in a monolithic form, DITA is a topic-based approach to authoring that leads authors to write small, standalone modules that can more readily be reused in multiple document types.
• DITA is an information architecture - To communicate information with the greatest effectiveness and efficiency, DITA defines an approach to writing and presenting information that embodies decades of experience. DITA specifies that technical information be classified as a Concept, Task or Reference, which guides authors to write more specifically with the fewest possible words.

Arbortext & DITA

As the first vendor to support DITA at IBM, the first vendor to support DITA editing, and the only vendor to participate as a founding member of the OASIS DITA Technical Committee, PTC has led the way for DITA adoption. DITA was developed on Arbortext.

Arbortext DITA provides access to all these capabilities:

• Conversion capabilities for legacy content
• Complete authoring support for maps and topics
• Link management
• Integrated support for review and approval cycles
• Conditional text to produce documents that meet the needs of individual readers
• Reuse of content across multiple documents
• Integrated functionality for content management
• Localization
• Rich publishing capabilities for any media, including print and electronic

Covered in the seminar:

• Getting Started
• Background & Resources
• Structured Application and Authoring Concepts
• Introduction to
• Arbortext Editor
• Working with Structured Markup
• Structured Information and Document Structure
• Creating Tables
• Introduction to DITA and the Arbortext DITA Application
• Creating DITA Topics, Concepts, and Tasks
• DITA Cross-References, Links, and Content References
• DITA Maps and Relationship Tables

You can deepen your understanding of working with DITA by attending the Content Management Strategies Conference right before this seminar, April 19-21 in Santa Clara. Liz Fraley will be presenting an case study of long-term Arbortext Customers.

You can learn more about Arbortext by joining the SF Bay Arbortext PTC/User Group or by visiting the Community Resources page hosted by Single-Sourcing Solutions.

This seminar is designed by Liz Fraley, software engineer, technical trainer, and secretary of the SF Bay Arbortext PTC/User Group and is based on the Certified PTC training courses offered by Single-Sourcing Solutions.

Trainer bio:

Liz Fraley

LIZ FRALEY, Single-Sourcing Solutions, is a long-time Arbortext user, customizer, stylesheet developer, and implementer. In addition to being a senior software architect, she is the founder of Single-Sourcing Solutions, a qualified, experienced reseller of Arbortext Products. She has over 15 years experience programming, teaching, and supporting knowledge growth in the community, including more than a decade of experience with the Arbortext product line. She has presented at industry conferences including Content Management Strategies and the PTC/User (formerly AUGI) conferences annually since 2003. She holds degrees in Computer Science and English from the University of College Park and a Masters in English from the University of Southern Mississippi. Single-Sourcing Solutions provides software and services to support the implementation and operation of a DITA-based publishing system: Arbortext.

Reserve your seat now

Or, It’s not just about the technology

A techpubs department is a techpubs department and one is not so unlike another. At a certain level that’s true, but the devil’s in the details as we all know.

I’ve found, over the years, after watching countless techpubs departments implement single-sourcing solutions of all kinds and using products from nearly every vendor: Those who go through a true, enterprise-wide discovery process have a far higher probability of, not simply succeeding, but realizing their return on investment. Not only do they realize ROI on the technology, buy they see that return in their staff costs, their implementation costs, and everything related to getting a project like this done. In addition, we’ve found that those who don’t go through this process, end up spending a far greater amount of time and money in the end than those who do.

Choosing an end-to-end system like Arbortext, discovery is critical for us and for you. To assure your successful implementation over the long haul, you want to understand three things:

1. The technology
2. Your existing processes and business goals
3. Your people, skill sets, and their interests

We all know that when you implement Arbortext, it’s not just about the technology. In fact, the technology is only one of the three major legs supporting an implementation like this which has lasting, process-level change effects on the entire enterprise. You’ve got to look at the processes you have in place today, your organizational culture, and your people—the skill levels, what you have, what you need. It’s not just the technology you’re looking at: You’ve got to look at all these different things, together.

Asking for a demonstration before you understand exactly what it will take, from an organizational perspective, to implement single sourcing in your enterprise is no different than watching a video or reading a white paper (both of which take less time and cost less in terms of your resources).

Rushing into a demo reduces the conversation to the technology, internally and externally. Management will give you less resources, reducing the chance of successful implementation because your business case is weak. It’s technology-focused rather than business focused. (And, let’s face it: business-focus is what drives budget allocation.)

In addition, a generic demo doesn’t serve you very well. Vendors show generic features and maybe they hit on the one thing you’re looking for and maybe they don’t. You and your team sit through an hour (or two or three) of features and functions of a very complex system and set of applications. (If you’ve got 8 people sitting in a 1-hour demo, you’ve just wasted 8 hours.) Looking at generic features, you can’t possibly know what ripple effects the technology that you didn’t see will have? You also don’t know what will help bolster your case because a feature you didn’t see helps someone else in the organization.

In all the work we’ve done with customers, we’ve often found that while other parts of the organization currently don’t use the documentation in it’s present form there is a desire to do so if the documentation could be easily customized or configured for their purposes.

Don’t get me started on the ripple effects of seemingly simple process changes.

The goal of any vendor evaluation is not to see whether the product works. Software always has bugs; there’s always a learning curve; and, unless you’re dealing with an out-of-the-gate startup and haven’t spoken to the vendor’s references, then you know it’ll “work.”

What you’re really looking to understand, in the most efficient way possible is this: What will let you advance the sales process or end it. What will get you what you need or get the sales person to stop calling.

What tells you that?

A clear and precise understanding of exactly why Arbortext applies to your unique situation, processes, people, and business needs. Without that, a demo doesn’t add value to your conversation.

The latest PubWright Podcast has been posted to here and on iTunes here.

This week’s podcast features Kevin Dietz from Timpani Software. Timpani Software develops BuildBeat, a comprehensive automated software build management system, and MergeMagician, an automated merging server designed to work with your existing software configuration management (SCM) system.

…is really more of a pyramid

Side #1 – The Theory Side: The Information Management Consulting Companies Side #2 – The Technology Side: Programming Nuts and Bolts Side #3 – The Product Side: Vendors and Application-Specific Specialists

There’s a new way you can read the Single-Sourcing Blog. That is, if you own a Kindle. It is not free. The charge is $1.99 a month and you can try it out for 14 days at no charge. If the Single-Sourcing Blog is worth a couple bucks a month and you own a Kindle, Voila! 

If you don’t know what a Kindle is, I’ve included a video about it at the end of this post. In short, it’s electronic wireless reading by using the same connections that cell phones use. Devices like the Kindle are bringing news and more to yet another digital format: one that is light, easy to carry, and saves a few trees.  You can read more about the device and what it costs here.

I’ve added a subscribe widget to the blog as well:  

For us here at Single-Sourcing Solutions, the benefit is obvious and easy: It’s just another example of another output format — another rendition of your content.

Kindle Blog Subscriptions

• Kindle Blogs are auto-delivered wirelessly to your Kindle and updated throughout the day so you can stay current.
• It’s risk free—all Kindle Blog subscriptions start with a 14-day free trial. You can cancel at any time during the free trial period. If you enjoy your subscription, do nothing and it will automatically continue at the regular monthly price.

Here’s a snippet from the PTC Express newsletter. You never know how well something will get saved and archived, so we’ve reproduced it here. Generally, we post code to the Adepters Code Archive. However, this article went out from the PTC Express news letter and it’s getting some attention in places that don’t really understand the impact or how to present the information as the code that it is. So, we’ll join the club, improve the article, and continue adding value to the community at large.

Let us present:

Creating an Update Button in Arbortext Digital Media Publisher (DMP)

It’s a fairly straight-forward process to add a button to Arbortext Digital Media Publisher (DMP).  This is a gentle introduction to getting started with customizing DMP. In March, the Arbortext PTC/User group meeting was all about DMP. There’s some amazing things that you can do with it. The full application is customizable.

DMP is part of the Arbortext Publishing Engine.  If you don’t already know DMP, you should get to know it:  DMP assembles documents and multimedia files for publication. Out of the box, DMP provides capabilities for distributing updates to your digital media via the web.

This article describes how to create an update button that your customers can select and that will initiate a download all of the updated pages on your website and store them locally for later offline viewing.

Arbortext Help Center is produced with DMP by PTC documentation staff

Creating the button that performs this functionality is not completely intuitive. Here’s what you need to do to create the “Update” button with DMP.

First: Configure the dmp.properties File

To make certain that the button will work properly once it is placed into the DMP image, set the following options in dmp.properties file to enable update pack generation:

# updateStatus: how the update should be proceeded. There are three possible values
#
# 0:  Disable update, no update is generated (but update button and
#     configuration is available in Consumer).
# 1:  Manual update: Consumer user can update content by clicking
#     the update button
# 2:  At launch: Consumer will check for update each time the application
#     is starting
# 3:  Auto update: Consumer will update at scheduled time. updateSchedule
#     property must be set.
#

updateStatus=0

# updateFrom: the version that the update is based on. The version must
#             be older than the version of the current project. This
#             property uses the same format as the "version" property
#             defined above, e.g. 1.0, 2.0.1, etc. It is an error if
#             the version is the same or newer than the current version.
#

updateFrom = 1.0

# updateUrl
# The url pointing to the folder that contains update
# manifest file. It can be a remote URL or local directory.
#
# For a remote URL, it should always starts with "http" or "https".
# For a local directory, if a relative path is specified, it is relative
# to $CONSUMER_INSTALL_DIR/project/WEB-INF. For example, if updateUrl is
# set to "../_updateFolder", it indicates
#
# $CONSUMER_INSTALL_DIR/project/_updateFolder
#

updateUrl = updatepack

# updateConfigurable
#
# Whether Consumer user is able to configure update properties.
#
# true: show update configuration page in Consumer
# false: don’t show configuration page
#
# Note: update configuration page will not show even if this value is set
#       to true if
#       1. Consumer is not installed, or
#       2. Web deployment and the request is not from localhost
#

updateConfigurable = true

# updateSchedule
# The schedule to run update. This property value is required only if
# updateStatus is set to "3". The updateSchedule has the format of
# [UpdateType]:[DayOfMonth]:[DayOfWeek]:[HourOfDay]
# UpdateType: d (daily), w (weekly), m (monthly)
# DayOfMonth: 1-31
# DayOfWeek: 1-7
# HourOfDay: 0-23
#
# For example, to run update at 3am every day, the updateSchedule should
# look like: updateSchedule = d:0:0:2
# To run update at 3am every Friday: updateSchedule=w:0:5:2
# To run update at 3am the fifth day of each month at 3am:
# updateSchedule=m:5:0:2

updateSchedule=w:0:1:0

Second: Check other properties in the dmp.properties File

Make certain the following property values do not change between the base release and all updates. Changes to these values will cause updates not be properly recognized.

• companyName
• projectId
• projectName
• outputroot

Third: Create the Button

To create an “Update” button in DMP, follow these steps:

1. Create a new “Welcome” page by opening up your text editor and placing the following template code at the top:

   <?xml version="1.0" encoding="UTF-8"?>
   <jsp:root xmlns:jsp="http://java.sun.com/JSP/Page" version="2.0"
             xmlns:c="http://java.sun.com/jsp/jstl/core"
             xmlns:fmt="http://java.sun.com/jsp/jstl/fmt"
             xmlns:atidmc="http://www.arbortext.com/namespace/dmc/consumer">
      <jsp:output doctype-root-element="html"
               doctype-public="-//W3C//DTD HTML 4.01 Transitional//EN"
               doctype-system="" />
      <jsp:directive.page language="java" errorPage="error.jspx"
               contentType="text/html; charset=UTF-8" />
      <atidmc:layout var="layout" />
      <html>
         <head>
            <title>Cerify Automated Verification and Checking System</title>
         </head>
         <link rel="Stylesheet" type="text/css" media="all"
               href="${layout.template}/css/dmc.css"></link>
         <body>
         <div id="config">
   

2. Under the <div> tag, place the content you want displayed to your users as the first page of your DMP image. For example:

   <p>The DMP Help Center itself is an application of DMP. You can search
   the documentation using the search controls on the upper-left side of
   the window. The Help Center also provides an index to the HTML and PDF
   documentation. (To view the PDF documentation, you must have Adobe
   Acrobat Reader installed and correctly configured on your system.).</p>
   

3. Directly under this new content, add the following code:

   <form method="get"
         name="update"
         id="update"
         accept-charset="utf-8"
         action="../_update"
         enctype="application/x-www-form-urlencoded"
         target="_parent">
      <input type="submit" name="update" value="update"/>
   </form>
   

4. Remember to keep your tags balanced and make sure you close the code with the following:

         </div>
       </body>
     </html>
   </jsp:root>
   

5. Save this file as:

   <DMP_INSTALL_DIR>\templates\simple\yourcompany.jspx
   

   Note: If working with frames, choose the frame directory.
   

6. Open the template.xml file.

   <DMP_INSTALL_DIR>\templates\simple\template.xml
   

   Note: If working with frames, choose the frame directory.

7. Change the Page type=”Intro” to the following:

   <Page type="INTRO">
   <Url locale="*" href="yourcompany.jspx" />
   

8. Run the dmp.bat file. You should see your button displayed. When you click this button, DMP will check for updates.
   

   see the results

   You should see that the content shows “current” because an update pack has not yet been generated.

9. Create and deploy the updates pack.Click the update button again. You will notice a series of steps being performed by your image followed with a completion status.Your content is now up-to-date. Practice adjusting the duration of the updates by opening the dmp.properties and setting the updateSchedule to different codes. The default code is:

   updateSchedule=w:0:1:0

   Setting the update to Schedule as described above will set the update to run on the fifth day of each month at 3am.

   The update button will work in both standalone mode and Web application mode.

This month the implementation date for MDD directive (2007/47/EC) comes due. There are a number of issues that require action by manufacturers (changes to the Essential Requirements, clearer requirements for clinical evaluation, changes to the QA of third parties) and the Notified Bodies ( higher requirements for auditing Technical Files).

After seeing how well positioned Medtronic is, we started wondering how well the other device manufacturers were doing. Interestingly enough, none of them have really started the discussion about the benefits of dynamic information delivery in their business.

I might have expected it. The first step is always: let’s meet the requirements and be done with it. But when translation costs are $50/page when they could be less than $5/page, it makes me wonder when the life sciences companies are going to really join the discussion.

We had a customer in the manufacturing space — not life sciences — and their translation vendor offered them a steep discount if they got out of FrameMaker. Their translation costs went from $20,000 to $6,000.  Just getting out of proprietary systems and into a standard XML publishing system got their costs down by 2/3.  (Reuse is the second driver of cost reduction.)

Is there something about this I’m missing? Medical Devices have a huge margin, why wouldn’t any medical device manufacturer want to maximize it as much as they could while improving their odds against failing an FDA audit due to product information development processes?

Recently, I was privileged to talk with a customer that implemented Arbortext in 1999: Greg Johnson at Medtronic. Greg heads up what his executives have told him is “the most successful project Medtronic had ever done.” We’ve already talked about how dynamic information delivery systems can really drive down translation costs. This time we’re talking about what reuse brings to the table.

It’s important to remember that this is the value that Medtronic has come to after 10 years of investment in the dynamic information delivery system. They continued to develop it over time and to focus on the things that drive those numbers down.

What drives it down?

According to Greg, it’s reuse. Their reuse is at 90%. Their writers are reusing written, translated, validated translations that they can prove are untouched and reused as is.  In their system, they can certify, based on the validity of the system, that they are using exactly the same content. They haven’t touched it; there are no changes: they haven’t created a revision that triggers other processes and other costs.

They’ve seen the translation savings numbers trend downward for for years. However, even if you’re investing more for the authoring side, you still see the win in each language. You still see it even if you’re only producing documentation in English. You still get the win because you’re still reusing modules rather than rewriting, validating, and producing large amounts of content.

At Medtronic, just on the authoring side, the data is equally impressive:

* Before: the first manual for a new model would see 95k new words.
* After: new models have 55K new words; all the rest of the content comes from reusable modules.

Writers and management collaborate to plan documentation for new models taking care to look ahead to other models, so they keep getting win after win for follow-on products.

Today, Medtronic has over 100 models, with variations on features and things that are all over in terms of price points.  Greg’s team can crank out the manuals for new models right and left. All the features have been netted out to all 100 models and beyond.

Over the last 7 years, they’ve been tracking what it would have cost if they hadn’t had a dynamic information delivery system. They’re keeping a running total. It’s a long process to see the return on the up front costs, but the return is utterly dependent on the scope of the problem.

Greg doesn’t sugar coat it. He admits that it’s a painful process, a long fight, but there are awesome returns. “Reuse is king,” he says, “and you should push for it; don’t start on authoring side thinking you will band-aid in localization later, understand up front, partner with localization partners to do this right.”

See previous article: Dynamic information delivery systems reduce translation costs