Really Strategies

• Articles
• Presentations
• White Papers
• Newsletters

Oh Really! 5 Questions With...

by Ed Stevenson    Nov 2005

In each issue, this section presents an interview with someone in the field of content management. In this issue we interview Gary Cosimini, business development director of creative professional products for Adobe Systems Incorporated.

[1] We've heard and seen many good things with Adobe's adoption of XML within the Creative Suite product line. What are some of the highlights of the products' XML capabilities, especially within InCopy and InDesign?

Gary: InDesign and InCopy support can depend on XML in many different ways; most XML experts are surprised when they see how far Adobe's support for XML goes.

Import and export: InDesign CS2 can import XML files into a document for manual placement or, if the XML's structure corresponds to one that exists in the document, into a template or a model for autoflow or to replace existing content. You can create or load tagsets, tag content, view the tags, edit them, and export a well-formed XML file. You can also tag text in InDesign's tables. You can load a DTD and validate the document against it. However, there are some limits to InDesign's Structure editing if the DTD is complex, and InDesign does not enforce the DTD during editing.

InCopy: InCopy is a version of InDesign for writers and editors. It's used to edit text without changing the document layout. An InCopy file is actually an XML file, tagged with a proprietary InDesign tagset and used for several things by InDesign: saving backwards from CS2 to CS, for Snippets, which are InDesign page parts that can be shared, and to exchange text with InCopy. Dozens of editorial workflow systems from developers like DTI, SoftCare, WoodWing, BaseView/Harris and others, incorporate InCopy. Because all the text in their systems exists in InCopy, it can be extracted at any time in the publishing process as XML by means of XSL transformations.

Style-to-tag mapping: InDesign and InCopy can map character and paragraph styles to tags and vice versa. Along with a configurable Word and Excel import feature, content can be converted to XML from Word or RTF documents.

Scripting and extensibility: Everything in InDesign and InCopy is scriptable through AppleScript on the Mac, VBScript on Windows, and JavaScript on either. Every part of the XML tagging interface may be addressed through a script, which allows a great deal of customization. If you can describe something you'd like scripted, you probably can. For our C++ developers, InDesign features over 25,000 API or programmer interfaces, to create their own hooks

XMP: XMP stands for eXtensible Metadata Platform, a format for adding custom metadata to any file type. It's a basic feature of all the programs in the Creative Suite, including Acrobat, and lets you tag any file according to your own schema, or use the schemas that we provide with the application, and even add custom user interfaces to read and write your tags.

[2] And what are some areas in which Adobe still sees room for improvement?

Gary: There are a hundreds of things we'd like to do better with XML! Some of them are on the feature list for the next version, so I can't really talk about them, but many of them concern ease of use, smarter automatic flow and better control over incoming and outgoing XML.

[3] A growing area of interest is the use of XMP to store metadata for binary objects. What benefit do you see from using XMP and what are some practical applications for it?

Gary: XMP is truly the "missing link" in the asset management ecosystem. Formerly, metadata was stored in databases. Each database had its own schema. Connecting two databases required a communication layer that was fragile and awkward. To transfer a file and its metadata from one system to another required multiple transactions. And each database developer had a natural desire to make theirs the authoritative repository for the crucial business information. These factors contributed to a major problem: using and sharing metadata was almost impossible.

I think of XMP as "PDF for metadata," but better, because it's all based on XML. Here's a universal way to package information about content, classification, business transactions and workflow that can co-exist with content. It's a perfect companion to digital photography, digital advertising, electronic publishing, and document archiving and retrieval of all kinds. I honestly see a future for XMP similar to that of PDF.

[4] Why did Adobe offer XMP as a public standard, as opposed to keeping it proprietary?

Gary: Because it's based on a W3C standard, RDF, it was logical for Adobe to immediately make the specification public. We also put the software that we created to read and write XMP files into open source license, both to encourage other developers to adopt the specification and to assure the quality of XMP that was created or edited by third party software.

Adobe has learned a great deal from the trends in the software world over the past decade, and we want to be looked upon as good citizens. Remember, we charged for Acrobat Reader when the product was launched, and it didn't go anywhere. We learned from that.

[5] Does Adobe see any potential in using XMP beyond binary objects, such as embedding XMP in text based or word processing files?

Gary: Yes. In fact, that's one of the things our customers have been asking for. You can already use XMP with InCopy files, for example. Would it be interesting to see XMP become part of another document's schema? I'm sure there are ways to use XMP we hadn't anticipated. For example, take a look at the Stock Photo service within Adobe Bridge, part of the Creative Suite. Here XMP metadata is used to communicate business information both between buyers and sellers of stock images, and also by InDesign, which lets you know when an image is a "comp" version and how to license it at full resolution. The imagination of its users will guide XMP's applications in the future.