Write Once Read Many (Ipg Conference 2010)

Editor's Notes

• #2: Thank you for having me here today. Twitter opportunity, if relevant.
• #3: In developing this presentation, Bridget helped elicit nine questions that IPG members would want answered as a result of my talk. The broke down into two big groups: “Why tackle agile content” … (click) and “How do I make content agile?” Today’s presentation is structured along these lines, with time at the back end for questions. Oh .. We did have a ninth question – whose answer I’ll save to the end.
• #4: Let’s start with a baseline definition of XML Extensible – can be adapted or customized for specific types of content (click) Anyone signed up by the middle of last week received an e-mail from IPG a week ago with (among other things) links to two glossaries of XML terms and related buzzwords that we published on the Magellan blog. Both can be found at the link shown here, and they’ll remain up and available after the conference.
• #5: To provide a bit more background on XML, let’s compare how most books are created today to how they might be created in a more agile approach. At the top, you see how we typically work today, combining content, document structure and design decisions in a single, document-driven approach. This workflow readily supports one output opportunity per document, and repurposing can be hard. On the bottom, you see what Phil Madans of Hachette Book Group calls a “content-driven” approach. The source document still links content and structure, but applies deign attributes separately, supporting cost-effective output across multiple formats
• #6: Essentially, disengaging design increases agility, allowing you to use one source document to support multiple formats and uses
• #7: Although XML documents tagged for structure are very useful, certain types of books can be tagged for meaning, increasing their value across other new or recombinant uses.
• #8: Agile does come with a cost, though; most publishers have to become more standard in their content workflows. On the left, a “write once, read once” workflow allows you to do things many different ways, because rendering the end product, typically a printed book, lets you take many paths By comparison, an agile approach, as shown on the right, requires that publishers use reasonably well-defined processes to create content that really can be “write once, read many”
• #9: So why do I need to know about XML .. Three environmental factors
• #10: In preparing for today’s session, one of the questions asked was, in effect …. Short answer is “no” It’s not just about XML, though
• #11: I don’t know about you, but this is the only survey I have ever seen in which a group of editors agreed on one thing, let alone two.
• #12: You see this in other ways, as well. Here’s a chart that shows areas in which various publishing segments see promise. Fixing print-to-web processes is a priority at most imprints that are not professional; and everyone is looking at ways to do more book formats
• #13: There’s a lot of survey data, and I’m happy to provide links, but in the interest of time, a bottom line
• #15: So that’s the “why” argument; we’re switching gears to talk about “how”. Planning and implementation are key here
• #16: On this last point, I’d like to offer up what one of my IPG reviewers called a “scary chart”. I hope I can make it less scary for you. On the left, you’ll see one dimension, the ability to divide a book into stand-alone components, what we call “chunks. On the bottom, the axis measures the relative frequency or potential for content reuse. In the chart, we’ve listed types or genres of books. The placement is directional or representational, not absolute.
• #17: For certain books … here, cook books, education titles, reference, travel, among others … the likelihood of updating, reuse and components or “chnks” is quite high, and these are the types of books for which a substantial investment in content tagging might pay off
• #18: For other books, a significant investment in XML would probably be overkill. Novels don’t get reused or updated much, and they offer few or no chunks. They still benefit from XML’s ability to support cross-platform uses, but it makes sense to focus largely on tagging these types of books for structure.
• #19: Last year, we published a research paper, still available at oreilly.com, that outlined five considerations underpinning the business case for XML These included … The first four of these are publisher-specific, and today I am focusing on the ROI piece
• #23: We talked before about the Copernican shift, with content at the core. If you start at the 12 o’clock position, you see the print book and then e-book; these are most common direct outputs today. Increasingly though, other uses could be supported, if the conversions costs were manageable As you move further clockwise, those uses are spelled out: direct publishing to the web, increased use of large print, keeping content in print using POD technologies, and offering higher-value, annotated editions. There are other, untapped opportunities out there…
• #25: So, there’s a rationale, and an ROI … how do you do this? Three buckets (explained). The fewer titles you produce, the more likely you are to stay closer to the top of the list (commercial solutions, outsourced services, standard transforms) Multiple solutions (increasing in flexibility and cost)
• #26: Here’s the truly scary chart … I don’t want to make you XML experts in an hour, but that write once, read many model has a few complications – you need to use XML applications to transform the original document into other formats. Sometimes that takes one step, sometimes it takes more. The transforms are called style sheets, and they come in a variety of flavors: CSS (simplest), XSLT (changing one type of XML document into another) and SXL-FO (creating a formatted, printable document)
• #27: The more you invest in style sheets, the less you pay in prep costs
• #28: In effect, then, you want to match your investment in style sheets to the volume of work against which those style sheets can be applied. If you are small, this doesn’t mean you can’t use sophisticated style sheets. It does mean that you’d get a better return if you adapted where you could to use commercially available or shared style sheets.
• #29: In our research paper, we talk about a number of better practices uncovered in our research. This is an excerpt from the full list, and I’d like to focus on four items
• #32: This can take a while, especially if you are moving a backlist, as well
• #34: Here’s a practical list that ties back to the survey results
• #35: On the revenue size, there are also opportunities Before the last click… you’ll remember that I had a last question to answer.. What was it?
• #36: Given all you have heard, what do you think I am going to say?
• #37: Here are a handful of links and ways to get a hold of me. You’ll note that the first link is the one I mentioned earlier, the glossaries developed ahead of this presentation. Thank you for your time… With that I’ll open the discussion up for questions.