View Document Solutions site map Document Solutions Home Page
Site Map Contact us
Document Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDF
Document Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDF
Document Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDF
Document Solutions - Getting the most out of PDFDocument Solutions - Getting the most out of PDF
Document Solutions - Getting the most out of PDF
Document Solutions Home Page

PDF News and Tips
January, 2005

Tune Up your PDFs!

You'll find examples of derelict PDFs everywhere, from the simplest to the most sophisticated websites.

Ever wonder why PDFs often get (and earn) a bad rap for usability? Missing navigation and interactive features (bookmarks, links, form-fields, optimal display settings, security and accessibility features) force every user to spend more time getting the information they came for.

What's the most common single complaint? Files are simply far too large, or their size is unaddressed and therefore poorly handled, contributing needlessly to server and network loads. The result is poor performance, which can exclude low-bandwidth users, and frustrate even those with lots of bandwidth. Worse, most crude attempts to control file size often leads to poor quality, fuzzy and distorted images.

What's the simple problem nearly everyone gets wrong? Believe it or not, missing or bogus Document Information metadata is one of the chief causes of user frustration. Why? From Google to IIS to other Content Management Systems, file metadata (usually, the Title) is used to display the file in search results to users looking for information. lacking a Title, the user often can't pick out their file from the other search results.

To ensure that you deploy quality PDF files, first determine precisely what you need them to do. If you envision them downloaded for future reference, for example, then you'll want to include links back to your website. If your users typically possess low-bandwidth connections, you'll spend more time ensuring your files are as small as possible. If good images are very important to your users (for example, with product literature), then you'll want to focus on retaining quality over shrinking the file- size.

Some other key considerations for preparing PDF files for online deployment:

Decide what version of Reader you need to support. Reader versions 4, 5, 6 and 7 give you different options for image compression and security, but not all users have upgraded. Perhaps you should prompt them to do so? Perhaps some of them can't upgrade?

Consider the length of the document. Longer documents have special characteristics that imply different handling. Bookmarks really help to alleviate user frustration. Longer documents, especially those with many images, can be much larger in size, so you might be choosing between a reliance on byteserving, delivery by the chapter, offering high and low resolution versions, or simply prompting users to download before opening locally. You may want to provide a "portal" PDF as well.

Make every PDF open in a way that appropriate to the content. For presentation PDFs, make them open Fit Screen. For portrait-oriented documents, Fit Width is usually best... unless the document opens to a full spread, in which case different options are in order. Users are put off by PDF files that don't "present" well.

Ensure Document Information fields are filled. Missing this step will compromise search results forever - don't do it.

To reduce file size, try higher compression before reducing resolution. Reducing image resolution is your last resort for reducing file size.

Perform a "Save As" as your last operation, and be sure Fast Web View is set to "Yes".

What's your PDF Strategy? Give us a call. We're here to help.

Find out about DSI's unique PDF Tune Up Consulting

SUCCESS STORY: The Journal of Light Construction

The Journal of Light Construction is the only 100% paid-subscription magazine read my remodeling and building pros nationwide. The staff and writers of the Journal, based near Burlington, Vermont, have spent almost two decades building a huge repository of advisory, educational and how-to information for the building and construction trades.

In 1998, the JLC began a program to leverage their content with a for-sale CD-ROM. The full-scale conversion of their legacy content from paper back- issues to publication-quality PDF files formed the bulk of the initial expense, but, JLC has covered this cost many times since.

A huge hit with the readership since introduction, JLC's CD-ROM now contributes a regular revenue stream, boasts it's own subscription base and helps drive web-site sales. Almost as importantly, the disc has helped cement JLC's reputation as a leader in serving the construction industry with timely and valuable information.

In 2005, the JLC will release the eighth CD-ROM set in the series. Spanning two discs, the collection may be installed to Windows or Mac computers, or simply accessed from the CDs themselves. Updated and expanded, the disc is set to make another substantial contribution to the bottom line for the 8th year running.

Want to know more? Read the Case Study. (PDF, 297kb)

Visit the Journal of Light Construction web site, and see how they market the JLCD-ROM.

Is there an opportunity for your publication? Learn more about DSI's solutions for periodicals.

What you need to know about OCR

There are a few simple truths about OCR that software manufacturers and service bureaus generally don't bother to mention. Keeping our eye on these balls is part of how DSI has stayed on the leading edge of publication imaging since 1996.

OCR for Search or for Extraction? These are two radically different requirements, but OCR software usually doesn't assist users in making this decision. If you are OCRing to make a document searchable, then automated OCR will often deliver acceptable results, particularly on simple or very clean documents. However, if you are OCRing to extract and reuse the text in Word or as XML, you should factor in the cost of a manual correction phase before the OCR work is done. It's far better to correct OCR errors using capable OCR software than afterwards!

Know your text. Have you really looked at the text you are trying to OCR? If there is handwriting present, you may need to aggressively clean the scans pre-OCR. Do you see lots of italics? These tend to be search terms on many types of content, and italics are tough on OCR engines. Consider scanning at a higher resolution (400 or 600 dpi). If you are OCRing articles, perhaps you'll need to capture the image-captions as well. Often, caption text is small, or located on a shaded background. If this text must be captured, you'll need to descreen your images and allow for some manual correction to assure good OCR results. Tables add other challenges, because the grid-lines can interfere with the OCR process. The only solution is to TEST your engine and EXAMINE the results closely before proceeding.

Know your images. It is vital to consider the implications of using color or bitonal scans or deciding whether and how they should be mixed. OCR results on color images are FAR more sensitive to image- quality than are black-and-white scans. Color scans are far larger, and require more attention to resolution and compression in the final product. Consider a process such as DSI's MultiResolution for documents with occasional color images.

Learn more about DSI's Imaging Services!

Get FREE PDF News & Tips
Email:
Get Adobe Reader 8.0 from Adobe.com 
Copyright © 2008, Document Solutions, Inc.