Welcome to MSDN Blogs Sign in | Join | Help
Hello. I'm back from parental leave but am now moving to another group within Microsoft. Please refer to Jeff Bell's blog for information about saving to PDF or XPS within Office "12."

This may be the shortest-run blog ever, but I am beginning parental leave very shortly and Jeff Bell, a Lead Program Manager in Office, has kindly agreed to continue the Save as PDF discussion on his blog along with the other topics he is covering. He has been on the project since the beginning so will be an excellent resource for information and questions.


Thanks for your interest; I’m looking forward to continued discussion when I return!

Thanks for your comments and questions about PDF—I appreciate hearing about how you are using PDF today and what solutions would work for you. I’ve answered a few posted questions below (if you’ve been reading Brian's blog, you’ll have seen most of these but I wanted to reiterate and update them here), and I’ll post more over the next few weeks.

Is this a one-way publish only, or do we support opening and editing PDFs as well?

Save as PDF is a one-way "publish" operation only. We are neither shipping a special viewer nor doing any work to make PDF files readable or editable by the Office applications.

Why is this built directly into Office as opposed to a print driver available to all applications?

Because this functionality is native to Office applications, we can capture richer document data further upstream than we would be able to do with a print driver. For example, we annotate our PDF output with tags for accessibility and we support internal and external hyperlinks. An additional advantage to capturing document data further upstream is that we render some visual effects, such as transparency and gradients, better because we are working with native information that has not been converted to bitmap for printer output.

Do we support tagged PDF (for accessibility)?

Yes, Office PDF output contains tags to assist with reading order for screen readers, and alternative text on images and on rasterized (bitmapped) text (to avoid interrupted text flow when portions of a text run are rasterized due to text effects or other document constructions). We also tag documents and, in some cases, text runs with language tags to assist with accurate pronunciation, and we export Unicode text for nonstandard glyphs to enable screen readers to access the actual text content. I appreciate the feedback we have received in this area as to what tags people would like to see implemented.

What choices were made with font embedding/subsetting/outlining?

For our Standard intent, we both embed and subset embeddable fonts (when permitted). For our Minimum Size intent, we do not embed "Web safe" fonts, but we do embed and subset all other embeddable fonts. We are not doing font outlining in the general case.

Do you support automatic bookmarking?

Yes, document structure is implemented for some applications to support bookmark navigation in PDF viewers. In Word, for example, you have the option to generate bookmarks by page, by heading, or by Word bookmarks.

Okay, first post. I’m Cyndy Wessling, and I’ve been at Microsoft for eight years and have been a program manager in Office for the past four years, with the last year or so spent working on the Save As PDF feature among other things. This topic has already gotten off to a great start on Brian Jones’ blog, and I’m looking forward to addressing some of the questions and comments I’ve seen posted there.

For me and many of the people on my team (all of the Office teams, actually), this has been a long-standing customer request that we’ve been wanting to satisfy, and we are excited to hear your thoughts about Save as PDF, so please let us know.

To expand a bit on Brian’s introductory PDF post, the addition of Save as PDF to Office “12” supports overall themes of content sharing and workflows and supporting complete customer scenarios around content. Office-generated PDF fits into these scenarios in a couple of ways: distribution and sharing, where customers can easily create PDF versions of their Office documents for distributing broadly, either online, via e-mail, or printed, and archiving, where customers can preserve their work for later retrieval and reuse. And, as Brian points out, this fits in well with the overall direction we’ve been heading with the Office file formats. It’s why we decided to move to Open XML formats as the default formats, and it’s what motivated us to build PDF support directly into the products.

In order to support these scenarios, our goals for PDF generated by Office “12” are that it: maintains fidelity to the original, is printable, is accessible, is navigable, and supports the PDF features important to the success of these scenarios.

Here is a quick overview of features in Office-generated PDF:

Native PDF creation—Saving as PDF is supported from within Microsoft Office “12” applications Word, Excel, PowerPoint, Access, Visio, Publisher, InfoPath, and OneNote.

Quality/intent settings—Most applications include support for two “intent” settings: Standard, for publishing files online and printing, and Minimum Size, for online sharing. The intent settings control the level and type of image compression for various image types and font embedding and subsetting options. Publisher has additional intents for commercial printing, and I will devote at least one future post to discussing Publisher’s commercial printing capabilities.

Internal and External Hyperlinks—PDF documents preserve internal and external hyperlinks assigned to text and other objects in the original file.

Tagging and accessibility features—PDF for most applications includes basic document structure with tagged content elements. The tags support logical reading order, alternative text on images and on text that is represented as an image in the PDF output, and Unicode representation for nonstandard glyphs.

Document outline—AKA “bookmarks” used for document navigation in PDF viewers.

Document properties—Metadata properties associated with the file.

Future posts over the next few weeks will go into more detail in these areas, as well as some of the specific questions I’m seeing on Brian’s blog. Of course feel free to post comments with anything else you’d like to see here.