With the increasing spread of PDF/A as the ISO standard for long-term archiving, unfortunately a few misunderstandings have been popularized as well. After nearly four years, some DMS providers still seem intent on “riding out” the PDF wave. But as I see it, the saying applies that “you snooze, you lose!”

Myth #1: TIFF is secured against tampering, PDF and PDF/A are not

This assertion is clearly incorrect. There is no document format which is inherently secured against alteration and compliant with auditing requirements. A TIFF-file can be modified with simple tools just like a PDF/A document or any other format. “Inalterability” of documents can only be achieved using a signature. If files must be archived in compliance with auditing requirements, then a system or process is necessary to ensure protection against changes.

Myth #2: PDF is a standard from one provider, TIFF is a disclosed standard

Yes and no.  TIFF is a de facto industry “standard”, but it has never been standardized by an international standards organization such as ISO or DIN. Both PDF itself (ISO 32000) and PDF/A (ISO 19005) are disclosed ISO standards and are thus not only de facto but also de jure standards.

Myth #3: PDF/A does not support signatures

On the contrary.  PDF/A even permits embedded signatures – including qualified electronic signatures. To do this, the signature provider must simply apply the product in a PDF/A-compliant manner, but there are still some signature providers who have not yet accomplished this with their products.

Myth #4: PDF/A does not support compression

Wrong. PDF/A permits all common compression methods to be used, such as JBIG2, JPEG, etc. The exception is LZW, where at the time of the standardization patents were still in force. For these reasons of time, JPEG2000 was not incorporated in the PDF/A-1 standard, but it will be covered in the new version (PDF/A-2).

Myth #5: PDF/A does not allow OCR for scanned documents

Wrong. OCR is possible in both PDF/A-1b as well as PDF/A-1a, of course. A minor point – perhaps the cause of the confusion - is the exception that this invisible font does not have to be embedded.

Myth #6: PDF/A files are too large due to font embedding

Yes and no.  It is true that fonts (except for OCR) must be embedded. Based on practical experience, this is only a problem in the particular application area for bulk outgoing mail. In this regard, one can apply font reduction and subsetting or pragmatically omit font embedding in a solution tailored to individual company needs. These files are then no longer PDF/A-compliant in a strict sense. However, except for the deliberate exception they retain all the advantages of PDF/A.

Myth #7: PDF/A does not support metadata

On the contrary. XMP particularly facilitates standardized metadata in PDF/A. Metadata can be managed in the surrounding systems as before. An advantage of PDF/A is that these data can also be embedded inseparably in the document.

Myth #8: PDF/A is not supported by DMS systems

Yes and no.  Simply put, an ECM system which can handle PDF can also support PDF/A well. However, (unfortunately) there are still a number of DMS providers wedded to their outmoded TIFF viewers, and that can sometimes be a stumbling block in practice.

Myth #9: PDF/A is only supported by a small group of local German providers

Not at all! It is certainly true that PDF/A was first accepted in German-speaking countries – and that the PDF/A Competence Center originated in Germany. However, in the meantime many countries and industries recommend PDF/A or even require it by statute. Moreover, the PDF/A Competence Center now has over 100 members from about 20 countries!

Myth #10 PDF/A is expensive!

Yes and no.  Of course the deployment of PDF/A tools requires an initial investment. Sometimes the ROI from highly compressed PDF/A files can be calculated within a few months even without an Excel spreadsheet, for example with the Sparkasse savings banks.  But that is perhaps more of an exception. The problem here is assessing the benefits: how much is it worth if unifying formats saves training time and expense as well as viewer license fees. And when fewer migrations are necessary in the future? And last but not least, how do you place a value on a “good” archive thanks to standardized PDF/A files?

Thomas Zellmann is an executive board member of the PDF/A Competence Center