A week ago I acquired an updated version of my eCopy PDF Pro Office software with much more comprehensive facilities for creating PDF/A documents. Since then I’ve been exploring what those facilities are and using them on the files I’m converting in my PAW-PERS collection. The updated eCopy software provides support for PDF/A-1a, PDF/A-1b, PDF/A-2a, PDF/A-2b, and PDF/A-2u. Broadly speaking, PDF/A-1b seems to be the most basic level of conformance required and aims to achieve a reliably rendered visual appearance. PDF/A-1a supports additional features such as Tags and Language, while PDF/A-2 (which was published after PDF-A-1) also ensures that layers, transparency and embedded files are preserved. eCopy enables you to check whether a document conforms to any one of these standards, and I used this facility to check that the documents I was converting to PDF from other formats, complied with PDF/A-1b. On almost every occasion, even though I was using the eCopy software to convert the documents into PDF, the compliance check threw up errors – below are examples of some of the most common ones.

• xmp: CreateDate Bad XMP Date: ‘2015-01-27T09:56:01Z:P’ Page;1 Number;1 (The XMP metadata stream should conform to XMP specification);
• Mismatch between xmp:CreateDate (‘2015-01-27T09:56:01Z:P’) and CreationDate (‘D:20150127095601Z’) (The XMP metadata stream should conform to XMP specification);
• DeviceRGB used in image but no output intent (Device-specific colour space used, but no Output Intent is defined for the file)
• Output Intent missing (Non-Device-independent colour space is used but no OutputIntent is not defined).
• Missing PDF/A identifier (the PDF/A version and conformance level of a file shall be specified using the PDF/A identification extension schema in the XMP packet)

eCopy also provides a “Fix” facility which in most cases cleared the errors – though only if the resulting file was saved with a different file name. In some cases however, even the Fixed file still had errors in it which were only cleared by a further “Fix” and saving the file to yet another file name.

This turned out to be a rather tortuous process so I decided that I was only going to check and ensure PDF/A-1b compliance for the files that, at the start of this exercise, were not in PDF format at all. The remaining 800+ files which were already in PDF format at the start of this exercise will have to stay as they are for now. I’ve checked a few of them and they all have several compliance errors, but to ensure they all complied with PDF-A1-b would consume more time than I am prepared to spend right now.

The key findings from this phase of the work so far are that it is vital to fully understand the file formats you are targeting, and to become very familiar with the software you intend to use, before creating the Preservation Plan. Without that knowledge the Plan is likely to be unrealistic and almost impossible to stick to.