Preflight errors during PDF/A-1a validation

Has no one a solution for this? I experience exactly the same problem with Acrobat 9.3.3 and PDF/A-1a. I have first structured my document in InDesign and I was disappointed when the structure was not exported as PDF-Tags. So I added tags to all elements in Acrobat. But when I try to convert to PDF/A-1a with preflight I get this same message.

I have not found one single valid PDF/A-1a document on the net... I don't understand why it's so hard to meet this standard. Any help is greatly appreciated.

You will find it effectively impossible to convert pre-existing documents to PDF/A1-a in Acrobat 9 if the original is not already compliant in terms of XMP metadata and structure. The Save As.. and Preflight tasks run a detailed check and can fix certain elements (such as changing the compression mode or adding an output intent) but they don't fix all the errors in the structure or the metadata, as many of the problems are semantic (requiring a human to make the decision).

If you export from MS Word etc using PDFMaker, you can make a PDF/A1-a file very easily as the original document can be analyzed and restructured before it's assembled into the PDF. Saving to PDF/A1-a from Word for example is no harder than saving to any other PDF version.

PDFs exported from other applications (Creative Suite included) do not use PDFMaker, and invariably they add or omit some of the structure or metadata. In the case of InDesign, it's an extra section of XMP that is always written into the file no matter what export options you choose, that Acrobat cannot remove and isn't permitted under the PDF/A standard.

You can of course print the file as PDF/A1-b (with no tag structure) by using Acrobat's virtual printer, but even the PDF/A Consortium say that creating an A1-a file is only practicable within the original application - and Creative Suite doesn't support PDF/A. The standard is not particularly hard to meet if you start with the original document, but it's nigh on impossible if you start halfway through.

There are countless A1-a files in existence (millions are created every year) but they are pretty much all the result of PDFMaker output from MS Office, or OpenOffice/NeoOffice direct export. If you want some examples to play with, try these:

http://www.pdfa.org/doku.php?id=pdfa:en:articles

Thank you for your detailed answer! I have noticed the metadata problem too. It seems like there is no possibility to save documents in the creative suite using an old xmp standard which is required by PDF/A. However I found a workaround for this: Removing all metadata with Acrobat's PDF-Optimizer and putting it back in Acrobat.

I don't think that metadata problem is linked to the structure error we are discussing here ("Structured PDF: Type entry ..."). I think in this case there is just one single untagged element in the structure and it would be cool to find it. The tagging itself is easy in Acrobat. I did a search for untagged elements in Acrobat but I get no results after having finished tagging.

Thank you for the link. Unfortunately using MS Word is no alternative to me as an InDesign user. As long as it is impossible to generate PDF/A-1a documents from professional applications this is a dead standard.

Just a sidenote: I checked the first three documents on that site and none of them validates in Acrobat 9.3.3 ...

This problem is resolved in Acrobat X.