Bits and Pieces
I think I have left a few loose ends in some of my previous blogs so this one is just to rehash and reemphasize some points I might not have covered or not covered well.
Archiving Documents
In my previous blog on this subject I may not have made it clear that I do not believe that PDF/A or PDF files are the total answer to archiving documents. There is a wide spectrum of needs when archiving that depends upon the anticipated uses for the documents in the distant future. What I did say and do believe, is that a vast number of archiving needs are of the ePaper variety, where I just need to save a birth certificate or a wedding license or a record of a business transaction. I do not see any need in those cases for saving an editable form. In fact, I want to save those documents in a form where I can lock down the appearance and be assured that in 50 years a person will be able to see exactly what I see today. PDF/A is great for that.
I wrote about CDF because in the first few weeks of November it started popping up in the news, primarily because the Open Document Foundation had announced frustration with the direction that ODF was taking and was switching their support to CDF. I found that rather puzzling since CDF and ODF are such different things. At least I think so. So I decided to make clear in my CDF blog what CDF is and what it isn't. I personally do not think it is a very good horse to bet upon for a ODF replacement unless it takes a rather different direction.
I am a little slow on this gossipy news but in the first few weeks of November a rather strange sequence of events was reported. First Sam Hiser (I guess with Gary Edwards and Marbux, the reported the leaders of the Open Document Foundation), announce their switch of alliance to CDF after having formed the organization to promote ODF. Then the most startling thing was that shortly after that announcement the Open Document Foundation closed down. There are some blogs about this here and here.
I am still puzzle over how shallow some people who write with great authority can be, and amazed at how they can collectively make a widespread and continuing story out of so little. They just fill in between the lines with great imagination. There were forecasts of the demise of ODF just because Hiser claimed their vital support for ODF was moving to CDF. I guess the switch in letters of the alphabet from "O" to "C" had press appeal. In fact, it seems that the ODF Foundation's vital support for ODF has diminished greatly over the years and at this point in rather immaterial.
Document Formats
Another thing that I have blogged about previously and still bugs the @#$% out of me is, again, the technical shallowness of some of the people that write about these things. In fact, I have found a lot of this same technical shallowness in some of the feedback we have gotten about PDF and standards. Come on! You have to be a little responsible and learn a little bit about the technology, don't ya!
I have been repeatedly asked if ODF will wipe out PDF. I guess it is a reasonable question, but I suppose I know too much for it to sound reasonable to me. Using some old fashion terminology, ODF is a word processor format used to save a document while you go to lunch. PDF is a communications format used to send information between people. Yes I know that my characterization of ODF is rather shallow itself, but really folks, that is where these formats like .doc, ODF and OOXML have come from. They may have tried to leave the old neighborhood, just like PDF has been trying to leave the old neighborhood of a captured version of what you would print, but those roots show through. Note in the diagram below that PDF and the other document formats live in different places.

I guess this all stems from peoples desire to read/write something interesting, something dicey. The controversy surrounding Microsoft and OOXML versus ODF is just such a thing. I might dismiss it as being a rather stupid argument except that it is at the root of a potentially huge financial loss for Microsoft. So it is big business.
Anyway, from my technical viewpoint (ODF and OOXML), PDF, CDF are all three different things. I do lump OOXML and ODF together although there are important technical differences between them as well. But they both do belong at the top of my diagram and neither is very well suited to be a communications document. Frankly, I haven't figured out where CDF fits in the diagram.
Document Format Profiles
I have written that document format subsets or profiles are a bad thing. Yet I also think that the PDF subset PDF/A is a good thing. How to reconcile these apparently conflicting positions. Well it took some thought, but I think I got it. Let's try to use my old colored telephone analogy. That was a what-if we had red phones and blue phones and red phones could only connect to red phones and blue to blue. That would be terrible. You can make it worse, even, by introducing black and blue phones and then make up a bunch of nonsymmetrical rules about which color of phone can talk to which other color of phone. Communications document format subsets are like colored phones. And PDF is one of those communications document formats.
But PDF/A is ok because it is not driven by the capabilities of devices and it doesn't restrict the key attribute of PDF, namely the reliable display of the document's content. What is does do is restrict the file format for functional reasons. The files should be such that the colors, the content and the ... cannot age or change from device to device or over time.
Let's see if we can fit it into my phone analogy. What if I said that for certain conversations you were not allowed to use foul language. That would not be the same as blue phones versus red phones. That is the kind of restriction that PDF/A imposes. It isn't a restriction that inhibits the communication of information, it is a restriction on the kind of information that can be exchanged. It is OK in my communications metaphor. I can still call and talk to someone with a different color phone, I just might have to watch my language.
I hope you got this because it is really important that we have support for PDF/A and understand its role in the bigger PDF picture.
Contact me at: jking@adobe.com
Comments
John-
Thanks for this cogent assessment of PDF -- a great & useful thing.
Sad to see even you couldn't parse all the disinformation about the OpenDocument Foundation.
We formed the Foundation in 2005 upon the ISO ratification of ODF v1.0. Purpose was to safeguard the public interests in interoperability to save the format from rapacious vendors.
Obviously, we needed more fire-power.
Please know the Foundation was an extension of a thread of interest and influence we exerted on the concept of a Universal Document Format dating back to 2001, when Gary and I began our involvement with OpenOffice.org.
The Foundation's mode & motives were distorted because we disagreed with Sun's & IBM's refusal to make ODF useful through interoperability with the MS legacy docs. We were never public facing and did no self-promotion outside the hermetic bubble of the standards effort (though we promoted ODF vigorously) so were helpless when the FLOSSY rabble with pitchforks were convinced by IBM that our name was the problem. This non-sequitur dominated by a force of viral cascading influence and succeeded in deflecting attention away from the hypocrisies encumbering ODF's development trajectory.
At this point, among other things, I am glad we at least have PDF; and we of the Foundation can focus our energies where they will be productive.
Posted by: Sam Hiser | February 21, 2008 07:14 AM