OpenRAW

Sorry, Nicholas, but it seems you totally misunderstand the OpenRAW work group and their goals.

DNG is a great idea, but currently there is NO camera out, which writes this format natively (however, 2 cameras are announced). What this means is that there are millions and millions of files stored away in proprietary and worse undocumented file formats, which probably at one stage become unreadable.

What we want is just the open documention of these file formats, just like the ingredients of a tasty cake. Programmers don't need the receipt (the source code or company secrets HOW these files are interpreted), if they see the cake (the results, the final image) and the ingredients (the documentation of WHAT is WHERE in the file).

With this information somebody could code a converter which exports ALL data into the DNG format and if this is still around in 100 years from now, the image is rescued.

Don't mix up apples and bananas at this stage.

Nicholas,

If you are so convinced that camera manufacturers will not publish the file specifications to their RAW files because they are "central to their products," what makes you so certain that they will conform to an open specification (DNG) that they will see as the lowest common denominator among camera manufacturers? They argue that they need proprietary file formats to ensure their creative freedom in hardware/software development.

No one with OpenRAW is asking them to "hand over their software" or source code - what we're asking for are the file specifications (or "floor plans") to their RAW files. We are NOT asking for the algorithms they use to render images. Given the file specs, a third-party developer can write their own algorithms to render the file into an image - of better (or lesser) quality than the RAW processor provided by the camera manufacturer. This mechanism may be the only way to support RAW files in the future after the manufacturer drops support of a given camera (e.g., as Canon has done with the D30).

What we want is the right to "develop" our "film" using whatever processor we choose.

Juergen, "The goal of OpenRAW is to encourage image preservation and give creative choice of how images are processed to the creators of the images".

If Canon and Nikon and the rest said "we intend to continue to proliferate Raw formats, but we promise to publish the specifications for them all", would the goal of OpenRAW be met? I suggest not.

Publication of formats can't simply be treated alone. Camera manufacturers mustn't be allowed duck the real problems and fob people off with statements like the above. If I were Canon, I would be tempted to publish the specifications and simply continue to proliferate new standards. And loudly promote the fact that I was now implementing the "OpenRAW" policy. Could you deny it?

It must be considered as part of an overall strategy.

You say: "DNG is a great idea, but currently there is NO camera out, which writes this format natively (however, 2 cameras are announced). What this means is that there are millions and millions of files stored away in proprietary and worse undocumented file formats, which probably at one stage become unreadable".

I simply convert my Raw files to DNG at the first opportunity. (Once I have 2 copies of the DNGs in different places, I delete the original Raw file - I don't archive it).

DCRAW can already decode over 150 Raw formats. The DNG Converter can handle over 70 Raw formats. Both include the D30.

I'm not arguing against publishing those formats. But OpenRAW's advocacy needs to accept the realities of that 150+ and 70+. Until you place your site in the context of that existing work, it will confuse people who know a lot about DNG and wonder if you don't. See:

http://avondale.typepad.com/rawformat/2005/04/dng_support_sta.html

Why don't you add a page that positions the OpenRAW solution within the context of DCRAW & DNG, and those 150+ and 70+, so that people can see where OpenRAW fits in the overall picture? It would make it clear to people that you do in fact know a lot about those products and the DNG standard.

Barry,
In your response you stated;
If Canon and Nikon and the rest said "we intend to continue to proliferate Raw formats, but we promise to publish the specifications for them all", would the goal of OpenRAW be met? I suggest not.
Quite simply, if they did release the specifications of their formats our goal would be met for that manufacturer. Full DNG support would then follow quickly. No heroic decoding effort would be needed by Dave Coffin, Eric Hyman or anyone else.

You also stated;
I simply convert my Raw files to DNG at the first opportunity. (Once I have 2 copies of the DNGs in different places, I delete the original Raw file - I don't archive it).
I'm glad that you have a workflow that works for you. For some though, your workflow might be anything but a flow. The DNG format is not perfect and is very new. Adobe warns that you should save the DNG with the RAW file included to ensure that metadata is preserved.

You then suggested that we add a page about DNG and DCRAW and how it fits' with OpenRAW. Maybe we should, I have posted already some of my thought's on the matter but perhaps a full page statement would be appropriate.

Larry, you say: "Quite simply, if they did release the specifications of their formats our goal would be met for that manufacturer. Full DNG support would then follow quickly. No heroic decoding effort would be needed by Dave Coffin, Eric Hyman or anyone else".

This is an intriguing glimpse into your thinking. You obviously have a view about where your advocacy fits into a broader picture that includes (or doesn't include!) DNG, DCRAW, etc.

Good! I hoped you all did. My problem has been that I didn't know what it is. To me, because I spent decades in the business of designing complex systems, I want to understand that big picture, to know whether all the bits will happen, or we will be left with some holes.

Obviously I don't expect any one person or organisation to handle all of the picture themselves. It is a multi-vendor problem, and will need a multi-organisation solution. I think there are some people posting here who think that publication is the end of the story, but as you suggest, it will lead to other vital things. We handle our images via software, not specifications.

(If I come across as negative, this is because I don't repeat each time that I think "OpenRAW" is a great slogan, injunction, and rallying call, and the description of the problem is at least as good as any other around. I'm qubbling about a few sentences in the rather brief "solution" part of the statements).