Their positions are based on ignorance, arrogance and a misguided sense of ownership. Very typical Japanese corporate thinking...(and I say this from personal experience dealing with these companies not out of some racial bias–anybody who has had direct dealing with these companies correct me if I'm wrong).
+1. I have never worked with Canon or Nikon, but I have spent time with other Japanese consumer electronics companies, helping them design new products, and participating in standards organizations with them, and completely agree with the summary. Well, maybe not the "ignorant" part. They're probably completely aware of what's going on. I'm not even sure about arrogance (certainly the Japanese have no monopoly on arrogance). But definitely "misguided sense of ownership".
I would like for there to be a single standard. And I have nothing against DNG, would be happy if it turned into the one single standard (I convert my CR2s to DNG on import). However, it's really not that simple.
For starters, it's naive to think that Canon, Nikon, et.al. are "reengineering" their raw formats for each new camera. They are not. They add a wee bit of metadata here and there. Maybe just change a version number. But imply/infer that they are redesigning things from the ground up is naive. As evidence of this, I'll point to two things: 1) it really doesn't take Adobe very long to release the next version of ACR that reads the raw files from the latest cameras. and 2) go read dcraw.c, a widely available open source raw file parser.
More importantly, and a good concrete engineering reason why Canon, Nikon, et.al. can convince themselves that their own formats are preferable, is ease/speed of innovation, coupled with keeping secrets for longer. As an example, look at Canon's GP-E2 GPS unit. It embeds directional (compass) information into the pictures in addition to GPS coordinates. This is new. Lightroom doesn't know how to display this. More importantly, the EXIF/IPTC/whatever metadata for this hasn't been standardized. Let's pretend you're an engineer at Canon, with bright idea of delivering compass information. Do you want to play with a standards organization, that may have their own schedule, for a long drawn out decision about how to record compass info? Heck no. Worse yet, do you want even tell a public standards body that you want to store compass info? Hell no, that gives advance warning to competitors of your product plans.
[Ok, I just looked it up, and EXIF does have a way to define compass direction. My point is the same, though: if a company wants to do something truly new they certainly don't want to wait on a standards body to tell them how, much less tell their competitors (on the standards body) about the new things being planned]
Of course, the way this is dealt with in the industry is to allow people to have their own custom metadata. If that happened in DNG, then Canon could provide their own private stuff in DNG. And Nikon their own private stuff. And so on. And at that point, your common standard has solved nothing: it's just a well documented wrapper for embedding undocumented stuff.
That, in addition to Japanese corporate decision making, is why any hopes that a standard will be adopted by the big guys is just wishful thinking.
