Your article is pretty good, but there are a few points that stood out to me:
You said: >>>>>>>>>>>>>>>> Well, this varies according to what type of photographer you are. If you are only interested in taking snapshots, and will never try to do anything with the images later, except maybe reprint them on your color printer, or take them to a processor for prints, then the JPEG mode is fine. <<<<<<<<<<<<<<<<
I don't know of a single sports shooter or PJ that shoots in anything other than JPG. JPG has limitations, but for plenty of professionals, the benefits (file size, speed of file writing, and lack of need for post-processing) outweigh the liabilities. I wouldn't make JPG sound like a snapshooter-only format.
You said: >>>>>>>>>>>>>>>> JPEG converts the image from 12 to 8-bits and then compresses the image, permanently throwing away large quantities of image data in the process. <<<<<<<<<<<<<<<<
TIFF goes from 12 to 8 bits too. I also wouldn't say it throws away "large quantities of image data". There is the issue of "significant bits". This is discsussed better than I can do it here: http://www.scantips.com/basics14.html
I know this discusses scanners, but the underlying principal is the same. The camera has a pretty good idea which bits is should keep and toss. Not perfect, mind you, but not bad.
You said: >>>>>>>>>>>>>>>>>>>>> You can later uncompress the JPEG image by letting the computer add data back, but it does this by "interpolating" the data, which means it makes an educated guess about where and what data to add back. <<<<<<<<<<<<<<<<<<<<<<<
Please explain your basis for saying the computer adds data or interpolates a JPG upon opening. I don't think it does.
You said: >>>>>>>>>>>>>>>>>>>>> When a JPEG image is made generally only about 25% of the original image data is maintained. <<<<<<<<<<<<<<<<<<<<<
Again, please explain the basis of this statement.
I don't think your discussion makes an adequate distinction between bit depth and data compression. IMO, you should treat the two more clearly. The above statement is a good example. Besides the fact that I don't agree with the 25% number, you don't say whether the loss is in shadow/highlight detail (bit depth) or in resolution/detail (JPG-induced data loss) or some combination of both.
You said: >>>>>>>>>>>>> JPEG mode can be used and modified to a degree, but your image will degrade VERY quickly upon manipulation. <<<<<<<<<<<<
I disagree with this. Done properly, there is little degradation.
You said: >>>>>>>>>>>>>> Surely, within the next year or two, a standard will emerge, and this problem will go away. <<<<<<<<<<<<<<
I'd be shocked if that happens, but that's strictly my opinion and I am only adding this as an editorial comment. You accuracy is not in question.
You should also mention that Nikon Raw image formats require the use of special software to open them. Until recently, if you wanted to open a NEF, you had to pay a ridiculous amount of money to get Nikon Capture. Now Nikon View, available free, can open and manipulate NEF files, but it's a little limited in capabilites and won't do batch processing, which Capture does for (I think) $150. Bibble and Quimage are less expensive options from third-party developers. But as of right now, QImage (my favorite) does not decode NEF files from the D100, only the D1/X/H.
Last, it is important to remember that as of right now, I'm not aware of any monitor or printer that works in more than 8 bits. You can open a NEF, but you'll only see it in 8 bits. You can convert it to a high-bit TIFF, but again you'll only SEE 8 bits of it. You DO have the luxury of choosing the 8 bits you want upon converting the NEF to anohter format. The same applies for printers. You CAN choose the most important bits (which mainly means shadow detail), but we are still in an 8-bit world. That, of course may well change.