I think that Jonathan made a good summary of the present situation. I'd just mention that we always have the option of sending off our picture to a lab which can use photochemical reproduction like Durst Lambda.
A current good-quality inkjet print lasts longer, has more accurate color, and has (with enough image pixels per area unit) greater sharpness and detail than an optical film print. This is especially true of large prints; one can print at the same 360 PPI quality regardless whether the print is 4x6 inches or 4x6 meters, as long as the large-print source file has the necessary resolution and one has the computer hardware necessary to render and spool such a large print. Inkjets can already print more detail per area unit than most people can see without magnification, but increasing the DPI capabilites of the printer means more and smaller dots per pixel, which means smoother tonal gradations and less-visible dithering patterns.
Improvements are still desirable in the areas of durability (fade resistance as well as how well the print tolerates handling, scratches, fingerprints, spills, etc), dynamic range (the difference between the whitest white and blackest black), color gamut (the range of colors the printer can reproduce), print speed, and the cost of the printer, paper, and other consumables.
The megapixel race is driven more by specmanship than by overall image quality considerations, especially among point-and-shoot cameras. In some cases, the increases in noise due to cramming smaller pixels into the same sensor area offsets the resolution advantage of the extra pixels, and the new model is no better than the old. Buying a camera on the basis of megapixels alone is kind of like trying to evaluate the qualities of a woman solely on the basis of her bust size; there are many other equally important factors to consider. But it's an easy way for marketers to "compare" their camera to someone else's.