Probably the first question is whether you need to become an expert in computer image manipulation to use a digital camera. The answer for most people is "no". You can use the bundled software to download pictures, maybe make a few simple adjustments like cropping and red-eye elimination, and then either print or use some simple web-publishing software, or one of the online photo-sharing services.
But to really get the most out of your images, you should get to know the image manipulation software. In my experience, colors are not perfectly balanced right out of the box, and contrast is often too low to make a really snappy image, at least on a computer monitor. I find myself wanting to make more sophisticated adjustments as well, such as correcting perspective or lens distortions. This gets one into the whole "digital darkroom" arena.
You can use online services like Shutterfly or Ofoto to get prints, and they'll probably do a fine job with "straight" files from the camera, since they do their own adjustments. For web display, if you're a perfectionist like me, you'll probably want to tweak them for color balance, contrast, cropping, etc. Cameras will often come with software, like Adobe PhotoDeluxe, which is fine, but does take a bit of learning to use to full advantage. Higher-end cameras will more likely come with Adobe Photoshop Elements, which is more versatile.
The full (and expensive) Photoshop application is probably unnecessary for most people reading this. But it does have some capabilities that are interesting, such as 16-bit and RAW-format processing. Thankfully, there are other, much less expensive programs that have some of these features, though perhaps in a less convenient or integrated fashion. I've used the very versatile and powerful Picture Window, from Digital Light and Color, which does 16-bit processing and offers a range of interesting image transformations, including perspective and lens distortion corrections.
For now, I've standardized on Adobe Lightroom, which is oriented to higher-volume workflow. It's both an "album" program (helping one display and choose from many pictures) and an image-processing program. I find that it helps me process hundreds of pictures from a trip or event in a less unreasonable amount of time.
I like panoramas a lot. This means "stitching" individual images into a wider (or taller) whole. Most consumer digital cameras have panorama modes to help you take the individual pictures correctly. There may be a bundled panorama software package that comes with the camera, and/or you may want to buy your own. I use Panavue's Image Assembler. Photoshop Elements does have a built-in panorama feature, but Image Assembler is more powerful at combining the not-perfectly-matched images one typically gets doing it hand-held. Ideally one would use a tripod and a panorama head, but I can never be bothered.
A more basic need is simply to browse image files from the camera, to select ones to print or work on further, or to show them in a casual "slide show". All or most of the image manipulation programs can do this to some degree, but I like the freeware IrfanView which is fast, simple, and supports a vast number of file formats.
The last stage is displaying one's pictures on the web. Many of the general-purpose picture-display programs seem to come up with some sort of album-generation mode, which produces the HTML for a web-site. I used to create my own HTML, but this got out of hand. I've never liked the the free sites such as Flickr and Picasa, since the results don't look that good to me. I now use SmugMug, which is a silly name, but produces high-quality results. Unlike many of the free sites, it does cost money, but one can upload an unlimited amount of pictures, and in full resolution for people with large monitors and good download bandwidth.
A separate discussion is RAW format processing, which pertains to feature of more advanced cameras, and that is tackled next.
(By the way, I don't put any value on in-camera processing, such as red-eye reduction, odd color modes, etc. That all seems very gimmicky. You are much better off processing after the fact, using the original image as shot. That one way you can change your mind, revert adjustments, etc.)
Digital cameras process the bits coming off the digital sensor in-camera, typically producing a standard JPEG-format file. More advanced cameras also offer the choice of producing a so-called RAW file, which has not been processed as much. The advantage of this is that one can select different processing parameters after-the-fact, and that the RAW file is the true "digital negative", with all its versatility intact, and nothing thrown away by the in-camera processing. For instance, one can often get an extra stop or so of exposure latitude from the RAW file.
A complication, though, is that the RAW format is not standard, but rather each camera vendor has its own, and so specialized software is needed to process RAW files and turn them into standard JPEGs for viewing. The vendor's own software seems typically to be somewhat lame, and so a third-party industry has sprung up. I used to use the Phase One software, which is highly regarded, but it got expensive, and I now rely on the Adobe RAW Plugin which is a component of Photoship and Lightroom.
I've tried RAW files from my older Canon G2, which doesn't seem to be that much of a win. JPEG is an 8-bit-per-color-channel format, and the G2's sensor theoretically gives 10 bits per channel. So using the RAW format, one can capture that in a 16-bit TIFF output file (almost 24 MB!). I was hoping to see the gain in exposure latitude, but didn't really, in practice, at least using Canon's own software. But at least one can do exposure leveling and color balancing in 16-bit mode, so that the final JPEG can still preserve a full 8 bits of per-channel resolution, which makes for smoother tonality. (Note that the typical "consumer" applications will only do 8-bit processing, so for this one does need full Photoshop, or one of the other more sophisticated image-manipulation programs out there, such as Picture Window mentioned previously.)
RAW files from digital SLRs have more to offer. Their sensors have 12- and even 14-bit resolution that yield a lot more exposure latitude. Whether using PhaseOne or Adobe RAW processing, exposure correction seems genuinely to pull a couple of stops more latitude from images.
So RAW seems to be the way to go at least for more sophisticated cameras and the most critical work. It does make everything more complicated though, including simply sharing images files straight from the camera. At least IrfanView (mentioned previously) does have the ability to access and display the embedded JPEG that's part of the RAW file. Another issue I haven't quite wrapped my mind around is archival storage: TIFF and JPEG are likely to be readable for a long time, but what about proprietary RAW formats? Adobe has a new Digital Negative (DNG) format that might take hold.
Copyright Richard Schooler, 2001-2008 schooler@alum.mit.edu