I have to say that for a begginer at least, I really feel that books are very useful. The big advantage that books have over the "internet" is a degree more reliability. To get a book published by a mainstream publisher of textbooks take a bit of work. The publisher is going to be concerned with their reputation, which usually means that as part of the publication process the work will have received a degree of peer review, I guess this will be more so for a new author compared to an established one, but even so there will have been a review process. The internet has no limits on publication, anyone can buy a domain name and some hosting, and publish anything they want. Actually you can even do it quite well totally for free if you want to. All of this means that I trust very little information on the web to be totally true and unbiased, anything that allows just anyone to update posts is essentially unreliable. Even places that many seem to quote as sources, such as Wikipedia are very poor. I have read many articles on areas that I have a degree of expertise in that are completely erroneous, although most articles on science and engineering seem to be reasonably OK as far as i can tell. The only places that I would rely on for information on the web are those associated with the traditional publishing industry, or other companies that have made a decent name for themselves as online providers, such as Lynda.com, who have a reputation to maintain, just as the traditional print providers did.
Now more on the subject of the OP, the one thing that always annoys me is the person who says that I do NOT need to post process my work, I always get it right in camera. This is not true now for digital images, and was never true for film work either. Actually with film you always have to process the film to get any sort of useable image. How that processing is carried out can make significant differences to the results that you will achieve. even with transparency film, changing the development conditions can have a significant effect on the result. As soon as you include producing a print you are adding a significant degree of post processing work, even for films that have been processed through a 1hr automated machine, as the machine systems will do a significant amount of adjustment to your image. Use a pro lab, that offered hand printing, or do the work in your own darkroom and the amount of processing applied to a film based image was significant.
With digital imaging every image is post processed, here though you have two basic options, decide on just what processing parameters you want to use in advance, set them into the camera and make your exposure. Change some of the parameters and there is a good chance that you would have to alter the exposure applied to suit the new processing. In Canon DSLRs the classic case in point would be changing the picture style. A photographer with real understanding will realise this, and if they want optimum control will shoot an image to give the optimum raw data file, that will suit the processing steps that they plan on using to get the image that they initially envisioned. I guess the classic example of this, which of course dates back well into the days of film is Ansel Adams and his Zone System. A system that requires understanding of not just how you will expose the film, but how you will develop the negative, as well as then go on to make the print. I guess one of the closest similarities that I can think of in the digital world is using ETTR. If you shoot ETTR and expose correctly for it, then the in camera processing will return a very poor image, most likely looking over exposed. Yet take that correctly exposed ETTR shot, and apply the correct processing to it, and you will end up with a much better image than can be pulled from a supposedly correctly exposed image. Do not try to do ETTR using Canon's software though, it is not designed for it really, and is unable to work well with the image data in the highlight areas. I do not like to think of it as highlight recovery, as you are actually working with image data that was recorded, just that the "standard" processing methods ignore it.
Of course the ability to make the most of the camera and other systems in producing an end result, an image, doesn't mean that the resulting image actually has any artistic merit. I have to admit that I am essentially self taught, and as such do not have the full understanding of the language of art that is used in such discussion amongst those with the requisite training. My training and education has been as an engineer, and that is the professional language that I speak. I unfortunately like a lot of the rest of us have to get by just knowing what I like and dislike, both in the visual and other arts. For a lot of people, that bar is extraordinarily low, as I well know.