Well, speaking as someone hopefully of their standard, I think I should have a few things to add. I for one can easily testify that more is not always better. Choosing only 3-4 prints to send off is because (and don't laugh, these are deadly serious and important reasons):
[1] The editor only wants 3-4 pictures.
[2] These photographers know how to edit their best 3-4 pictures.
As I've been trying to demonstrate via the portfolio forum, editing is not something many photographers get to grips with in a hurry. When you are working on a deadline and your editor only wants 3-4 pictures, you only send 3-4 pictures, not 34. That is, if you want to keep your job.
Put it another way. If you know the magazine is only going to use 2 pictures, there is absolutely no point sending in 10 or even 6. If you feel that you might not pick the right ones to send, then that's because you don't know how to edit your stuff. Professionals can edit out the ruff and send in the best 3-4. If those don't fit the bill, then the other 34 aren't going to and there's no point sending them. If there were something completely unique that the editor is after, then he or she would have told the snapper before hand, if not, would put in a request after the fact, and the photographer would go and look through his mountains of frames.
I shoot as much as I can, but I don't make sure I send in one picture of Beckham, one picture of Keane, one picture of Giggs, one picture of whoever, just on the off chance a news editor might be focusing on that player. I shoot, send in my normal lot, and if they do need a picture of Giggs for example, they'll ring my desk who will ring me, I will sift through my lot, find a picture of Giggs, and send it along. Sure it take additional time, but you don't want your editor to do the editing for you (ironic I know, but...)
We shoot lots for all kinds of reasons, stock being a major reason. Which means while a small number of our shots might be shot for immediate news use, a good number is actually shot, catalogued at a later date, and shunted to a stock library. And eventually makes the agency more ££ than live pictures. So they shoot more for a reason.
Also, sport is probably the one area (not exclusively, there are others) where you cannot shoot one frame and get the shot. Yes you've seen these guys work. Have you seen film snappers shoot sport? They shoot every bit as many as digital shooters. You pile up a small mountain of film canisters by half time. In fact, I probably shoot less with digital than with film, simply because my film camera happens to churn through frames a lot faster.
When you are working professionally, cost of film is really a moot point in general, which really does take professionals out of the equation in addressing the original topic.
Regarding shutter life, as we know both DSLR and SLR has same shutter life.
Not necessarily. It depends on how the shutter of the specific DSLR is designed, but because of greater flexibility in design of DSLR shutters, they have the potential to be constructed to take less strain than conventional SLRs and thereby last longer on average.
The same reason has the nifty side benefit that a D1 series camera can be made to sync with flash at around 1/2000 or even faster with conventional flash units. The only reason 1/500 is stated as the maximum is because otherwise the flash duration can become longer than the shutter speed...
SLR+Drum Scanner - is a perfect combination for people who thinks that DSLR depreciation is unjustifiable. Extra $1000 saving can supply film and developing/printing cost for more than one year for average shooter.
As you've already mentioned, yes a drum scanner costs a lot more than DSLRs.
