In reply to rexafarian above, there are many reasons to have a provable original.
In my work, I many times document incidents and accidents. This documentation can potentially be used in litigation. It must be original, undoctored images that I have.
In my avocation as an underwater naturalist, I again document things others rarely see. If I documented those with digital images, and had original observations of unusual behavior of the animals I'm documenting, then my observations cannot be sustained by my photographs.
For instance, I've documented in the 1970's sculpins using anemone tentacles for shelter. This is not unusual for some tropical fish, but this was in Newport, Oregon. I documented other symbiotic relationships with this anemone (then known as Tealia coreacia, but the name has been changed about 10 years ago, and I don't recall it right now--you may know it as the rose anemone).
An amphipod then unknown to science was found to colonize this anemone. It is a very interesting creature, with a distinct, "white," "V" on its back and gold strips going down its back on a purple background. I put the "white" it quotes because, under a dissecting scope, it wasn't white at all, it was transparent. What we were seeing was the white insides of the amphipod, and a close inspection revealed that we could see its beating heart, through the amphipod's back carapace! I've always thought this amphipod would be a great subject for some graduate student to study the effects of pollutents in the water, because we could actually see and count the beating of its heart without intrumentation. But I'm about the only one who knows (or used to know) where they live.
Also, a copepod was found to use the tentacles for shelter. This was not known in cold salt water environments, only the tropics and with different species. The copepod photo was published by Dr. Wim Vader of Norway in one of his papers.
These are two examples of where unmanipulatable images can be used to further science.
But if I came up with some startling observation, and brought only electronic images, the immediate question in everyone's minds would be, "How proficient is this guy in PhotoShop?".
So for the immediate future, until I've reconciled myself with this problem, I'll simply stick with film for my images. The original slide is there, and there are ways to determine whether it has or has not been manipulated.
I will use these in books/articles/papers in the future, and with scanned images too. But I will have an original to back them up. Now I need to get a slide scanner to do it, and luckily their prices are coming down.
SeaRat