Pro cameras - where you shoot in RAW mode, and do post-processing later at home on your gorgeous widescreen IPS display, curate it, and then upload to your favourite hobbyist sharing site (Flickr, SmugMug, etc). That doesn't really lend itself to requiring a wireless modem.
I can see integration of GPSes into DSLRs - geocoding is a hotly requested feature particularly by hobbyists, but modems?
The absolute top-end Canon cameras have WiFi dongles that can wireless upload photos as they are shot to a storage unit. This is really to facilitate quick review in a studio setting - you aren't tethered by a wire, but can review the photo on a big screen very quickly after a picture without popping the memory card.
But while you're out and about? It seems pointless. The uses of a pro camera do not include "hey guys I'm at a pizza place. Look at the size of this sucker!", which is really where a 3G modem seems useful.
But I agree, this isn't a huge deal.
Dave was specifically calling out a line from their ad campaign, saying that real cameras don't have ringtones. He then argues about an entirely unrelated point, that cameras without data connections won't be terribly enticing.
Except that his original point is still there, and still is pretty dumb. Real cameras won't have ringtones, because placing a call to a camera is a bad feature, that nobody wants.
A camera that has a wifi radio, though, is a cool feature, which everyone wants, which is why Nikon made a wifi attachment for their pro DSLR line. It's called the WT-4/4a, and they released it in 2007.
(Of course, being Nikon, they charge $700 for it, but that's another matter.)
> Optics is, more or less, a solved problem. We hit hard physical limits long ago.
This is only true for cameras which are designed around a principle of light passing through a “generalized pinhole” (i.e. go through a lens and are collected as a two-dimensional image in a single frame at the back, without doing any fancy math; the kind of set-up required by, for instance, a process based around film or our current types of digital sensors).
Look to biology though for inspiration and it quickly becomes clear that visual information can be captured in all kinds of ways, with all kind of crazy “eye” designs.
There’s a huge amount of information stored in the light rays passing through a camera which is currently not captured, and the cameras of a few decades from now will blast through all of those “hard physical limits”. We’re going to get much better approximations of actual light spectra coming in, along with information about polarization and distance to the subject; simulate any lens, any aperture, any focal length in software because all of the relevant data will be captured. (In my dreams, anyway, or even maybe my somewhat-informed guesses; we’ll see how well technology actually advances.)
Anyway though, our typical current digital cameras don’t even hit the limits given the traditional/obvious camera design approach: for instance, we currently have three different colors of pixels each of which filters out a different set of wavelengths, meaning we are only capturing a “third” of the information we need to make a metameric match from each point, and just tossing out a bunch of the other photons. That right there is worth a something like 2x improvement in data captured per area.
One last thing: even before we get crazier camera designs, improvements in computer processing speed are going to make it possible to create high-resolution images out of an image captured over time (i.e. a movie processed into a still): we can see the beginnings of this already with those panoramic cameras I’ve been seeing TV commercials for.
Plenoptic cameras are cool, and have a bunch of neat features, but each camera element still uses traditional lenses, made of specially shaped transparent materials, whose fundamental design has not changed in four hundred years. There's been a lot of neat hacks on the basic idea (achromatic lenses, etc) but nothing new. Except superlenses, but you'll never see a macroscopic superlens, so it's useless for photography.
http://stellar.mit.edu/S/course/6/sp09/6.815/courseMaterial/...
http://stellar.mit.edu/S/course/6/sp09/6.815/courseMaterial/...
My main point is that it’s quite possible that we’ll have cameras in 50 years which are the size of today’s compacts, at the quality (or better) than today’s full-frame DSLRs. Indeed, I’d be surprised if we didn’t.
The ad was advertising point and shoot cameras. Most of which have tiny sensors. They might be good quality, but only marginally better than the best of the cameraphones right now.
Do you really think the quality of a point of shoot is so much better than a cameraphone that you would carry one around all the time?
Or would you rather carry a phone with a decent camera, and take out your DSLR when you need better photos?
You'll never see an interchangeable lens camera with a cell phone in it, simply because high quality lenses have big elements, which gather lots of light. As I am sure you will be surprised to hear, big pieces of glass result in big lenses.
Optics is, more or less, a solved problem. We hit hard physical limits long ago. You'll never see a compact full frame 35mm sensor digital camera. It just can't happen.
Can you imagine talking into a pro DSLR? Using it as a every day carry phone? The D700 masses 995 grams, and that's just the body! Add another 500 grams for the lens; is that something you're just going to slip into a pocket? How are you going to use it, by holding it to your head by the lens barrel, then terminating a call by pressing the shutter release?
It's generally a good idea in consumer electronics design to not make your users look like fools. Nokia forgot this when they made the N-Gage, to their detriment.
(And, of course, he misspells Lumix as Lunix. That's pretty funny, though, I'll give him that.)