The next generation of cameras are going to be competing in a market that’s tougher than ever. While the launch of mirrorless camera hardware from Nikon and Canon may have prompted some to upgrade, the next round of camera purchases also needs to offer something new in software. Want to know what needs to be there?
The Menu Experience
I want to focus on software because I feel like it’s the most universal element of the user experience from beginner to pro and brand to brand. I don’t think anyone is in love with their camera’s user experience, at least on the software side. Messy menus, arbitrary changes between bodies and generations, and weird limitations on customization are all symptomatic of a time before smartphones.
While phone manufacturers have realized that you need to let your user tweak the layout of apps, for instance, very little of the menu is actually customizable on cameras. My Z 7 has a custom menu, but there are menu items that can’t be relocated there, and a single “super menu” of favorites is literally the least effective from an organization standpoint.
This is not a desirable user experience in 2020.
To address this, open up the ability to truly build a custom menu. I’ve shot Nikon for over 10 years, and I’ve never once needed the retouch menu. Let me hide that entirely. The same goes for customizing buttons. Other than the power switch, why isn’t every button on the body re-mappable in software? Also, the menus themselves can become much denser in terms of information displayed. Between cheap storage and high-resolution displays, pulling up the “help” option on a menu item could display the entire manual entry, not a cryptic two-line explanation.
Things have gotten better in some cases. On newer cameras, a good number of the buttons actually are re-mappable, while the new implementation of the i-Menu in combination with a touchscreen has alleviated some of the need to have an entirely customizable menu. I’d still say that these are just steps towards a better experience, however, with a lot more distance to travel.
Support for Mobile
Everyone has a smartphone, but few cameras actually interface effectively with it. A smartphone offers storage, a high-resolution display, a blazingly fast processor, and a high-speed cellular modem, yet no camera company is making use of this hardware. To go back to Nikon, SnapBridge exists, but it’s not very useful. Trying to do even something simple, like shooting tethered, doesn’t work (if you don’t have a memory card in the camera). The same case follows for other manufacturers: Sony’s PlayMemories app sports an impressive 1.2 star rating on iOS.
Even moving the images between devices is slow, because manufacturers have made silly choices for connection standards. There’s no reason why a $3,000 camera shouldn’t have AC WiFi or at least be able to connect over USB-C to the mobile device. The same design mentality that gave cameras micro USB and mini HDMI have also given them inadequate wireless connectivity options. Moving forward, design with standards in mind. Cameras should last years, so users certainly don’t want to be stuck with the HD-DVD of connections.
Open Up Support
The camera industry is contracting, and now is not the time to be building a walled garden around your users. While Apple might be able to leverage their market position into locked-down software and soldered-in hardware, no camera company should try to emulate that. Instead, camera companies should be looking to open up. 
I had high hopes for the L-Mount alliance between Sigma, Leica, and Panasonic, even if not much has since come of it. At least they’re trying. Canon and Nikon, however, have kept the RF and Z mounts locked down. Not only has this hurt the availability of lenses, by limiting those options to what can be made by them and reverse engineered by the likes of Sigma and Samyang, it also hurts the sales of bodies. Nobody wants to buy a body that no lenses are available for.
Two things make the lockdown of the mount particularly frustrating: they don’t do it throughout their line, but they do extend that mentality to software. In Canon’s line, you can get other mount options once you’ve stepped all the way up to the PL mount cinema cameras. I get that cinema is a different set of needs, particularly when it comes to focus and lens automation, but you don’t have to offer an F mount R5. Instead, let third parties license the RF mount and produce lenses for it.
This mockup won't win any design awards, but instead consider the possibilities of combining the great software experiences of mobile devices with the image quality of bigger sensors and glass.
The same open thinking needs to be expanded to the software and firmware side as well. Release an API and allow tight integration with all the mobile apps that photographers already use and love. Imagine having access to PhotoPill’s AR overlaid on your live view screen, letting you set up the perfect shot for any choice of focal length.
Support for more complex procedures, like focus stacking, has taken years to make its way into the cameras and even now is mediocre in its implementation. I get that a camera maker isn’t a software company, and that’s okay. Even Apple, Google, and Facebook can recognize when someone has built a better piece of software than them. By keeping the platform so closed off, however, developers never even get the chance. Just look at what the team at Magic Lantern was able to build, despite all the difficulties of reverse-engineering the software.
Why Now?
A lot of these behaviors have been around since the beginning of the digital camera age, but they need to change. The whole landscape of technology has shifted under the feet of camera makers, and they haven’t dealt with that well. It shouldn’t be an excuse to quit, as some have argued, but a reason to look at the strengths and weaknesses of your current product and find a path forward. Protectionism and isolationism don’t generate sales, and in this case, don’t even seem to be doing a good job of preserving sales. The big manufacturers shouldn’t be looking at software as a pair of shackles to keep a C300 buyer from purchasing a R5. Instead, let the RP appeal to the dozen of iPhone photographers looking to step up to real lenses and higher image quality, but that’ll never happen if your UX is 20 steps backward.
They all have microphones so why not voice controlled menus?
That'd be interesting. I know cameras like the Dx series let you record voice memos - but a Siri style assistant would be cool.
They need to address the actual aspects that become a brickwall to shooting.
-Focusing in poor light
-Dynamic range
-Flash sync speed
-Noise
The whole EVF, eye focus thing is currently a distraction to really improving the basics. And a slight backwards step in some areas.
I see stacked sensors that allow shooting the same shot at different focus points, and at different exposures, which creates one stacked RAW file as the next big step.
Sensor cooling and shutter design are the keys to unlocking the next wave of photography improvement.
I'm also going to hang this out there...
For improvement in both Stills and Video, we need to stop believing they are both on the same pathway. They NEED divergence in technology for both to take the next step forwards.
I think for some of those you're going to start to run into physical limitations of current technologies and form factors. Canon's focusing down to -6EV, while you already have to massage existing DR to fit well into display devices, for instance. I think over the next generation of cameras you're only going to see marginal improvements in a lot of these areas, as other links in the real life to display chain need to catch up.
These software improvements could be made "tomorrow" and even back ported to a number of existing devices a la Kaizen.
I especially agree with the dynamic range. It should be a thing of the past. There must be some way to merge two consecutive raw files shot milliseconds apart in camera. It’s embarrassing when phone shots in high dynamic range situations look better than my $4000 gear straight out of camera.
As for most specific task device, the coding langage is probably not directly compatible with smartphones as is. Camera have a very very limited computing power compared to smart phones. Considering having less than 500 shots is bad press, imagine having a camera than can post on IG but only do 150 shots. Even more bad press!
I wouldn't use "must have", but now that we're pretty firmly entrenched in the mirrorless era, there are many things that could stand to be improved, and because mirrorless basically means 'the computerization of cameras', most of the improvements can be made with software.
I agree with your menus and mobile comments. More customizability would be nice too—for example, why can't I assign a name to a custom mode? It would be a handy way to keep track of what you use that mode for. On the EOS R and R5, there is no fixed mode dial—why can't I setup more than 3 custom modes?
Face/eye tracking has to be hands down one of the best things that define what a mirrorless camera can do that a DSLR can't (at least, using the view finder), and should be on everyone's "must have" list when upgrading. Other 'smart' things I would love to see in "real" cameras is detection for when you have camera shake, an un-level horizon, or when it thinks you may have missed focus.
I would like to see more extended functionality of USB-C ports for things like image backup, tethering over mobile devices, network connectivity, audio I/O, etc...
I could keep going, but you get the point.
Connectivity could sure be improved - for tethering, wireless remote; transferring files... I'm a Sony shooter and I find most of the software for ancillary features (like wireless remote) to be hopelessly clumsy - as bad as their initial Alpha 7-series menus. (But at least the menus have gotten a little better, and maybe much better with the 7Siii.)
I agree with the comment that a lot more could be done through the USB-C port
I was thinking about this the other evening when I took two photos of the same dark scene - one with my camera (Pentax K3 II) and the other with my smartphone (Oppo Find X2 Pro). The difference between the two images was very striking, with the Pentax needing a high ISO setting to produce a typical night-time picture, but the phone using its AI capabilities to 'manufacture' a far more striking composition.
For me, at least until the next big advance in sensor technology, I think the most important thing cameras need to embrace is artificial intelligence, so that our cameras aren't getting ever more out-smarted by smart phones.
I'd suggest it should be implemented as separate modes so we can ignore it if it's not the way we want to work, but make considered choices if we think it is appropriate for some of our shooting.
Real photographers don't need this sh..t! It's just motivated by these Insta influencers who always need something new to stay...
"...Nobody wants to buy a body that no lenses are available for." I can see why Nikon/Canon have not opened their mount specs. There are actually some lenses for their cameras, and there will be more. So camera vendors understandably want to sell _their_ lenses, instead of letting Sigma/Tamron/etc. do this.