Not a single camera company has really managed to get workflow of traditional cameras right for the Internet age, particularly with WiFi. The “best case” scenario is that you use a flaky and minimal iOS/Android app the camera companies have created to get images over to your smartphone or tablet, and then let them do the heavy lifting. I believe this is the wrong approach. Moreover, given the file sizes we’re talking about and the lack of raw support, smartphones don’t really solve problems, but create new ones.
Here’s the way it should work:
- Shoot your images
- When you get back to the hotel, office, or home, turn WiFi in the camera On
- The camera connects directly to your computer/network
- Software on the computer recognizes the presence of the camera connection and automatically downloads and renames all images in the camera to the location of your choice based upon your preferences/selections
That’s it. That’s how simple the “better” solution is.
Why is it a solution? Well, let’s take the tourist scenario for a moment. You left the hotel with your camera in the morning to go tour the city. You shot lots of pictures during the day and finally got back to the hotel after dinner. You probably don’t want to do much with those images other than get them saved off your camera because otherwise you’d be working with images all night in the hotel room instead of sleeping. But even if you do want to browse and select a couple of photos to send to friends, you usually want to use your usual digital software editing tools. So the images need to get to your computer, and perhaps even all the way to your home network or cloud storage.
Cellular transfer of images would be expensive (especially if you’re in a foreign country). You have lower-cost or no cost WiFi at the hotel. Your computer software handles raw files and batching of images better, and can be more easily programmed to do a sequence of useful things with them (especially Macs via Automator).
Thus, you really want the images to magically go from camera to your computer the minute you enter your room and fire up your computer. Then, if you’re using something like Lightroom, you want them to automatically get into Lightroom so you can use its tools to do the heavy lifting (hint: use a Watched Folder).
As you might guess from my scenario, the software needed from the camera companies is relatively simple: full peer-to-peer support for Mac OS-X and Windows via the camera’s WiFi (not iOS/Android peer-to-peer support!), and a Mac/Windows program that does Ingest/Rename ala Nikon Transfer that can recognize the connection and preferably complete the work automatically via settings you’ve already made. That’s it. That’s all that’s necessary to let users do what they want.
This scenario works for home and even some pro use, too. Indeed, it works for everything but the “gotta send it the minute I take it” scenario. The existing iOS/Android apps can work for that in a pinch, if they’d just get improved a bit and the camera made more of the controller (e.g. a menu selection “connect and send this image to Facebook/Flickr/whatever”).
I could go a lot further and specify things we do in the camera to make the whole Wifi-to-computer scenario work smoother, but we haven’t even got the first stone laid yet, so let’s start with getting that right rather than worry about all the other stones we eventually want to put in.
The irony is that while the camera makers haven’t been doing this simple thing, they’ve enabled a competitor to do it for them: EyeFi. Of course, EyeFi doesn’t get it all right, either, but at least I can get my images from the camera to the computer automatically without using a card reader.
In technical terms, for some reason we’re still in the Sneaker Net days with digital photography. You remember them: when you used to have to take the floppy out of your personal computer’s drive and walk it down the hall to your colleague who needed to work on the file. Well, that’s what we’re doing with images: taking the card (floppy) out of the camera (computer drive) and walking it over to the card reader on our computer (colleague).
I suspect some of this has to do with the Japanese language. Personal computers didn’t quite get into homes and businesses and used the same way in Japan as in the US and Europe. The complexities of typing thousands of Kanji characters and dozens of Hirakana/Katakana characters meant that it was just easier for business users to write on paper and fax/copy the paper to get information to someone else in the office. It’s not by chance that the biggest copier/fax product producers are headquartered in Japan. In essence, many Japanese businesses, including Nikon, are still using a form of sneaker net to move information about within their offices. Thus, taking a card out of a camera and putting it into a card reader seems like an elegant solution to them.
Unfortunately, the digital world just keeps on moving forward, and having these physical gaps that require physical interactions isn’t very modern. Certainly my generation grew into digital data that is moved/shared/stored willy nilly without any physical actions necessary. The upcoming generations won’t even understand that kind of thing (“what do you mean you had to drive to a store to buy something?”).
It’s time for the camera companies to join the 1980’s and enable true networking of image data. Oh, but don’t keep using DOS and its restrictions in what you do! The whole DCF and EXIF standards need a 21st century update and a strong move away from 8.3 filenames, for example. Though frankly: if a camera company got all the connection stuff right, they don’t need DCF any more ;~).
Update: after I posted this story, I got emails from people about existing products. For example, Fujifilm has a program called PC Autosave. That product has a few things about it that make it less than optimal for fixing workflow problems. First, your router needs to have a WPS button to use the simple connection procedure, and because it is router specific, you should probably be traveling with your own router. The manual installation on the camera side is a bit complex (and slow if I were to use my router’s very random long password). (I’m a little amused that Fujifilm suggests that you check the side of your router for the default password. I guess that is the typical case for a lot of people who buy routers and don’t know how to configure them, which is a whole different problem.)
But other problems include the fact that the software thinks you’re only going to put images into one destination folder (determined at installation). Since renaming files isn’t an option during the transfer, what’s going to happen when you get to DCF_9999.JPG? Good thing you can rename the files that the camera is going to create, I guess, but you’ll have to remember to do that each time you transfer files. Also, despite the name Auto Save, it isn’t automatic. You need to open something on the computer and select the PC Auto Save menu option from the camera, navigate to the destination computer you’ve created, then start the transfer manually.
The Sony cameras that support Playmemories Apps have a Direct Upload app that works a little differently. Basically you select images on the camera, then point the camera’s app at the proper network service (e.g. Facebook) when you’re logged into your WiFi network. Still not quite right. Again, it isn’t automatic, it doesn’t allow rename, it’s not applicable to raw files, and much more.
Technically, you can get a high-end Nikon camera to do what I ask with one heck of a lot of extra parts and configuration and some kludgy software bits. In studios, for example, you can configure a D4/D4s to wire into your Ethernet network and mostly do the right thing. With a bit of work you can have the images you shoot move to a Lightroom watched folder, for example. You can also use Nikon Camera Control similarly.
The problem with all the solutions I’m aware of so far are that they are cumbersome to set up, rarely automatic, often require specific gear, don’t handle a lot of things that need handling, and generally don’t improve the user experience or solve the user problem any better than using a card reader and some non-camera company software does.
Further Update: Two developers tell me that the real issue tends to be the low-cost parts used in the cameras and add-on WiFi units. They both tell me that they’ve tried to do something similar to my software idea, but the connection/download performance is atrocious due to the hardware. My 1300 NEF scenario might take days to download given the bottlenecks at the camera end.
The problem is compounded by a series of performance bottlenecks in most camera designs: lower cost electronics that read/write directly with the card, lower cost electronics that put data on the USB connector, lower cost WiFi chips that are also performance impaired.
This just all makes for a further comment in the vein I’ve been on for years now: where are the camera companies solving the user problem? Adding cheap WiFi that no one will ever use in a practical sense means that they tick off a marketing box (“WiFi compatible!”), which solves their marketing problem (especially vis-a-vis smartphones), but don’t actually deliver anything useful. That defines the ultimate type of design failure: great for marketing, does nothing for users.
What I’m talking about is quite do-able today. It requires correctly defining the problem to be solved, and then applying the right parts and code to make that happen. It’s not rocket science, but it does require a Captain at the helm who is capable of getting the entire crew on board with the same goals.
Finally a comment about whether the software idea is worthwhile or not. I’ve gotten several emails from folk that say sneaker net is fine. Saves battery life, for example. Well, battery life with the current systems is certainly a problem. But given that my smartphone can talk all day to my WiFi system and not run down its batteries while still providing very high speed transfers, obviously the problem is solvable. Not just not by current camera company engineering teams, it appears.
While I continue to use the sneaker net method of workflow, as do most pros, it’s because we haven’t been given a viable solution, not because it’s the best choice. I’ll stand behind that position.