When you're editing a photo, every pixel counts.
For the past two decades, Adobe has been the leader in maximising every pixel or analysing the colour, density and sharpness of an image, helping pro photographers and amateur snappers alike improve the quality of their work. The question is: where can its software go next?
To infinity, it seems. Future features are set to include facial recognition for meta-tagging images and online editing tools. We sat down with Jim Mohan, a Director of Product Management at Adobe, to talk more about these planned advancements and ask about the current and future state of photo editing.
TechRadar: We'll start with an easy one: how does Photoshop work?
Jim Mohan: I can give kind of a top-line view of what's going on behind the scenes with Photoshop. Really it's all about maths and pixel- level maths. Each pixel has a red, green or blue value, and any time you make adjustments to them you're altering those values either on a pixel-by-pixel basis or based on what surrounding pixels are doing. The mathematics can obviously get very complex and require a lot of calculations at one time. Since consumer PC processing power has increased significantly in the last few years, a lot more can happen on the consumer's computer. That said, sometimes to make Photoshop appear to run faster we'll do previews on a lower-resolution photo before finalising the change. So you might see some really fast responsiveness if you want to lighten or darken or change the levels and then when you press commit or OK, the processing on the full resolution photo will happen. As consumers are getting better cameras and the size of photos is increasing significantly, we do what we can to make sure you're not waiting for a very long time to see and evaluate whether you want to make that change.
TR: How have you reacted to other changes in technology over Photoshop's lifespan? Has your core customer changed?
JM: We are definitely seeing our consumers moving up in terms of camera [quality]. DSLR cameras are coming down significantly in price. What we do is try to take advantage of the more advanced features of the camera. One of the biggest trends in the recent past has been an increased adoption of raw formats – consumers wanting to do more of the processing outside of the camera so that they can have control over it. So we've got Adobe Camera Raw at its full capacity in Photoshop CS3 and what we call an 'elementalised' version of Adobe Camera Raw in Photoshop Elements. We're seeing a huge increase in the number of people using DSLRs and experimenting with Raw. We're also starting to talk to the camera companies about how we can work together to think about the workflow more cohesively and think about building things into the camera that our software can take more advantage of. One of the most extreme examples is a prototype lens that Todor Georgiev at Adobe has been working on. This is a lens that's certainly nowhere near going to market yet. It captures over 20 images at one time and you can then adjust the focus after the fact. So I can take a picture of a crowded room and then afterwards have it focused on the far end or near end of the room. And that's an example that we see coming in the future. Adobe Camera Raw allows you to adjust lighting and exposure; this concept is about adjusting the actual focus after the fact.
TR: Do you, as a software development company, feel that certain functions should not be done on the camera? Is that the kind of stuff that should be saved for later Photoshop retouching?
JM: Not necessarily, no. Thus far our focus has been on listening to requests from consumers about making the software easy to use. So we've done things like improving selections by creating 'smart surround', which checks what each pixel looks like in comparison to the surrounding pixels as a way to approximate an object. There were tools in previous versions that were very much based just on colour, but now we're able to take into account the colour, lighting and contrast of the surrounding pixels to do a way better job at figuring out what the object is. So our focus, thus far, has been on making processing easy in our software. If that extra processing is a decision that the consumer thinks about at the time of capture, then we certainly have nothing against that being done at the time of capture. One statement our customers – probably not surprisingly – agree with is that the majority of the fun and the majority of the value of a photograph comes after the time of capture. That's not really surprising given that these people are our customers, using our software. But agreement with that statement is something that we've seen increasing over time. The consumers who are shooting every day are freed, if you will, by the post-processing aspect. I can just snap the photo and I know if it's not the proper exposure, I can deal with it later. Some people cringe at that. The artistry of the capture might be lessened for those individuals. But everything in our business is about segmenting the customer base; and general consumers think very, very differently to pro photographers.
TR: I'm curious what your thoughts are on the web. When will we see a full-blown photo- editing application running out on the cloud?
JM: Yeah. That's a great million billion dollar question. I don't think anybody can predict right now when the world will be ready for that. Bandwidth is a huge part of it. Particularly upload bandwidth can still be very slow today. Ourselves and other companies are providing solutions that give decent responsiveness and a good kind of fluidity of experience, but we can only do that to a certain level. So they are more basic applications. Photoshop Express is aimed towards a mass market; it uses Photoshop technology under the hood but it's geared towards the person who has simpler editing needs. It's our real first big foray on the imaging side into a completely server-based application. There's debate internally as to how fast we should go, how long will it take and how much of a role should we take in pushing that, but when it comes down to it, it seems like the commonality is bandwidth. And the ubiquitousness of bandwidth is the key issue.
TR: Talk a little bit about Adobe Labs. Are you peeling back the layers, saying 'you can have a peek at what we're doing'?
JM: Yeah, I'd say we are and I think that's increasing over time. One of the factors is our joining up with Macromedia. There's been a cultural shift within Adobe because of that. A lot of Adobe Labs' foundation, if I'm not mistaken, is coming from Macromedia's lab side. And you're right. There are more and more projects up there. We're exposing things earlier. We've had a lot of success with public betas, particularly with Adobe Photoshop Lightroom, and that's something that we've been looking at for a lot of other products. There's definitely more openness, and more people are of the opinion that there's no substitute to many people trying something at very early stages. There are still folks internally that cringe at some of the openness and some of the things that are being revealed, but by and large the cultural shift has occurred and people see the value of it.
TR: With such open development, you must get thousands of messages from customers making suggestions. How do you get through them?
JM: I wish I could say there's a super formal process, but there's definitely not. We do have a very formal process internally, in terms of looking at what should be implemented in the next versions of the products, but the specific system around that differs by product team. Everybody on every team is looking at the forms, is looking at the responses on the blogs as well as looking at what is out there in the competitive space, which has exploded in recent times. We have an internal 'competition list' that is bubbling over with online photo editors, online video editors and small downloadable applications, so we keep a close eye on those. We've got more than enough to do.