The first issue is about engineering mature software products.
It is a painful reality that all software decays. Some call it ‘code rot’ or ‘software erosion’, whatever the name, it is inevitable that software will slowly become difficult to maintain or even unusable.
First, the hardware that the software must run on is constantly evolving, so although you may have a working software and hardware combination today, one day the hardware will fail and you will have to replace it. At that moment you will discover that the old software does not run on the new hardware, or some key feature fails to work (this is particularly the case where the software meets the hardware in graphics cards and other peripherals).
The user requirements don’t stand still either. They will face new file formats and standards. User expectations, particularly in the realm of the user interface, ease of use and responsiveness will change over time too.
At the same time a mature software application becomes harder and harder to add features to. Think of it as a complex mechanism. Every time you add new functionality, that complexity increases, and that gives you more things to consider when adding the next new feature. The chances of breaking existing functionality when you add new features goes up dramatically, and consequently testing becomes harder and more time consuming.
Ultimately this pile of complexity becomes so bad that the engineers insist on starting again and ‘rewriting from scratch’. To achieve this takes a whole heap of time and you inevitably end up with fewer features (as the pressure to release the product eventually overtakes the engineering attempts to recreate legacy features). The inevitable reaction from the market is “Why did we have to wait so long for less features?”
The second issue is about selling a successful mature software product.
If the product has gained significant market share, there will be a diminishing pool of new customers. New functionality is added to swing those customers. Remember that each new feature is more and more expensive because of the ‘code rot’, and each new feature attracts fewer and fewer new customers. So the cost of acquiring each new customer is going up exponentially.
At the same time the company goes looking for ways to expand its market. For any company that is in a tiered market (‘professional’ to ‘consumer’, ‘specialist’ to ‘generalist’, etc) it will inevitably seek to go ‘down market’.
The Final Premiere Pro X Cut Story
What is interesting about the latest round of the FCP X story is that it is linked to its biggest competitor, Adobe Premiere, by (at least) two people. I’ve patched together a combined history.
Star video software engineer Randy Ubillos starts writing Premiere at Adobe and they release the first version in 1991. Randy stays for 3 or 4 releases and then jumps ship to Macromedia to start again with Key Grip, shown privately at NAB 1998. At around the same time a product manager called Richard Townhill starts work at Pinnacle (who also do editing systems).
After some horribly convoluted licensing issues between Microsoft, Truevision and Apple (don’t bother to ask), Apple buys Key Grip, Randy follows, and they show it publically for the first time as Final Cut at NAB in 1999. In 2001 Richard moves to Adobe and oversees the re-writing of Premiere, which after around 12 years must have had some ‘rot’, so they renamed it Premiere Pro. He hung around for a year or so before heading off to … Apple, in 2004.
In 2007 Randy tires of FCP and starts work on iMovie.
After Version 7 of FCP in 2009, it looks like they started work on building a completely new FCP based upon the iMovie code base. A ‘rot’ based decision? This is released in 2011, 12 years after the first release. Richard does the presentations at NAB where he announces that although Final Cut has 1.5m ‘users’, they are aiming for 3m with FCP X. That was the hint that the professional video users should have picked up on.
Confused? Here is a simplified timeline for you:
I get a feeling of déjà vu, Randy and Richard must feel like they are stuck in ‘Groundhog Day’.
Who is FCP X Aimed at?
So when FCP X finally appeared, what were the features that were lost in the re-write? Well, certainly not the user interface which looks gorgeously slick and neat. Ease of use is right up there too, with lots of things to make it easy for people to get the footage looking nice without having to try too hard (by the way, professional editors don’t make the footage look nice, they get colourists to do that for them).
Capturing footage off tape or recording it back to tape? That’s gone. Loading old FCP v7 projects? No need for that. Handling complex multi-camera productions? Nope.
I’ve written before about the small size of the film and television post-production industry, but Ron Brinkman, who I like and respect a great deal, has related that to Apple’s decision in a recent post: http://digitalcomposting.wordpress.com/2011/06/28/x-vs-pro/. He guesses that there might be 10,000 high-end professional editors in the world. I think there might be 200,000 professional editors in the world (not necessarily ‘high-end’). What the heck does it matter? Looking at Apple’s target market of 3m users, Ron’s figure represents 0.3% and mine 6%. Either way, I’m sure it is below the radar as far as Apple Product Management are concerned.
Are you getting the message yet?
A Final Chasm
I think Apple faced a classic problem (The Chasm to those in the know) in developing the market for FCP. They had saturated the professional market and had to look to the semi-professional and pro-sumer (professional-consumer) markets to grow beyond it. Apple are a consumer products company, for whom professional users concerns are a bit of a pain in the ass with a whole bunch of complex needs to solve everyday workflow problems. As Ron Brinkman puts it: “Apple isn’t big on the quotidian.”
So often, as you move from your early adopters to your majority market, you have to change your focus and concerns dramatically. Your majority users will have very different needs. In this market they are: ease of use, quick to learn, digital media camcorder formats, make it look nice, low skill, etc. This compares dramatically with the professional market which is concerned with workflow, depth of features, complex productions, delivering video tapes, and speed of use. They’ll learn how to use the most complex feature if it saves them time. The fact that they managed to keep both professional and hobbyist users happy with FCP v7 is remarkable and a testament to their product design.
So to make the product what your majority market needs, you sometimes have to do the opposite of what your early adopters wanted. If you piss off a few thousand gnarly old editors whilst chasing your $1bn market, well that’s just collateral damage like having to break the odd egg to make an omelette.
How should the professional users act in such a market? Ron has some advice:
“So if you’re really a professional you shouldn’t want to be reliant on software from a company like Apple. Because your heart will be broken. Because they’re not reliant on you. Use Apple’s tools to take you as far as they can – they’re an incredible bargain in terms of price-performance. But once you’re ready to move up to the next level, find yourself a software provider whose life-blood flows only as long as they keep their professional customers happy. It only makes sense.”
Think it through, guys. If you want to get the attention of manufacturers, you’ll need to offer them a respectable market. If we asked you 200,000 editors to support a 1$bn market, you’d need to be paying $5,000 for your editing software. If we asked just the 10,000 high-end editors … oh forget it.