Subscription to software has been on my mind a lot lately thanks to my ongoing review of Ulysses, which transitions to this business model a few weeks ago. So when the Mac Power Users podcast recently published their Developer Roundtable, I was was hoping the question of subscriptions would be discussed. I wasn't disappointed.
To back up a little, the roundtable feature developers from the OmniGroup, Smile Software and AgileBits, the developer of 1Password. Each of these companies make productivity software for the Mac and iOS and have a number of different business in place for their products.
The developers from Smile and AgileBits gave compelling arguments for subscriptions, include: lower barrier to entry for consumers, better support outcomes and more sustainable forms of income for the developer. All of these points make sense and the last thing I want to see is developers I depend go bust.
The developer from the OmniGroup however made what I think is the strongest counter-argument against subscription and that is life-span. When you buy a software licence it's not unreasonable to expect that you can run that software for as long you have a computer that will run it.
I've bought several productivity apps1 that are either dead (Filemaker Bento, discontinued (PDFClerk Pro) or that have a very uncertain future (VoodooPad). I also have licences for several apps I've chosen not to upgrade for various reasons, such as Adobe Creative Suite 5.5, Panic Coda, Mellel and Mariner Write.
Each of these apps continues to work. I can still open them and retrieve data; if I choose I can still use them for productivity in the way the developer originally intended2. The only requirement is maintaining a computer with an operating system that can still run the application.
If these apps had been released under a subscription model, I would not be able to do this. Developers can discontinue products, developers can go bust and the minute the app phones home and finds its server has gone dark, the software will either lock you out completely or reduce the application's features to a much more limited set.
I've reconciled all arguments in favour of subscriptions. I cannot yet reconcile the risk of losing access to a productivity app should this happen. No developer has adequately or publicly stated what they would do in the event of going out of business or deciding to kill an app. The moral thing to do would be to release the software's source code and allow the community to take over its development. Short of that, the next best thing would be to release a binary version of the software that is no longer bound by subscription, activation or other forms of licensing3.
However, if the developer goes belly up or is killing a product in favour of something new, I can't realistically see them doing either. Ulysses developer, The Soulmen, killed their product Daedalus Touch in part to focus on Ulysses, of which its move to iOS made Daedalus obsolete. I understand the reasoning but it's the precedent that bothers me.
Still, there's equally an argument (mostly made by developers) that buying into a subscription service makes long-term success of an app much more likely. This assumes there's no consumer backlash.
As more and more high-quality applications adopt this model our choice as consumers will be to hop along for the ride, or duck out in favour of the few apps that keep to the old model or switch to open-source alternatives.
I still have computer games I bought in 1990s that still work, albeit on emulated operating systems! ↩
I don't any more, with the exception of PDFClerk Pro and the Adobe Creative Suite. ↩
Scrivener for Linux and Adobe Creative Suite 2 both adopted the latter approach when abandoned and TextMate more or less adopted the former. ↩