5 ms·

While much of the blame still falls on Apple for pushing developers toward subscriptions in the absence of a straightforward paid-upgrade mechanism, requiring a

While much of the blame still falls on Apple for pushing developers toward subscriptions in the absence of a straightforward paid-upgrade mechanism, requiring a subscription for an app with no cloud dependencies remains an instant disqualification for me. It's buying into a promise that your monthly payments are going toward continuous improvement of the product – a promise that might be kept, or might not – and even if it is, for how long? If updates stall out after awhile, or if the new features don't solve anything for you, it's not like you can choose to skip an upgrade. And the more you use the product, the higher the barrier to switching away. I agree, Craft looks lovely, but there's no way I'm locking myself into its business model. I say this every time, but this is a solved problem. Developers of software like Sketch and Bitwig Studio have the right kind of subscription model, where you subscribe to updates, not access. If you don't care for the latest version, you can let your subscription lapse and keep using the last version you're licensed for. Apple has all the pieces necessary to simply gate App Store updates via subscription, but they apparently have no motivation to do so.

_jal 6y ago
Related, but the push to subscriptions coupled with my reaction has devalued my phone substantially. I just do not subscribe to many things. I don't like assigning myself future financial tasks, and I don't like dealing with recurring payments[1]. As a result, I've stopped using a bunch of things that went sub-only. There are more I'll be dropping. And this has moved a lot of things I did on my phone elsewhere. This, combined with the fact that the Iphone X (I have no need to upgrade it) is a bit too big, and the Screentime thing tells me I use it an average of 17 minutes a day, which is going to mainly be 2FA and texting with family. I mean, I kinda welcome this - I don't like phone dependency. And I expect I'm not the median case. But Apple's strategy has backfired with me. [1] I realize this is not strictly rational, and I have paid more for one-time-purchase software that I used once than a sub would have been. It doesn't matter.
- hackerfromthefu 6y ago
  Completely agree! Subscriptions cost future hassles and distraction, as well as the money. I'm busy. No subscription, no deal :D
- stjohnswarts 6y ago
  with 17 minutes a day you definitely are not the median case. i think you're more of an outlier these days (probably outside 3 sigma that people actually care about). I can see why you wouldn't want any subscription services.
opsy2 6y ago
Gating app store updates is not so simple- what about security patches or fixes when a new iOS version breaks everything? Still could work but isn't easy, and probably throws a wrench in the whole modern idea of continuous develop -> deploy.
- novok 6y ago
  TBH I think it's fine to prevent access to updates that update the app to work with the updated OS. That is work and I think it's fair to charge for it.
  - alisonkisk 6y ago
    It's absurd to use an OS that will arbitrarily lock you out of your old apps when you update
    - psifertex 6y ago
      It's absurd to think that OS updates should in perpetuity guarantee compatibility with all prior versions.
- mortenjorck 6y ago
  The App Store already does exactly this, though. The only difference is that the gating is by OS version rather than payment. An Xcode deploy target can require a minimum OS version, and when an app's minimum version increases with an update, the update is not available to users who haven't updated their device. If the device doesn't support the new OS version, the user is stuck with the last version supported on the old OS, and will not receive any future patches.
haswell 6y ago
I hear what you’re saying, and even agree with it in principle. To provide a counter example, I subscribe to Ulysses. They embody “subscription for access done right” (at least so far). I liked the app enough that I bought into that promise, and so far, the developer had delivered with excellent updates over time. What I dislike is that there’s no guarantee this will be the case. But to be fair, this is how reputations will be built: both good and bad. Developers who build a bad reputation will not succeed in the long run. It sucks that someone had to be the guinea pig though... So while there’s risk in buying the promise, it can pay off. I’d still prefer the “subscribe to updates” approach.
z3t4 6y ago
As software development works today, the subscription model is the only logical business model. Unless it's like a single player game where you only play through or use once. Software today is an continuous process. When a software product ships it's far from finished. The subscription model is good for both users and developers, as it allows releasing a software product early, without a huge marketing and advertising budget. The incentive is not to make you buy it once, the incentive is to make you keep using it, by improving it, or keeping it good, take good care of customers, etc.
- setr 6y ago
  > The incentive is not to make you buy it once, the incentive is to make you keep using it, by improving it, or keeping it good, take good care of customers, etc. The problem is that this isn’t the only incentive, or rather, not a well-defined version of the incentive at play. The incentive is to simply keep the customer as a customer, over time. Multiple strategies will work, which includes being nice to the customer, but also includes lock-in strategies like data availability (proprietary format, no export), logic availability (custom proprietary language, stdlib, etc), etc That is, Jetbrains and Oracle both run on subscription models, with very different interpretations of how to process those incentives.
- trevyn 6y ago
  Not everything has to have a business model. Some people write code because they want to, and then give it away for free. If you plant a seed in the earth, it will give you food for free, too. Also, we’re starting to reach the point where money is an ineffective incentive.
  - dismalpedigree 6y ago
    I am sure there are a handful of rent seeking startups trying to figure out how to make sure you pay a subscription for planting that seed also.
    - sojournerc 6y ago
      I laughed, but then then felt sad about my deep cynicism. Merry Christmas!
andyfleming 6y ago
The problem with web-based software subscriptions like this is that you don't really have any option to fall back to a perpetual license with the version you're at. It's ok for shorter term use cases like specific projects, but it's hard to commit for longer-lived documentation or personal documents you want to maintain over years. Your usage may ebb and flow. Aside from that, the company may go out of business or just change its priorities.

_jal 6y ago

Related, but the push to subscriptions coupled with my reaction has devalued my phone substantially. I just do not subscribe to many things. I don't like assigning myself future financial tasks, and I don't like dealing with recurring payments[1]. As a result, I've stopped using a bunch of things that went sub-only. There are more I'll be dropping. And this has moved a lot of things I did on my phone elsewhere. This, combined with the fact that the Iphone X (I have no need to upgrade it) is a bit too big, and the Screentime thing tells me I use it an average of 17 minutes a day, which is going to mainly be 2FA and texting with family. I mean, I kinda welcome this - I don't like phone dependency. And I expect I'm not the median case. But Apple's strategy has backfired with me. [1] I realize this is not strictly rational, and I have paid more for one-time-purchase software that I used once than a sub would have been. It doesn't matter.

hackerfromthefu 6y ago
Completely agree! Subscriptions cost future hassles and distraction, as well as the money. I'm busy. No subscription, no deal :D
stjohnswarts 6y ago
with 17 minutes a day you definitely are not the median case. i think you're more of an outlier these days (probably outside 3 sigma that people actually care about). I can see why you wouldn't want any subscription services.

opsy2 6y ago

Gating app store updates is not so simple- what about security patches or fixes when a new iOS version breaks everything? Still could work but isn't easy, and probably throws a wrench in the whole modern idea of continuous develop -> deploy.

novok 6y ago
TBH I think it's fine to prevent access to updates that update the app to work with the updated OS. That is work and I think it's fair to charge for it.
- alisonkisk 6y ago
  It's absurd to use an OS that will arbitrarily lock you out of your old apps when you update
  - psifertex 6y ago
    It's absurd to think that OS updates should in perpetuity guarantee compatibility with all prior versions.
mortenjorck 6y ago
The App Store already does exactly this, though. The only difference is that the gating is by OS version rather than payment. An Xcode deploy target can require a minimum OS version, and when an app's minimum version increases with an update, the update is not available to users who haven't updated their device. If the device doesn't support the new OS version, the user is stuck with the last version supported on the old OS, and will not receive any future patches.

haswell 6y ago

I hear what you’re saying, and even agree with it in principle. To provide a counter example, I subscribe to Ulysses. They embody “subscription for access done right” (at least so far). I liked the app enough that I bought into that promise, and so far, the developer had delivered with excellent updates over time. What I dislike is that there’s no guarantee this will be the case. But to be fair, this is how reputations will be built: both good and bad. Developers who build a bad reputation will not succeed in the long run. It sucks that someone had to be the guinea pig though... So while there’s risk in buying the promise, it can pay off. I’d still prefer the “subscribe to updates” approach.

z3t4 6y ago

As software development works today, the subscription model is the only logical business model. Unless it's like a single player game where you only play through or use once. Software today is an continuous process. When a software product ships it's far from finished. The subscription model is good for both users and developers, as it allows releasing a software product early, without a huge marketing and advertising budget. The incentive is not to make you buy it once, the incentive is to make you keep using it, by improving it, or keeping it good, take good care of customers, etc.

setr 6y ago
> The incentive is not to make you buy it once, the incentive is to make you keep using it, by improving it, or keeping it good, take good care of customers, etc. The problem is that this isn’t the only incentive, or rather, not a well-defined version of the incentive at play. The incentive is to simply keep the customer as a customer, over time. Multiple strategies will work, which includes being nice to the customer, but also includes lock-in strategies like data availability (proprietary format, no export), logic availability (custom proprietary language, stdlib, etc), etc That is, Jetbrains and Oracle both run on subscription models, with very different interpretations of how to process those incentives.
trevyn 6y ago
Not everything has to have a business model. Some people write code because they want to, and then give it away for free. If you plant a seed in the earth, it will give you food for free, too. Also, we’re starting to reach the point where money is an ineffective incentive.
- dismalpedigree 6y ago
  I am sure there are a handful of rent seeking startups trying to figure out how to make sure you pay a subscription for planting that seed also.
  - sojournerc 6y ago
    I laughed, but then then felt sad about my deep cynicism. Merry Christmas!

andyfleming 6y ago

The problem with web-based software subscriptions like this is that you don't really have any option to fall back to a perpetual license with the version you're at. It's ok for shorter term use cases like specific projects, but it's hard to commit for longer-lived documentation or personal documents you want to maintain over years. Your usage may ebb and flow. Aside from that, the company may go out of business or just change its priorities.