I can name several companies off the top of my head that couldn’t ship an update quickly enough and went under because no one would buy until the update came out. Income was extremely spiky and hard to manage.
If this is the case, the software industry has only itself to blame. Whether knowingly or not, software developers have conditioned their customers to expect a never-ending parade of updates and new features. I remember when software updates—and especially major upgrades—were released fairly infrequently, and computer users bought software when they needed it. The notion of waiting for the next update didn’t really exist; or if it did, it bore no resemblance to the sometimes feverish anticipation that nowadays drives many people to start speculating about the next update on the same day that a brand new update is released. This phenomenon isn’t restricted to the software industry either. It looks to me like the consumer economy has become obsessed with novelty for its own sake. New features are not necessarily improvements at all—they’re marketing gimmicks.
My memory is that back then, initial purchases were very expensive but upgrades were cheap. The market was expanding so quickly that new purchases paid the actual bills, and the upgrades mostly covered the cost of sending out the disks and packaging rather than supporting ongoing development. More expensive upgrades and/or subscription plans came about once this initial growth curve flattened out and new purchasers were no longer sufficient to subsidize the rest.
(And many new purchases were wrapped into hardware purchases, at significant discounts compared to standalone prices.) Student discounts were also significant, with the hope that if more students used the software, it would increase the likelihood that their future employers would buy it at the more expensive price.
I will note that I spend my days programming in SAS, which has never been available as anything other than an annual license/subscription.
Why are you limiting your poll to digital subscriptions? I started an audit of my subscriptions to help me decide whether particular bundles made financial sense (SetApp, Apple One), and ended up amazed at the amount I was spending on subscriptions of all sorts, from AppleCare (which is a load of separate monthly subscirptions these days), to automatic repeat orders of products (i.e. from Amazon, purveyors of vitamin supplements, &c), to podcasts. At the moment I find myself adding to the list all the time as I remember more things that I pay for on an automatic recurring basis. Some of the ones you list already bleed over into the analogue world (publications, fitness gear). One digital category you’re missing is networking/communications (Zoom, SRFax, Slack, VPNs).
Perhaps left out of this discussion are the subscriptions associated with cable plans (a subscription?). I have Spectrum and they are now including services such Disney+, ESPN+, Hulu, Peacock, Paramount and others are a part of their cable package. They also bundle, so your internet package cost us based on what your cable package looks like. Finally, you can contact them from time to time a get significant price reductions without giving up any access.
Purely to keep it manageable and within scope for TidBITS. But thanks for calling out the networking/communications category—I’ll make sure to include that.
Sure, but there was no way to avoid that happening without collusion. In any competitive landscape, Company A is going to see an opportunity to ship before Company B, which only drives Company B to have an excuse to ship sooner the next time.
And, to be fair, today’s software doesn’t live in the happy vacuum that used to exist. Apps were isolated from one another and Macs were largely isolated as well. Now every app and computer is online 24/7, with all the good and ill that brings. Just look at the updates for BusyCal and BusyContacts that I just published. There’s an incredible amount of interoperability that’s involved, so if Busy Apps does nothing, its software starts working worse and worse.
Not so cheap. @omalansky remembered spending $199 every three years for Photoshop.
The market was expanding, but it was vastly smaller than it is now, so the people taking advantage of the software could make a good living at it and were willing to pay for the tools. The issue now is that although the market is much, much larger, the tools have all been commoditized for most uses, such that they’re free or very, very cheap. And when that’s the case, the chance to make money on upgrades becomes far riskier, especially if the upgrades do nothing but maintain functionality in an ever-changing world. Even if they add features, users may be uninterested in them and choose not to upgrade. Then there’s no revenue for the ongoing work of staying afloat.
Subscriptions also remove any incentive to provide regular updates if the application can only be used as long as the customer keeps paying. The money is coming in anyway and the customer can’t decide to stop paying without losing access to the application. Jetbrains has what I’d consider a fair and reasonable subscription model - you get updates as long as you keep paying, but can continue using the last version when the subscription comes to an end.
I understand how this could happen; but Adobe has done the opposite. They’re always updating their apps, and some versions are much buggier than others. They never leave well enough alone.
But thanks for calling out the networking/communications category—I’ll make sure to include that.