Keeping customers retained by scheduling updates at regular intervals, keeping lapses in software licensing to a minimum, is standard practice for many software companies.
I think it is a bit more complex than that. In the olden days until like 15 years ago one would buy a software license for a major version that typically would include minor updates for this version. When a new major version popped up one would have to buy an update to that version and the process starts again from the beginning. So no need for frequent updates but rather for big bang ones from time to time as these were the things that created income (and to deliver that they had to have relevant features).
Some software vendors would allow to leave out one major version and still being able to upgrade to a new major version, some did not. Those who would allow to leave out major versions would typically charge a bit more for such a version-jump-upgrade.
Optionally one could buy support, typically used in more professional or enterprise environments, which would typically cost between 15 and 50% percent of the license fee per year, depending from the support level (standard to platinum) and wether major updates were included.
Time based licenses (that would allow you to use the software you bought only for a year or so) were pretty uncommon back then. This model worked well for many years but started to change about 15 years ago - in two different directions at the same time:
• on the one hand, Apple introduced the app store which had two effects: software became dramatically cheaper as people were not willing to pay relevant money for it, there were loads of fee apps available that got their revenue via in-app advertizing and via the central app store people were using (or at least downloading) way more apps than they did before. As Apple did neither implement the possibility for free demo versions nor the possibility for paid version upgrades effectively in the early years you bought an app once and had all updates free forever. Users loved it, but surprisingly, this turned out not to be a sustainable business model for software developers. And so in recent years more and more followed the alternative route:
• on the other hand with the upcoming of easy to implement automated build and deployment systems and easier implementation of automated testing it became way easier and cheaper to release new software versions than before. And with distribution moving from physical media like CDs or DVDs almost completely to internet download thanks to fast and cheap internet connections plus easier options for online payment and online distribution frequent releases of software became way easier. And so updates became smaller and more frequent. This was good, as new features were delivered faster now plus smaller updates created less risk in comparison to bigger ones as less new code an less code changes were invented.
But at the same time with this approach it became harder to create new major versions (still needing a big bang featureset to justify the version change) and artificially holding back features while there was no technical need to do so felt awkward (and rised the risk of bugs additionally).
As a consequence software as a service (SAS) became the new hot **** for software vendors: You no longer buy the software but rent it for a period of time for a monthly or yearly fee. Often enough, it doesn't even run on your own hardware but in the cloud and you don't have to care for update or maintainance. Bugfixes, new features and enhancements are silently delivered in a constant stream. Plus the entry barrier is lower as you only have to pay a small sum per month instead of a huge sum upfront. Adobe was one of the first companies to invent this modell for their Creative Suite and many many others followed.
This model works brilliantly for software companies as they now have a steady and calculatable monthly income that is not coupled to having to release a major version (and only then earning money) while in between having to deliver updates and maintenance for free. Plus they do have a honogenous base of installations instead of having to support and to fix various versions on parallel as it was before. This way in theory the quality of software (from a craftsmanship perspective) was expected to go up as there was now less pressure to deliver shiny new features en bloc under time pressure but rather deliver continuously smaller features in smaller batches whenever they are ready and strictly focus on only one version only. Software companies and enterprise customers jumped on that train fast, and - having no choice - consumers did as well over time.
Fast forward to today: Over time customers have increasingly started to realize that with this new model they typically quickly pay way more money per year than with the old model for basically the same value while software vendors have realized that they are in a very comfortable situation, owning a money printing machine (especially with cloud based software) and getting a sustainable monthly revenue w/o a real obligation to deliver anything. Which often enough is the case: SAS is an expensive money drain for many.
Coming back to XenForo: XenForo is still pretty much on the old model (which I like and which was one of the reasons going for XenForo) while, with XenForo Cloud also serving the new market. Personally, when starting with XenForo two years ago, I was alien with the idea only to buy an extention when a major or relevant update pops up and still am alien to that approach. I planned to pay for the extention constantly to have constant support and updates (and stacked up a couple of years last year when there was the sale before the price rise), also because this was what I was used to from many professional environments over the years.
However: It turned out that, until now, community support via the forum was sufficient - no need for an official support ticket until now. It has often been criticized that time between XF releases would be to long - which may have been true over the last couple of years. Currently I feel it is rather fast in the area of
planning at least, rushing through a couple of 2.3.x releases quickly, aiming vor 3.0 and more or less spontanouosly planning a 2.4 in between. Until now I find it more irritating than useful as each update creates effort and risk, so the need for a bunch of bugfix releases is a bit of an annoyance. More of a moving target than anything else. Which is then eased up as for the moment I am stuck on 2.2. anyway due to add ons and - as the new releases did and do not deliver much in terms of features that I would be desperately eager for - I am mainly forced (and more and more eager) to upgrade as add many on developers only add features to their 2.3. versions while at the same time I cannot upgrade due to other add ons not being ready for 2.3 yet.
So today I do have a way different perspective on XF than I had when I started over with XF two years ago:
• back then I though XF would be the main playground and I may or may not need support for it. Being a commercial product It will be developed further in terms of features continuously and decently fast, I may however in rare circumstances have the need of a third party add on (but this will - given the pretty basic needs that I have - be the exception).
• today it turned out (until now at least) that XF itself is rather a kind of rough foundation - it is way more limited that I assumed, it hit borders really quickly and the "need" for using add ons is immense as soon as one wants more than the basis of the basis. On a positive note XF itself does run w/o any issues, stable as an old Mercedes Diesel taxi and with the help of add ons you can achieve many things easily. But the price for that is complexity: The need for a lot of addons makes maintainance and upgrades of software way more complex due to many (uncoordinated) players (each following their own rules, policies, strategies, characters, price-models and timelines) and dependencies between them and it makes it way more expensive than I initially thought (as the cost for add ons stack up to more than the XF license quickly) and calculated. I vastly underestimated what XF themselves state: That they are a platform that things can be build on top of (not explicitely telling the part
"and you will probably need to do so quickly" 
). So in the end in comparison it is rather the add ons that bring me to or stop me from upgrading the XF version, not so much XF itself (as I - while clearly honoring the efforts going into new XF versions - do until now barely see features or changes that would make an upgrade of the main stack mandatory or even worth for my needs). But - and this is exceptionally good - this is only the case because XF does not hop from one security hole to the next but seems to be on a rather good quality level as far as I can judge and therefor there seems to be no need for fast updates (or updates at all) from a security perspective.
Ironically this situation is self enforcing: Add Ons are developed because XF lacks a certain feature or functionality. XF relies on add ons being developed and therefor does not include too much new functionalities and features in the software itself (makes life easier for XF plus it would be annoying to make a well perceived add on worthless and cut the income of the developer who sells it). So the situation stays as it is and becomes even more complex over time. But this lack of features makes XF updates less attractive and that makes license extensions less attractive while at the same time even with a license extension an update of XF may be cumbersome or even impossible due to add ons. The fewer license extensions are sold the less money does XF make and the less financial capacity is there to develop features. Which, as a consequence leads to slow update cycles with only little features which then again stops people frombuying license extensions. A self-enforcing process to a degree.
So it is not only the frequency of updates but also the content of updates (in terms of features that are relevant to me), their quality (in terms of bugs) and the dependencies in regards of add ons that come into play and that may influence the willingness to upgrade and to buy yearly extensions constantly. I am eased up for the forseeable future as I have stacked up extensions but can in the meantime understand people that rather do a little cherrypicking in terms of extending their XF license.
We’re good. I promise.
