Hacker News Reader

> It is unrealistic to expect to be able to pay once and use an app "forever".

No it's not. This is very realistic: pay once for a license to use the software, and then you pay a maintenance fee every year to get the updates and new versions. You stop paying, you stop getting the updates, but whatever you have already still works. There's no shortage of examples but I'll give a shout-out to one of my favorite software products, PRTG: https://shop.paessler.com/shop/prtg/new/?preselected_license...

That works for a stand-alone app, but not SaaS, a web-app, or an app that uses 'cloud features'.

Okay, but there are plenty of subscription apps that could just drop the cloud features and like 99% of the user base would be fine with that, Office and Adobe suites being obvious examples.

Sure, I don't intend to dispute that at all. There are also cloud-based apps that absolutely could not be, and at least part of the reasoning is probably to make it easier to charge a subscription fee.

I think the model's seen as being more lucrative, so even businesses that don't inherently need it so much try to find a way to adopt it.

Agreed.

You've pretty much described a subscription, with the difference that you expect that whatever you have already still works and that you completely ignore the problem of support.

We can pretend it will work for a while, but I do not think this is realistic.

Support is not something you can ignore: people will request support, post negative reviews and opinions, and you will need to deal with that.

My point is that we should all stop pretending and just accept that apps cannot be used without regular maintenance anymore.

> It is unrealistic to expect to be able to pay once and use an app "forever".

That's the other extreme though.

There's a middle ground that http://reaper.fm/ uses. You buy the current x and x+1 versions. Then you pay for the license again for x+2, x+3 if you want to. The price is very reasonable and there's no subscription as such. If you're happy with the old major version, you can use it as long as it continues to work.

In that case the number of users who would buy x+2, x+3 is probably not half as many as users who bought x, x+1 and the developers would no longer be incentivized to build any more versions, making such a business unsustainable. And newer versions of operating systems, especially Windows, break old software like Office 2003 to ensure that you are using the latest version of Office 365 so they can profit. You can stay on Windows XP to avoid that, would you?

Not true, some things need to be updated with the OS, like Parallels.

There are other and more important motivators in the world than money…

It is of course necessary to make ends meet and plenty of software shops do just fine with the business model of making a good product and charge for upgrades.

I'm not sure why you make that assumption. X+2 is going to be bought both by users happy with x+1 and by new users who only just learned about it. Why would that be lower than the original buyers?

> You can stay on Windows XP to avoid that, would you?

People do. But realistically either you'd upgrade your computer since XP was sold and get a new license included, or just pay for the update once in almost 2 decades. I don't get the issue you're trying to point out.

I got my license back when 5 came out just as I was getting into music in 2015. That covers me to 6.99, and it's taken 7 years to reach 6.64. That's one heck of a bargain.

I didn't realise that. Now I actually want them to cut a new version soon. They deserve my money again and then some.

That's the neat thing. They're always making a new version. They dropped a whole notation editor in with nothing more than an item in the changelog. The video compositor is always improving.

You get all of whatever version you start on (Even if you start at x.99), plus the next, and then your license is done unless you renew it. So no matter what, I don't get updates after 6.99 unless I renew.

Most of what I do music-wise nowadays is making songs live in Ableton Live on Twitch, and I rarely bother to take them to a "produced" state, so what I have in the current Reaper is plenty. I probably won't upgrade, but I'll keep recommending it.

Oh, I know they release new versions, I've been using them for over a year. I meant that I wouldn't mind if they released a major version every year or two. Still worth it.

No one should be able to stay on an outdated version, both for security and compatibility reasons. Supporting old versions of you have an api attached is a cost in it self as well.

It's a music editing app, not a public internet facing service. There's an argument to be made for providing updates, but "No one should be able to stay on an outdated version" is just being a dick to the users. Don't disrespect the users.

Shouldn't software be cheaper than ever? Features certainly haven't increased much. And we now have magnitudes more processing power, memory, bandwidth and storage space. And improved tooling should have made it much much more efficient process? So what has failed? The entire field? Do developers really deserve the expensive pay if their efficiency has not increased?

That's like: Metal and materials are better now than ever before, so shouldn't it be simple, to design new airplanes, nowadays? Why do the aerospace engineers get paid so much

Due to all the frameworks, n0de junk, comp0se blather, I've seen devs write 200 lines of code, for what would have taken 20k lines in 2005.

Devs have it do easy now, that many are so abstracted away from DB, OS, network stack, persistence, that they literally have no idea how anything really works.

This is desired of course, devs today are in short supply, just like the bow vs the crossbow, years of practise turns into months.

But the point is, it is indeed far far far easier to deploy software today.

(why the 0 and now double u usage in place of w with auus?

I suspect auus, n0de, c0mposer dependant entities, such as auus,spend time defending their business model. I want honest discussion, not "omg, defend n0de" discussion, which latches on mention of such tech and pulls in the army)

What's "auus"?

I think systems nowadays are larger and more integrated, more complex.

Deployments sometimes are simpler, whilst other things instead are harder. Software will always be pushing the limits.

Aha auus is AWS, I didn't read that part of your comment properly.

Maybe you don't want the comment to be found by search engines

Features have increased, at least for the kind of apps that I write. My software does things which are quite complex, and were definitely not present in this class of software several years ago. Better languages and libraries make this possible.

In general, I do think software should be cheaper than ever. The main reason it isn't is the crazy pace of changes in platforms. I maintain a web app, which helps me keep my sanity, but I'm not even considering writing mobile apps: there is no way for me to keep up with the pace of changes and updates that vendors make there. It's just way too expensive.

It may be unrealistic for your business plan and your product (which is fine, no criticism here), but the existance of software that still uses the pay-once-use-forever model proves that it is possible for many apps without being unsustainable. E.g: Serif / Affinity, JetBrains (as you still get a perpetual fallback license to the last version you payed for, so it still can be viewed as a one-time-payment, IMO - just cancel immediately).

Infinite support or updates and a one-time-payment is unrealistic, yeah, ruling out cloud-based SaaS. But for common desktop apps and on-prem stuff, that just isn't neccessary. Drop support at some (previously announced) point and the customer can decide if they want to upgrade. Adobe CS6 still runs fine.

One thing is that this doesn't currently work in the mobile space, because somewhat regular (~biyearly) updates are necessary to simply keep stuff working on the newest OS and hardware, and the majority of new software targets mobile, often even exclusively.

The problem with a subscription model is that there is a mathematical limit to how much MRR you can make. You can configure the limit by changing pricing, marketing, etc, but then that effects churn. There is simply no way to bypass this limit, no matter what you do unless you can remove churn completely, which is not possible with a subscription model.

With a one-off purchase, you can sell the same thing to a customer multiple times, you can sell upgrades, and you "magically" remove churn.

You own a SaaS, how long do your customers stay customers? If it's 8 months, sell a one-off purchase for the cost of 16 months. Bam, you've made more than you ever would from an average customer, and you don't have to hunt them down and try to get them to stay a customer. You can just focus on building an awesome product and not worry about churn at all. In a year or so, think about a new feature or two that increases the value of your product, sell that new feature (eg, discounted) to existing customers and increase the price for new purchases.

What is the breakdown of your software maintenance costs?

One of my ideas is that we should package manage solutions to problems at a level higher than just code and subscribe to them. We all depend on BSD sockets due to history. BSD sockets Just Works™ and people don't really evolve the BSD socket API over time. If only the rest of the software stack was as feature complete as BSD sockets.

Imagine a labour market that was also a package manager.

When library "xyz" upgrades from version 1.2.0 to 1.3.0 there's an interface change that breaks everybody. If there's 1,000 dependents of "xyz" and it takes 1 hour to upgrade and 1 hour to pass all QA tests and get into production, then that upgrade takes 1,000 man hours + 1,000 hours automated testing.

If we don't depend on library "xyz" throughout our codebase, we reduce the number of callsites to change. If we use a facade, they're all in one place. If we use a Service Provider Interface, we isolate ourselves even more and add flexibility of implementation.

If our facade or SPI is in escrow on a labour market, someone can fix it for us for a cost of their time. Ideally the software maintainer fixes it so they don't break so many people.

> The usual take is that businesses are "greedy" and subscription models are "lucrative".

"Greedy" as a criticism of a for-profit business makes about as much sense as "predatory" does as a criticism of a cat.

It does make a lot of sense actually. Because cats are predatory, we need to regulate our domestic cat population by keeping them indoors, otherwise their predatory nature will destroy local populations of birds and other animals.

So a for-profit business is greedy by nature, so we need to regulate its natural impulses since it can’t control them itself, much like the cat.

I think greedy is still a valid criticism - just because it's "for profit" doesn't mean "max out the profit at the expense of others".

If we assume that not all people (as it is people who drive these organisations) are greedy, we can also assume that greed is not baked into every profitable org.

Windows 3.11, Office, Windows 95 etc seem to have done ok without monthly payments.

And they were all funded in the paid update model. Subscription is a consequence of products running on the cloud (where you certainly wouldn't want to operate old versions for customers who refuse to update) and of the lifetime background updates established by app stores.

That's exactly what should be. Pay to upgrade to major version, no subscription. Functioning app, fixing errors should be included in the selling price.

Monopolies are great! /s

Microsoft respects developers. Google doesn't