Hacker News Reader

Comment by techsupporter | Hacker News Reader

At the time of this writing, all of the comments here are focused on content, but I want to touch on the author's bit about software:

> It’s not just Bloomberg and media — it’s software too. I used to write everything in Ulysses, a syncing Markdown editor for OS X and iOS. I paid $70 to buy the apps, but then the company switched to a $40 a year annual subscription, and as the dozens of angry reviews and comments illustrate, that price is vastly out of proportion from the cost of providing the software (which I might add, is entirely hosted on iCloud infrastructure).

I have noticed this as well, that a not-small number of software programs have turned from "give us one money for this version" to "give us slightly smaller amount of one money per month, but do it in perpetuity if you want to keep going with the data you've entered."

That, to me, is crap and is incredibly frustrating, especially when the subscription model is coupled with a data lockout threat. A task manager, a word processor, even a drawing program...none of these ought to require a subscription to be able to use. Sure, have a subscription service for something that needs upkeep, like a syncing service (but get out of here if all you do is just lump it into the user's iCloud storage) or a data feed. But it's almost insulting to say "we used to charge you $50 once, now we're charging you $29.95 per month forever (if you want to see those notes in two years, and no, we don't have a data takeout API); it's such a deal!"

View on HN

A model I really like is the one IntelliJ uses. They encourage you to buy a subscription. But when you buy a year at once, you get a "perpetual fallback license" such that you can always keep using the version that is current when you buy: https://sales.jetbrains.com/hc/en-gb/articles/207240845-What...

That means I always will have access to my projects and to the level of functionality I start with. But I keep getting upgrades as long as I keep paying, and they get the steady revenue stream that means they don't have to play weird marketing games.

Developer Express has that too with their .NET components. We have dropped the subscription but are still using a version that's 3 years old but good enough for us.

You also get the perpetual fallback license if you pay month to month for a year, for whatever version you had 12 months ago.

E.g. I start paying month-to-month at 2016.2, after 12 months I have a license to 2016.2. But say after the first month, 2017.1 comes out. Then after 13 months I now have a license for 2017.1.

One thing that always rubbed me wrong about this is that, logically speaking, if you buy a license for a year, the perpetual fallback license should be the version you had at the end of the license, not the one that was on the market at the beginning of it. It doesn't really make sense the way they do it, and it feels like a particularly hamfisted attempt to grab an almost negligible amount of money, so I'm puzzled.

It's not the negligible amount of money that matters. The trick is that at the end of your yearly subscription you have already upgraded to a newer version, so falling back to the perpetual license means downgrading, which is both often a practical problem and (perhaps most importantly) something that I think many have an emotional aversion to.

I'm pretty sure that if the fallback license was to the most recent version you're entitled to during your subscription way more people would take leaps between the end of a subscription and the start of the next one (though Jetbrains offers you a further discount if you don't do so).

That makes sense, thanks.

In the end, though, when I discovered this was how it worked at the end of my PyCharm subscription, it left me quite a bitter taste in my mouth, and I just stopped purchasing any JetBrains products. These kind of details are akin to trying to trick people via putting stuff into EULAs nobody reads, and for me at least, it gives the distinct feeling that they're trying very hard to screw me over, so I try to not frequent nor support those kinds of businesses, IMHO.

So, is what they are doing worse than what most subscription models do (lose complete access)?

I've been using them for awhile and I felt they were really up front about what they were doing, but I don't know if they've always been as transparent.

It's definitely better than that. But it's still not good. This is all my personal opinion though, I already voted with my wallet by not buying their products any more.

Jetbrains actually wanted to cut off access completely at first.

Initially JetBrains didn't even want to offer rollback licenses when they went to subscription only model a few years ago.

There was a big uproar among programmer communities (including here on HN) and after few weeks of deliberation a compromise of sorts was reached.

Despite offering a nice product line, JetBrains did lose a bunch of goodwill including mine. But goodwill does not pay the bills.

Would it have been better if you "purchase the current version and get temporary upgrades for a year for free." Essentially the same thing, but up front and not implying that you get more than you do.

It'll be the same only if at the end of the year your version is rolledback.

I think this:

> The trick is that at the end of your yearly subscription you have already upgraded to a newer version, so falling back to the perpetual license means downgrading, which is both often a practical problem and (perhaps most importantly) something that I think many have an emotional aversion to.

Qualifies as

> a particularly hamfisted attempt to grab an almost negligible amount of money

OK, so maybe it's not as negligible an amount but it feels petty and I'd not pay money to a company that pulls stunts like that on its customers.

> falling back to the perpetual license means downgrading, which is [...] something that I think many have an emotional aversion to.

They're playing on loss aversion and the endowment effect. It feels worse to lose something than to never have it, so people are more willing to pay to keep it than they would be to pay to buy it if they were never given it in the first place.

https://en.wikipedia.org/wiki/Loss_aversion

> One thing that always rubbed me wrong about this is that, logically speaking, if you buy a license for a year, the perpetual fallback license should be the version you had at the end of the license, not the one that was on the market at the beginning of it. It doesn't really make sense the way they do it, and it feels like a particularly hamfisted attempt to grab an almost negligible amount of money, so I'm puzzled.

You are absolutely right! That's exactly how it should work but it's surprising that they don't do it that way. I can give you an explanation from accounting perspective: When the sale happens, the accountant would record the sale in their Books of Account as "Unearned/Prepaid Revenue" (as a liability). It makes sense as you have taken money "in advance" for a sale but haven't delivered the updates yet and hence you cannot claim it as Income. Every time you put out an update, you can claim a portion of the "Unearned/Prepaid Revenue" as "Revenue" and push it from liability to "Retained earnings" account as Income. When the year expires (note that the year is for the user not the financial year) the accountant would claim the remaining Unearned/Prepaid Revenue as Revenue/Income.

I wonder how they manage their Books of Accounts if they have defined that the license falls back to the earlier version. What they are doing doesn't make sense from an accounting perspective as well!

I'm not an accountant but presumably you treat it the same as the fairly common purchase+service contract. You have a sale of a license (of the version at time of purchase) and the prepaid service contract that either converts to revenue over time, or, if non-refundable, converts to revenue immediately. The proportion between the two components is fairly arbitrary. There is no constraint on how they implement this from an accounting perspective - they can change to allowing the user to keep the latest version they upgraded to without serious consequence (you can just as easily sell "product version X with one year of updates" which is very common as well, and mark that as revenue right away).

> I'm not an accountant but presumably you treat it the same as the fairly common purchase+service contract

I'm not a certified accountant either but I took the course just to understand workings of my business.

> they can change to allowing the user to keep the latest version they upgraded to without serious consequence

It'll be in their best interests if they did.

> you can just as easily sell "product version X with one year of updates" which is very common as well, and mark that as revenue right away

You definitely can. But you end up overstating your revenue and in turn your profits. The downside to that is you end up paying high taxes (unless you can minimize your profit by showing a ton of expenses as well). If it was instead treated as Unearned Revenue (which is a liability), you would be stating your true income and showing lower profits and thereby paying lower taxes.

Take for instance the case where your customer purchases a multi-year subscription. This causes a higher cash flow (assets go up) but you won't be overstating your income (as liability goes up by classifying it as Unearned Revenue instead of owner's equity). The liability can be reduced by claiming it as revenue over a multi-year period. If it was marked as revenue at the time of sale, the owner's equity (retained earnings) would go up. That would mean showing higher income which results in higher profits and you end up paying high taxes.

Remember that with unearned revenue you can carry it over to multiple financial years and spread your income between them. You lose out on this important tax saving component if you mark as revenue right away. This also has the advantage of portraying a true picture of your business in your books of accounts.

Slightly off topic, but semi related. My parents had a one year membership at Costco. The level where you get some percentage back. I think it was 2%.

Long to short, the "rebate" was paid ~60 days prior to the end of the 12 months. The remaining 60 days' purchased will only be rebated if they renew.

I presume this is in the small print somewhere. None the less, it's complete crap.

Sounds similar to employers who pay contracted bonuses late, don't pay if you quit before the bonus date, and use bonuses to make up for reduced salaries.

This is really annoying for me, how can I work for a full financial year (and get my shares allocated), receive my performance review but not get a bonus because I quit before October where they pay the bonuses on the last FY?

Hence the wave of resignations the week of bonus payouts. People play the game and wait to get paid, knowing if they prematurely signal their departure, the company will reallocate money to those who choose to stay. I'm not saying it's right but it happens.

And then you only get half a bonus at your new role if you are lucky (my last job I joined 3 days too late to be eligible)

It's not "complete crap" given that (a) it's in the paperwork they are given at signup time, and (b) if they are genuinely dissatisfied, they can downgrade back to basic membership and get refunded retroactively to the start of the membership year for the difference in cost of the higher membership.

A year is a year. It's crap, and shows the brand lacks class and respect. Who would renew for a bunch of a-holes?

That's strange, as that was not my experience not long ago. I'm thinking there was some miscommunication about the rebate with your family.

I received my rebate after I had let my membership lapse without any issue. You can use the rebate even without an active membership. You can even walk up to the customer service counter and get your rebate amount in cash if you want. I ended up getting pizza and took the rest as cash back.

Thanks. I'll report back to The 'Rents.

Actually, I might have been an edge case and not representative of their normal rules.

"The Reward is issued approximately 3 months prior to the member’s renewal date in a Renewal Statement and reflects rewards earned up until the issue date. A member must be an active, paid Executive Member when the Reward is earned and issued to receive the Reward. To receive a refund for the Executive upgrade fee the membership must be canceled or downgraded to a Gold Star or Business Membership and any 2% Reward issued or accrued will be forfeited. Any additional savings and services exclusively for Executive Members will be rescinded." https://www.costco.com/executive-rewards.html

  You can even walk up to the customer service counter and get your rebate amount in cash if you want.

Here, cashing out is the only way they redeem the vouchers unless it's a really high amount (4 figures+), in which case they cut you a company check (that happened to me once in 6+ years).

Why should it work that way? You get the updates based on the assumption you are a subscriber, and you get the fallback license based on the assumption that you made a one off purchase at the date of purchase. It's logical the way it works.

It doesn't for the simple reason that they provide updates for the entire year. Surely those updates aren't provided for free. They are provided for purchase you made in advance for a year. If I cancel midway I should be entitled for all the updates till the end of the period. Unless those updates were free and not part of the sale. In which case, the licensee should still get those updates anyways for the period. You look at it either way it still doesn't make any sense. Or, they should return the update fee they charged for the year during the sale on pro-rata basis.

Also, how would you account for it? You can't retrospectively change the nature of purchase from subscription to a one-off sale! That would be audit hell for their accountant.

They sell you version at the time of purchase (fallback license) + access to updates, all accounted at the date of purchase. So when your subscription runs out in theory you could get old version then apply updates but you have no access so no updates for you.

> then apply updates but you have no access so no updates for you.

But that is denial of service that I legitimately paid for. Either they should provide the updates till end of subscription or refund the collected update fees pro-rata (viz. if I cancel after 9 months of purchase, 3 months of excess update fee collected should be refunded).

My sense is you paid for the software as it is on the date of purchase.

With that you get a trial of all updates for a year. If you don’t like the trial you don’t pay and revert to the version you paid for.

Yes it's possible to think of it that way and it makes sense now! However, they should explicitly mention it in their license terms that the updates are provided on trial basis and valid for a year and those updates can be purchased as a version upgrade at the end of the year.

It makes sense to me as removing the choice between, "Do I buy one-off license or a subscription?" If you get the subscription, you get a free one-off license at date of purchase.

Your way also makes sense, but they both seem reasonable to me. And I don't really care much, in that I renew when I'm regularly using their product, as I want to support them. If I just had to open an old project for something, then I don't mind exactly which old version I'm using. (And if I somehow did, I could just put it in 30-day trial mode.)

And at least the way I use it, there's no lock-in. If you stop subscribing, you can easily open your code in another editor.

Wait, so if I've paid the subscription fee for 2 years I get the one at the start of the subscription period?

Some lag from the latest version seems fair, e.g. 6 months before the latest major version is perpetual...

You're in luck. You get the version 12 months before the subscription ended, so the version released before the 1 year mark.

I think the solution is delayed gratification

Don't upgrade until after renewing the subscription for next year.

so you miss out on the new features/bugfixes (which you presumably want, since you are paying already).

As I recall, the perpetual fallback thing was added to appease their customer base, who were enraged when IntelliJ announced that they were moving to a subscription model. The term "rent-seeking" was thrown around quite a bit then and seems appropriate now as well.

Another, maybe less flattering, way of looking at this is that you're buying or rent-to-owning the current version and then getting demo copies of updates so long as you keep paying your lease.

They had a much more customer-friendly model before - you pay yearly updates and keep the last update. Now you keep the very first version only.

That's not entirely true. You can keep any version that you've paid one year worth of subscription for. That means that if you subscribe for three years then cancel your subscription, you'll always have access to the version that was released one year before you cancelled.

I think parent’s gripe is that if you end your subscription, you can’t keep any updates that come out during your last year, so you likely need to go backwards with your installed version.

It is particularly annoying if they introduced a serious bug, ie datagrip not refreshing queries, in the version you got when you purchased it, the one after the trial that worked. Then, a year later you get downgraded to a version that is useless.

They also drop the subscription rate after the first and second years. And even at the outset, the “everything JetBrains” license is cheaper than any two individually licensed products. If you do, say, Java and C# or Python and C++, this makes it a pretty sweet deal.

I wasn’t aware that the perpetual fallback license only covered the latest version at the time of subscription/renewal. That’s less than ideal, but apart from that, I think they’ve pretty well nailed the subscription-based model. It’s actually the _only_ software I’ve used where I feel the subscription model terms were actually an improvement over what I had before.

There are other software providers that charge annual subscription fees that will just stop support and upgrades when you decide to stop the subscription. They don't retard your licensed version down to a previous version. Some of them also allow you to buy a license for the current major version, with minor version updates for free, instead of going all in for a subscription. Heck, I spent 3 years getting free updates from one publisher. Having the choice of being able to buy outright or go on a subscription is, IMO, far better than subscription only models. I'm also pretty surprised that jetbrains took the decision to not only take away paid for updates, but not provide a rebate for people who don't want to continue with the subscription part way through. That just rubs me up the wrong way.

JetBrains made me super salty at how expensive their package is.

I graduated from college and to see their lowest cost package was ~300 USD IIRC was infuriating.

I went full Netbeans for PHP. It does the job, the debugger works, and it doesnt cost me a yearly fee.

It's gone up since then to $500/yr. It takes you two years before you get the $300/yr price again. By then you've already spent nearly $900, with the third year making it $1200.

Android Studio is free.

I think you're looking at the pricing for Businesses and Organizations (the prices displayed by default). Those are floating licenses. Most individual developers would buy an individual license. Individual licenses for the All Products Pack start at $249 for the first year, $199 for the second year, and then $149 for the third year onward.

Individual licenses can be used for commercial purposes just like a business license. AFAICT, the only difference is that a business license can be used by any single developer within an organization at a time, while the individual license can only be used by the named licensee.

There's also a 25% discount for new/recent graduates, which eases the sticker shock a little bit for new graduates.

I bought a copy of Office 2016 when it came out, it has now lasted me 2 years and I paid about the same as 2 years worth of Office 365. I intend to keep using it until it is no longer supported at all and I am sure I will have saved quite a bit of money for effectively the same product.

I got Office 2003 and it works for all of the home based stuff that I do now. I paid $70 on a sale. If I had been doing 365 since then at $48/year, I'm looking at $720.

I agree that office 2003 has enough features and is good enough for 99% of the things I use. In fact, starting with 2013, there are anti-features (integration with OneCloud, online docs) that I'd prefer were not included.

Unfortunately, Office 2010 already has issues opening files made with 2013, 2016, etc. Most of my PCs are on 2013 as I was able to grab several HUP licenses at the time. I seriously doubt 2013 will be 100% compatible in another year or two, and 2016 will be the minimum version one needs in a work environment where you're collaborating on docs made with newer versions.

I have no doubt that Microsoft has employees on the payroll who have no other function than creating annoying incompatibilities to encourage users onto the 365 treadmill.

I haven't kept up with the doc format story - are the ODF formats published, so LibreOffice can at least try to work with them? Or has MS closed them up again in the new versions?

I've been lucky enough to never have to actually develop ODF software for integration with Libre Office / MS Office, but from I've read, the specifications were half-heartedly published, and it is now basically a game where developers have to figure out all the exceptions to the interfaces that are within Office. Some work-around that might have worked yesterday might not work after an update of Office (which happens every few weeks).

Again, the cynical part of me assumes that MS employs full-time staff to ensure that Libre Office, Google Docs, and other apps that work with MS Office will never be even 95% compatible.

And from my experience, even simple spreadsheets and formatted Word docs don't display properly when opened in Google Docs or Libre Office, and large, complex business docs with macros and esoteric formatting don't have a chance.

There are a few developers in every place I've worked who use Libre Office because they're on Linux or refuse to pay for Office. And you can tell when they collaborate on a doc with MS Office users because it renders poorly when they send it back. It doesn't bother me, but I've seen how folks in management, marketing, etc. consider it unprofessional and assume the developer is just lazy or can't use Office well.

The one silver lining is that Linux users can use Office 365 online for less complex docs; it's not a full replacement for offline versions, but it's a better alternative than Libre for those of us who believe it better to be compatible with others even at the cost of a 365 subscription.

I got LibreOffice.

This can become problematic when an OS update breaks the perpetual fallback license software. As a developer, you can tell the customer that they have to upgrade the software (for a fee) because they upgraded their OS, but customers aren't happy about this either.

Isn't this basically what a lot of paid software with licenses used to do anyway? It's still how paid forum software and CMS plugins tend to work, with the rule being that anything released while your license is active you have access to forever and anything released after it expires you have to upgrade to get.

Why can't all subscriptions work like those old license schemes do?

It feels pointlessly complicated to me, the IDE version of picodollars. Their pricing is already a great value and I'd happily pay them a premium not to ever have to look at another weird 'perpetual fallback' Gantt chart.

About Ulysses specifically... they say when they were charging a monthly fee their users still expected regular updates after they paid.. and that sales would spike after a release, then fall to an unsustainable level: https://medium.com/building-ulysses/why-were-switching-ulyss...

"that price is vastly out of proportion from the cost of providing the software"

That's just an unfounded assumption by the author. He says he knows this, even though he doesn't know how many developers are working on it, what they get paid, how much time each release takes, how large the code base is, or how many users the product has. Essentially, he has made up a cost in his head, and decided that the costs must be less than his fantasy. For all we know, he's one of 2 users, and the cost for keeping his editor available is $100k/yr/user.

This also completely ignores the fact that prices for software aren't generally set using cost plus pricing.

well, that is economics 101. consumers price goods by value, not by costs. if your costs are higher than consumers are willing to pay, then you will at some point go bankrupt.

Exactly. No one cares what the product costs to make, they care what it costs from their own pocket. If the app isn't worth $40, then it doesn't have the features and/or the polish to have multiple full-time developers working on it. If the price people are willing to pay is lower than your cost of goods sold, lower your COGS or raise your value proposition. Those are the only options.

> If the app isn't worth $40

Ay, there's the rub. The problem is also related to price anchoring [1]: People [2] accepted the price of $70 for the perpetual value provided by the app (including cloud syncing), but balked at $40/year. The latter price makes the app more expensive over 2 years, and the developer promising updates doesn't have quite the impact to justify the change in the eyes of the angry customers.

If the app initially cost, say, $160, maybe the newer pricing would've been better accepted. I hear Adobe and Microsoft are doing quite well with the subscription models...

[1] https://en.wikipedia.org/wiki/Anchoring

[2] The people who bought it. This is important, because it ignores all the people who thought this wasn't good value. Maybe some of the initial non-purchasers became subscribers. Market analysis is hard.

On the other hand, there's a million different competing not-microsoft-word text editors for writers. Whereas there's no real alternatives to adobe or microsoft products if you need them for your career.

> if you need them for your career.

if you need 'em for your career, then make the person providing your career pay for it.

if you're self-employed, then use an open alternative.

So I guess you’ve never used one of those “open” “alternatives”? Often broken mess, lacking features, incoherent or antiquated UI, etc., ending up costing you more in time needed to ge things done than the paid alternative.

> No one cares what the product costs to make

The production cost is one of the metrics customers use to decide if a price is fair or not, to them. There are other metrics, like the time that can be saved by using the product (ie. some way to quantify the product "use value"), but the production cost is often quite useful to detect abuses--and boy, aren't there a lot of attempts of such abuses.

> production cost is one of the metrics customers use to decide if a price is fair or not

only approximately, and only if the consumer really knows the cost.

Think jewlery (like diamonds), or expensive brand name goods like designer clothing. The production cost for those goods is "low", but the charge is in marketing and status symbol representation (which is modified via marketing).

It could be argued that those two examples are good cases where the customer could be concerned about the value they get, even including the status factor. Diamonds are notoriously high-priced because of a quasi-monopoly, and high-end clothing is pretty often of poor value, quality-wise. If you like designer clothes--and you're not on unlimited budget--you eventually learn what designers/brands to avoid, because the value is just not there, even for that industry.

So this looks like a case in point, where having a rough ideas of the production cost (some parts hand-made, made in France/Italy/England, expansive material like quality cashmere etc.) is important.

> consumers price goods by value, not by costs

But the point is he's a consumer and he isn't pricing by value - he's pricing by costs.

I disagree, I saw it as him pricing by value, but mentioning that the costs don't seem to make sense either.

"consumers price goods by value, not by costs"

That is in itself a great argument for the subscription model. Consumes receive value from continued use over time, so why not fit revenue to the same shape?

I feel like Ulysses wouldn't have gotten nearly so many complaints if (a) they'd given some way cross-apply the license purchase price to a subscription, instead of expecting recent purchasers to shell out a premium price a second time to get updates, and (b) they'd actually had any substantial feature updates to show for it, instead of almost all attention since then going to just making the iOS app act more like the desktop app.

> some way cross-apply the license purchase price to a subscription

They did do this. From the footnote to to their original announcement[1]

Recent purchases of the Mac app will unlock up to 12 months of free use, and purchases of the iOS app will unlock up to 6 months of free use. Your individual free-use duration is calculated from your actual purchase date, respectively, and if you bought Ulysses on both platforms, we will add together both periods.

[1] https://ulyssesapp.com/blog/2017/08/ulysses-switches-to-subs...

The real innovation of the *aaS model is that you no longer need to compete with prior versions of your own software or hardware infrastructure. You can hike prices, cut engineering costs, cut some nines off your availability, defeature, or just plain embrace incompetence, and, in the short term, it will improve your balance sheet.

Of course, in the long term, customers will revolt, and do so swiftly, and without mercy.

I don't know. I think a lot of people have decided they like things like the Adobe products subscription.

Instead of the several thousand dollars you would have had to pay for the various softwares in the lineup up front, you just pay $50 a month. Which is reasonable of you previously used more than just Photoshop and maybe upgraded every 3 to 4 years.

If the end result is a similar cost, I would much rather keep my capital in the business and make a monthly payment.

There are some benefits in terms of product as well, if Adobe knows they have subscription revenue then they may be more focussed on fixing bugs and keeping existing customers than adding dubious features in an attempt to get upgrade revenue.

But 50€ a month is way too much for casual users.

I do some vector design work maybe once a month. I'd love to buy a license for Illustrator, if it cost whatever it used to cost before Creative Cloud. I would use it for a long time.

But 50€ per month is too much. So I bought a license for Affinity Designer instead. It covers most of my needs, and it's no subscription. I also bought a license for Sketch, since the subscription is optional.

I'd love to use the same tools as professional designers, but they've become too expensive.

They never became more expensive. They were extremely expensive to begin with and now they’re a lot cheaper. Previously it was like 500€ or so? That's too much for the casual user as well.

Here’s the reality. I used to pirate Adobe software because there was no way I was going to pay 1500k or whatever Photoshop was. Adobe got $0 from me.

Now I subscribe to get access to Photoshop. Adobe now gets £100 from me a year. I’m not the only person I know who went through this.

Of course, cheaper alternatives always existed. Adobe’s adoptance of subscription pricing hasn’t changed that.

I just checked the illustrator website. One year costs 287,77€. So it's only cheaper if you updated at least every two years. But I use my software a lot longer than that.

Software used to be an investment. At university, our physics department had a lot of old computers standing around, whose only purpose was to run some specific legacy software that was no longer maintained, but still worked perfectly.

During my time there, I convinced the theoretical physics department to buy a few Illustrator licenses so we could make better diagrams. I think the academic license cost 300€ per seat. They probably never updated, but I'd assume they still have a PC somewhere with one of those licenses installed, and if someone has a PDF from Inkscape with an incorrect Color Profile, they could still use that machine to fix it.

I think you underestimate how many people use old software out there. Sure, we tech folks update every year, but people who work outside of the tech bubble don't always upgrade a system that's working perfectly fine.

they are way way more expensive now. I only upgraded every other version so $199 every 4 years or $50 a year. current Photoshop subscriptions is $239 a year or nearly 5x the price

Honestly, the price I actually pay surprised me. I looked it up on my statement and I’m only paying £8.32/month (excluding VAT) for Photoshop.

Where are you getting your prices? I never remember being able to buy photoshop for $199. I remember it more like $600+. And the website shows $120 annually and that includes Photoshop AND lightroom AND 20GB of cloud storage.

If you need more than just photoshop, the deal is even better.

https://www.adobe.com/creativecloud/plans.html

It works for Adobe, ReSharper or MSDN because people make their livelihoods off it. It makes less sense when it's a utility (window management) or tangential to your actual revenue stream (keeping notes).

What? Am I doing my math wrong? Photoshop upgrades were always $199. Photoshop subscriptions are $239 a year. So upgrading every 3 year is $199 vs paying a subscription for 3 year is $747. How is paying nearly 3x keeping money in your business?

The $199 upgrades were only after the initial PS purchase price of $699. The PS-only subscription is $9.99 a month. The $239 a year one "Includes 100GB of cloud storage, your own portfolio website, premium fonts, and social media tools". It is way cheaper now unless you go 5+ years without upgrades.

I wouldn't put any value to those perks. The average use of that 100GB storage is likely to be a very small fraction of that and all will cost pennies or less for Adobe to provide. Yes, if you just so happen to be in the market for all of those it would cost you some amount to source them independently. But the trend of bundling low-cost tangential services as a pricing defence strikes me as rather cynical.

You skipped over the meat of his post. If you don't need those extras, you can get photoshop AND lightroom for $9.99 a month. So you are getting it cheaper than you ever could.

Releasing new features and improvements is highly incentivized in the *aaS model. It’s the main way of boosting ARPC, and as you mentioned, if you fail to do it, you’ll be very vulnerable to competition. It doesn’t alway suit every situation, but in those that it does, it benefits both consumers and merchants.

The best model I know is the one used by Sketch (https://sketchapp.com/pricing/):

  A year of updates.
  A lifetime of Sketch

  Your license comes with a year’s worth of updates.
  Once that year is up, you can keep using the last 
  version of Sketch you downloaded, forever.

* You can always open the files you have made

* There is no incentive for feature creep caused by having to "sell" the next version to your current userbase every so often.

* The developer does not have to keep supporting old versions forever.

IMO this is much better than what IntelliJ does.

From the developers perspective though it is way easier to run a business when your software is sold on a subscription basis. If you know to a decent level of accuracy how much income you have, you can easily work out how much you can afford to spend to keep running and growing the business.

If you have no idea how much income you are going to make, stuff become a hell of a lot more complicated, unless you have a substantial cash reserve available - not a luxury that everyone has. The directors spend a hell of a lot more time running around banks begging for short term financing, cutting costs and planning out 100 different potential scenarios.

If I had to run another business, I would definitely look to running some kind of subscription scheme..

I get this, but how much more predictable are subscriptions? I mean, unless you are holding customers hostage with data lock-in, they are going to come and go.

And if you get a reputation as a hostage holder, you will likely spawn competition that avoids your negatives.

I’m sure you will get some residual low hanging fruit in the form of people that buy subscriptions and don’t care enough to cancel.

Doesn’t seem like a winning business model though.

Indeed they are going to come and go.

But, generally this will be to a much more predictable level that if you are relying on upfront sales. Unless you really mess up your product, if someone subscribed last month, you can generally predict that there is a, say, 95% chance that they will still be subscribed next month.

If it is a one off payment, then you know you aren't getting any income from them and you just have to hope your sale people are going to do their job next month.

It has nothing to do with lock in. That's a bit of a red herring. You can have just the same (or more) chance of a repeat subscriber by creating a really good product as you will with some kind of hostage strategy. As you say, not a winning business model.

My experience here comes from more business to business where a months worth of expenses would be covered with 1-2 sales. I had times when the sales people would just disappear on extended unannounced holidays or decide to move to a different job without informing anyone. With our business strategy relying so heavily on upfront sales to maintain the cash flow, it was not a fun position to be in!

Things may well be different with consumer products where sales volumes are in the thousands of units.

I look at something like Overcast. Great software that I subscribe to and give him his money. Almost by definition, and through great pains from the developer, there’s almost zero lock in for a podcast client.

Yes but Marco said that only about 10% pay for subscriptions. That's why he created his own ad platform.

Subscriptions are MASSIVELY more predictable. I say this after working daily for years with FP&A team on building revenue projections for a product that had both subscription and non-subscription sales models.

You realise that your post sounds like the stereotype of every bad manager ever? "How much harder could it be to build the new feature if we did X, Y, Z? Surely you can get it done in just three days?"

I must say I actually love the subscription model that is becoming the de facto standard.

If you sell a one time license to somebody there is no incentive for the seller to keep making the product better, to care keep caring for you.

With a subscription model - especially a non-annually-paid one, they have to keep working to keep you.

It's also a way more sustainable business model and probably the only thing which can replace ads in practice. If a service like facebook gets a load of money out of you per year, there's no way they would switch to a one-time fee instead. If it were to be replaced by a subscription though, then you're still a constant money flow for them, which would work much better.

How many products need constant feature updates? Aside from security/compatibility there are plenty of products that already meet everything required of them (to near everyone). How many new features does a file manager need? An PDF reader?

For Facebook - I don't think forced subscription would work. People would be forced to think about how much they value the service (a less ubiquitous one at that).

> How many products need constant feature updates? Aside from security/compatibility there are plenty of products that already meet everything required of them (to near everyone). How many new features does a file manager need? An PDF reader?

Do you have any example of file managers or PDF readers with a subscription-based model?

Adobe Acrobat Pro, Foxit Phantom PDF. I'll concede I can't think of a file manager though.

But those are PDF editors, not PDF readers.

True. How many new features do they need? What about word processors? Photo editors? Email clients? There are new features that could be useful, but for many users their needs are already fulfilled. Why should they have to continue paying for Photoshop as a service if they're happy with the current version? I apologise for my poor examples.

> Why should they have to continue paying for Photoshop as a service if they're happy with the current version?

Ultimately - because the person who owns the rights for the software wants to keep charging them and since it's their software they decide what is charged for it, end of story.

If it's not a good deal for you, go somewhere else. There's no point arguing about prices beyond that. There's no 'should' in prices - only what people want to charge and what people want to pay.

Right, they pay more because the company wills it and people (I think) are more receptive to subscribitions. This is what the author of the article is complaining about and the reason consumer advocacy groups exist.

>can't think of a file manager though

http://dopus.com/ not exactly subscription, but you buy a major version, and each new major version is a new purchase.

Not only feature updates. Also everyday bug, security updates and also operating costs.

1. Operating system 2. Programming IDE 3. Password Manager 4. Well I pay subscription for Netflix and Spotify but that's not really for the updates. Partly for new content, partly operating costs. 5. Facebook (I'd prefer to cover the operating costs by subscription, I really dislike ads) 6. Email - same as above

Also, actually I'd like all the services I use to provide feature updates now and then and continue trying to innovate.

Edit: also, if you do feature updates every now and then. Like it was with Windows or Photoshop. Then you end up having to support a lot of parallel versions with security and bug fixes which is wasted work really.

The model works for you - great. There are users who don't care for constant feature updates and rather own a product than perpetually rent it.

I do understand that. I'm not saying there's one size fits all. I just don't agree that subscribtions are inherently bad.

But - if those preferring a perpetual license were to be a strong minority, there would be no business incentive to make products in such a way.

> 3. Password Manager

This one is a total non-starter for subscription pricing. I'm not willing to use software that will potentially cut access to all my accounts if I stop paying them monthly.

Check out 1password. Read access is retained, and you still have everything stored offline.

Most password managers retain read access. The subscription is to sync between devices and delegate shared access across teams.

Not too mention when the features just become some massive bloat and ruin the core features of the product. Chat applications are pretty much the worst offender for this kind of thing.

When you say "working to keep you" I think you're imagining the software becoming more valuable and useful to you, such that you are happy to keep paying them.

In reality, those companies who don't do scummy things like attempt to lock your data or similar, will be out competed by those who do.

> If you sell a one time license to somebody there is no incentive for the seller to keep making the product better, to care keep caring for you.

I strongly disagree on this one. If they want to sell you version 2.0 (upgrade), they better make sure those new features are worth their money.

With a subscription model, you either keep paying them or you lose it all. Why would they improve their product when you're already paying them x/month?

Because in this case their competition will catch up and I will switch. Exactly as I did with play music -> Spotify.

On the other hand, since we are on HackerNews, this is a HUGE business opportunity.

Some time ago I bought Affinity Designer because I don't want to pay Adobe for a subscription. And going back to that website, one of the first things they promote is "No subscription. Only €54.99".

Seems obvious that any big player that is switching to subscription only, is leaving a gap for smaller companies that support people like us.

There really isn't a better way to run a business (even if it's just a single app) than to get someone to pay you an amount on a fixed interval. It's predictable and generally weathers shocks pretty evenly. This is also why for most investors dividends are a big selling point - consistency.

Even Apple somewhat relies on this, except it's been in 2 year intervals. It's the point of so called "planned obsolescence."

So if you're an indie developer, you only have a few options.

1.Go through the rat trap of doing ads and trying to grow a huge userbase with marketing etc..., which everyone hates but most tolerate because it's free.

2. Charge a 1 time fee and hope that you price it right to support long term development and sustainment, knowing that you'll probably limit your market significantly.

3. Charge a recurring low(er) fee and draw a balance between losing a lot of very price sensitive people and having to run a marketing campaign and ad system.

No good option there but subscription seems to have the most benefits. By the way, everything needs upkeep.

Seems weird to call Apple out for planned obsolescence considering their devices are generally supported by the manufacturer for considerably longer than the alternatives.

It's not directly comparable, however the two year upgrade is what they really want people to do - after all it was pretty definitive that they were degrading experience as a trade off on the battery issue.

> the company switched to a $40 a year annual subscription, and as the dozens of angry reviews and comments illustrate, that price is vastly out of proportion from the cost of providing the software

As a software (SaaS) developer, I find this comment outright bizarre. I find it is hard to make a living with $50/month subscriptions, if you write niche software (and believe me, you DO write niche software, you are not Dropbox), much less support an entire company.

Have you actually done the numbers before writing that "price is vastly out of proportion"? Just take the total cost (roughly 2x salary) of several people, divide by that $40/year and see how many subscriptions you come up with. I'm getting ~25000 subscriptions for 5 employees, just to break even, and we haven't even touched the marketing required to get 25000 subscriptions.

> that price is vastly out of proportion from the cost of providing the software

Furthermore, only devs complain about the "proportion of providing the software". Didn't especially Indie devs here on HN take their time to learn the lessons of value-based pricing?

I do charge monthly for my app. It's 30 bucks a year for a B2C app. Expensive some might say, fair for others. But I can at least confidently tell that most users use my app every single day. And people who won't use it for more than a few months came away way cheaper than if the one-time-fee was 30-40 bucks to start with.

Oh and providing an app as a one-time-fee for $5 or some ridiculous undervalued price that is considered "acceptable" simply doesn't work in niches.

I think that most armchair-businesspeople underestimate the costs of running a business and overestimate the number of paying customers. It is particularly ironic, because they both do not want to pay for software themselves, and also think that masses of other people do want to pay for software.

People tend to think that the costs of a software business are just the time of the single developer that writes the software. And they also calculate the cost of that time as what developers actually get as their salary. But this is not how you run a business, not by a long shot.

This gap in understanding has been widening recently, because of the completely unsustainable app store pricing.

Have you considered that perhaps a fucking text editor doesn't need a team of five employees? This isn't a brain-surgery guidance program, it's a Markdown text editor that uses a sync system written and operated by somebody else.

Users are balking at the price because it's garbage to expect them to support some gigantic business for a simple thing. This is very common (1Password has over 75 employees!!!) and the sense of entitlement developers have to try to force their users to support a grandiose business, no matter how bloated and unnecessary the people are, isn't sustainable.

It seems to me that you haven't written an application of similar complexity. Also, Ulysses is not one app, it's a Mac app and two mobile apps, for iPhone and iPad, yes they are quite different. Have you actually used the app? Seen the customizable theming system? The export to PDF/Text/ePub/docx/HTML? The way you can rearrange and group sheets? The smart folder system? And most of that is present in the mobile apps, too.

You are underestimating the complexity of writing, maintaining, documenting, supporting and marketing apps of this size. I think my guess of ~5 people is about right, they might be able to do it with a smaller team, but probably not less than 3. I'd say 2 developers and one person to do marketing/support is a minimum.

True. A subscription to Office 365 comes down to 99 Euros per years. Of course you’d pay maybe 150 every three years or so for a new version back in the days, but still that’s double.

Also, I wonder whether this will lead to a slower development of the software. Before you’d have to create a leap forward to impress people to buy the new version, if the old version was still good enough.

I like that you mentioned Office 365 because I see that as "subscription done right." You can still buy perpetual licenses for the components of Office, you can easily store the data you created with those components locally, and if you stop paying for Office-as-a-subscription the same programs still let you view (but not edit) your data.

Carrot Weather App is another good one. Pay $5 once to get a really polished weather app with some basic "cloud" functions thrown in, like notifications to a phone device, some fun "achievements," and a pretty good weather data feed. If you want the super-accurate feed or if you want more advanced notifications and mobile device access (e.g. repeatedly on your Apple Watch), pay up for the subscription because it costs the dev more. And, if you stop paying, that $5 you initially paid is still valuable because the underlying app still works.

Jetbrains also has it done "right" I think... yearly license fee... and the version at the time of payment is "perpetual" while you get to continue getting updates as long as your license is "current".

Stop paying? When you aren't "current" you can either pay up... or downgrade to your perpetual version "owned".

Conversely, I find Carrot Weather ridiculous when Dark Sky is $4 one time and uses the exact same data source. They (DS) don't try to gouge you for Apple Watch features, which I interpret as "they're probably rich so why not?"

$4 a year (or $10, if you don't dismiss the 'ultrapremium club' popup and buy that by mistake) plus the $5 one time is not remotely what I would consider worth paying for a weather app unless Carrot was out in the wild building weather stations.

You'll also note that Carrot's website nor App Store description even mention the existence of this buy-up (except on the support page - but why would I look at the support page for an app I was thinking about buying?) or the pricing. They also deceptively show a photo of the Apple Watch showing weather, which is technically correct (the best kind of correct) - but the Apple Watch doesn't update in the background (or give you weather notifications) unless you pay for a subscription. Because weather apps that don't update are really useful.

Also, it doesn't "cost the dev more" to send updates to a watch instead of the phone (they're going through the phone anyway,) so I don't buy your justification for paying more for it. It's just a way to market segment and try to scam more money from people who have already paid above market for a weather app. You get Weather Underground data with the 'cheap' sub... but only on the phone. If you want it on the watch, you have to get the expensive sub. This also makes no sense. There's no way in hell wxug is charging extra based on the target device.

It is also surprising how much other subscriptions cost compared with Office 365. Office 365 contains the most complete office suite in existence, probably larger than many operating systems. It also includes 1TB of OneDrive, and free monthly Skype minutes. And the 99 Euros that you quote are for five persons.

Most subscription software is a terrible deal in comparison. I have tried to avoid subscription software as much as possible. But we have a family subscription for 1Password because $REASONS. We pay $60 per year for (in comparison to Office) a small software product, with small cloud storage, of which we only need a couple of megabytes for storing encrypted passwords.

Also, I wonder whether this will lead to a slower development of the software.

Most likely. Once you are locked into some proprietary data format, it is often more expensive to move to an alternative, and it is easier to subscribe. Once a substantial part of the user base has a subscription, there is no more need to entice users with a new version.

Subscription software does not benefit the user.

> Subscription software does not benefit the user.

Depends on software, subscription and benefits...

I think Office and Jetbrains are softwares done "right". They continually release updates - quarterly?... continually add new features...

I think more companies get it "wrong" than get it "right"... but done right, the company has reasons to keep developing - no new stuff? no continued monies. no new monies.

There was some low hanging fruit for companies to create subscription plans who shouldn’t have been adding them. The more subscriptions your target customer has, the harder it is to convert them on the new one. For the business side the ceiling is a lot higher, but for consumers there is a very low limit.

Each market is different, but when I look at an area like Photoshop or Illustrator, the competition keeps getting better. In some areas they aren’t even competing.

The workflow of a game artist will have them going through a bunch of different products to get to an end result. A lot of those steps can have alternatives. If a vendor has some proprietary format that doesn’t work with anything else they never even get a user base.

If a subscription is funding continuous development, and th company is doing a good job, and I use the software daily it doesn’t bother me. If I use it twice a year, I’m going to be looking hard at alternatives.

Its going to be a real problem for Adobe. Their biggest apps (PS/AI/IND) haven’t had real upgrade worthy updates in years. Competing apps like Affinity Photo are pretty much 1 to 1 replacements. Eventually some significant portion of Adobe’s core users are going to go elsewhere.

It's not even that - they basically are back to just the isolated professional market now as the majority of built-in apps do many of the enhancements that the cheaper subscription licensees would have sought out Adobe for. No one except aspiring professionals really needs the majority of Adobe projects to do competent looking amateur work anymore.

Subscriptions from Adobe were meant to soften the initial sale, but Adobe's own rigidity with the subscription schemes on release shot them in the foot, as organizations had to wait a long time for a model to come out that met the needs of small/medium businesses who previously could have made due with a few perpetual licenses. The storage was pretty sad even at the time of the switch, so that wasn't a compelling sales point either.

I think Adobe just didn't pull the numbers they were expecting when they switched, and now they're stuck with this model and infrastructure they don't know what to do with.

Maybe subscriptions will start going down, but Adobe's most recent quarterly earnings report [https://wwwimages2.adobe.com/content/dam/acom/en/investor-re...] says:

>Creative revenue in the first quarter of fiscal 2018 was $1.23 billion, up from $942.2 million in the first quarter of fiscal 2017 and representing 30% year-over-year growth.

So not only are they still selling, but subscriptions increased 30% over the past year. Clearly it's working for them so much that they're finding success doing the same thing for the document cloud:

>Document Cloud revenue in the first quarter of fiscal 2018 was $231.0 million, up from $195.9 million in the first quarter of fiscal 2017 as we continue to transition Document Cloud to a subscription-based model.

> Most subscription software is a terrible deal in comparison.

This is actually my way of making decision to buy a subscription-based product. It must provide comparable service. Office 365 single user license costs around 70$/year. Now:

- Evernote is asking 60$/year. Seriously? One note is already offering same service with 1TB storage.

- Pocket is asking 45$/year. WTF? It is just a bookmarking service which is already offered by one note web clipper.

- Proton mail is asking 50$/year for 5GB mail box. Dude, Office 365 offers 50GB add-free mail box. However NSA is able to read your emails, which is not the problem here. At least MS is not selling my information to advertisers.

Other than streaming services, It is getting really difficult to find any subscription-based product that gets close to Office 365.

> Also, I wonder whether this will lead to a slower development of the software. Before you’d have to create a leap forward to impress people to buy the new version, if the old version was still good enough.

I am not sure if that would be bad, though. Often enough, I feel that software vendors either just change the UI so they can show non-technical people how "improved" their new version is (MS Office), or they introduce arbitrary incompatibilities, so you need the most recent version to edit files that have been created/modified with that new version (I'm looking at you, Autodesk!).

With a subscription model, at least these vendors could focus on changes that actually provide value to their customers rather than making arbitrary changes just to make people buy a new version they might not really need.

That is a good point, regarding the stunting of the software industry.

Many economical models (including working 9 to 5, btw) came from the times of physical labor. At that time the part of labor was relatively small in the cost structure, and the part of materials was big. And since materials is a one-time expense, it was natural to have one-time pricing.

In the knowledge-based industry such as software the cost of labor has the biggest share in cost structure. And unlike materials, the labor itself is a subscription. Social policies that make firing hard make it even more so. No wonder that revenue structure also tends to shift towards subscription-based models.

A less charitable analysis is that the software industry has failed to deliver reliable software so spectacularly that it requires constant upkeep from a team of professionals to maintain a text editor like Ulysses.

Ulysses has a custom markdown feature that I thought was great. The only problem was that you couldn't use it without opting in to use their own file based storage schema. So if you wanted to store your writings in "regular" files and sync them yourself etc., you couldn't use the custom markdown feature.

I thought this was absolutely ridiculous since the custom markdown would have simply gotten written as plain text. I asked why they're limiting the user like that and got even more ridiculous response: apparently they didn't want other people to get confused in case I ever shared the files that were outside their own file based schema. This non sequitur infuriated me enough to ditch the software.

None of it made any sense until they announced the subscription model.

Ulysses is an egregious example as they cost more than MS Office in many cases, which is particularly noxious when you consider how little a markdown editor does in comparison.

I’m always annoyed about whiny iOS developers complaint about how they cannot make money without obnoxious schemes like this or gambling mechanisms. Software ISVs in olden times had a tiny market, huge expenses around retail channels and distributors, etc. They made money.

It's frustrating, but people wouldn't do it if it didn't work.

You will generally make far more money charging $5/mo forever than you will charging $50 once for identical software. Not to mention the steady cash flow.

An extreme example, but I always pirated Photoshop (as did probably half the people who ever used Photoshop). But now I pay $60/mo for it and a bunch of the other Creative Suite apps. I never would have paid whatever the 4-figure price was even when I was getting paid to make graphics at work. But I'll gladly pay $60/mo to have access to everything now even though I don't get paid for any graphics or video work.

edit: disregard my comment below, as others have reminded me, the yearly charge is for updates and will not stop you from using it completely if you decide not to renew it, you just don't get any more updates. So this isn't the best example

> I have noticed this as well, that a not-small number of software programs have turned from "give us one money for this version" to "give us slightly smaller amount of one money per month, but do it in perpetuity if you want to keep going with the data you've entered."

Like Sketch App which used to be just US$99[1] and is now US$99/year[2]. Although in this example, "slightly smaller" doesn't apply

1. https://web.archive.org/web/20160416212244/http://sketchapp....

2. https://www.sketchapp.com/pricing/

After that year, you get to keep using the product and your files though. You're just paying for updates.

But the website says:

Your license comes with a year’s worth of updates. Once that year is up, you can keep using the last version of Sketch you downloaded, forever.

That sounds ok to me.

What happens if there is a security hole in your version?

Air gap it.

I don't know about any software that is supported indefinitely. One day you need to upgrade, it's fair enough.

The same thing that used to happen to you before everything was a subscription--you will run insecure software.

Ah yes I completely forgot about that and you're right, that does sound ok to me

Look at our old software from the past decades. Old archived documents sitting in some obscure presentation program like Harvard Graphics. Now imagine not being able to load that data at all, even though you have the save file and the program, because the licensing server literally no longer exists.

Now imaging a world 15 years from now where Adobe totally goes under due to competitors making much better programs. Their servers go away and now, and you can't even start your licensed software without running it through a debugger and trying to crack the license check.

> Many founders choose SaaS simply because they want the recurring revenue. Just because you like the model as a founder, doesn’t mean it’s the right model for your customers!

(from https://justinjackson.ca/saas/)

> Sure, have a subscription service for something that needs upkeep, like a syncing service (but get out of here if all you do is just lump it into the user's iCloud storage) or a data feed.

This. If you sell an app and service together, you can charge a recurring fee. If your app is more traditional, then a one time fee for the app is appropriate. Note, this does not mean that you can just add a service willy nilly like and start charging a recurring fee. That's crap.

That's crap is not exactly a great reason for why one can or can't do something. I'm reading a lot of people with strong opinions, but I'd like to see the reasoning behind them.

A restaurant has some really popular plates, fancy looking things. Owner sees a way to generate revenue, so he sells the plates for $5 and every time you bring the plate back, you can refill it for $10 (which is a discount off normal prices). That's a pretty good deal for everyone, this is good.

Alternatively, same restaurant has some really popular plates, fancy looking things. Owner sees a way to generate revenue, so he sells the plates for $5 and a recurring fee of $1 per month. That's it, nothing else. Why would I want this? What's in it for me?

Why would I want this? What's in it for me?

Why must there be something for you? The thing in it for you is the same as before: the plates. If they're no longer worth it for you, that's fine, but that's the same if the owner had kept the same model but simply raised the prices or reduced the size of portions. I don't see why that's "crap" or "inappropriate".

The only exception I see is people who already bought a plate; I fully agree they should be allowed to keep using it as-is for free, since otherwise that's a breach of an existing agreement. But there's no such agreement between the owner and prospective buyers.

In scenario two, why should the plate get more expensive over time? It's the same exact plate. Nothing changes about it. I don't get a service where chips are fixed. I don't get a discount on food occasionally. Nothing. So, there's no _reason_ for the monthly fee, from the perspective of the purchaser.

You seem to be taking issue with my word choice, but I stand by that choice. The second scenario is a crappy deal. It's "Crap" for the consumer. I don't know anyone in their right mind who would go for deal number two willingly.

There is a difference between paying for tools that bring value, and discretionary spending.

If I am paying for software that I use to be more efficient and increases my profits, I am OK with a subscription model, or whatever model is offered. This is, as long as the cost of the software is considerably lower than the value bump I get from using it.

With discretionary spending, it takes a lot for me to buy a subscription. If I spend more than X% of a given month consuming what that subscription provides, I will consider buying it. Otherwise, I am fine not consuming that particular service at all.

I subscribe to a few magazines and Spotify. I read content online from many sources, but will not buy a subscription to each one of them. None of them are valuable enough for me to subscribe.

If there was a way to pay, say $20 per month, and have a set of credits that I could redeem at Washington Post, The Economist, New York Times, Bloomberg, etc, on a per-article basis (I make the decision after the first paragraph or two) I would jump at that right away. The system would need to be very user friendly and almost friction-free.

> If there was a way to pay, say $20 per month, and have a set of credits that I could redeem at Washington Post, The Economist, New York Times, Bloomberg, etc, on a per-article basis

i think this model doesn't work (but i'm willing to be pleasantly surprised). Imagine you read 10 articles a day, 30 days a month. That's 300 articles - that's about 6.6 cents an article, spread across many different organizations.

Those articles will need a readership in the 10,000's per article for any sort of return on investment with those sort of prices. Granted, a high quality article is going to be able to reach that, but there's limited time per person to read, and that 10 articles per day read is on the high end - most people read much less than that.

If there was a way to pay, say $20 per month, and have a set of credits that I could redeem at Washington Post, The Economist, New York Times, Bloomberg, etc.

blendle.com offers basically exactly what you're asking for. (Not in any way affiliated with the site, just an occasional user)

In the EU, the GDPR will make data takeout APIs (or some guy exporting as CSV and mailing it to you) mandatory.

Mandatory, yes, but the GDPR does not say it should be done for free. I offer data exports for all my services, so I'm compliant with the GDPR, but you will have to pay me for my time

Careful, there, it does say that in Article 12, (5):

"Information provided under Articles 13 and 14 and any communication and any actions taken under Articles 15 to 22 and 34 shall be provided free of charge".

(Article 15 is the Right of access by the data subject.)

You can only charge for subsequent copies or when "requests from a data subject are manifestly unfounded or excessive".

See also the ICO FAQ on the subject: https://ico.org.uk/for-organisations/guide-to-the-general-da...

This is not the case. (Article 12 (5)) https://gdpr-info.eu/art-12-gdpr/

> data takeout APIs

the data will be useless other than as a source of information. If you have data for an online vector editor, the vector data can be legally taken out via a common format (such as json), but there'll be no importers for this shape of data.

Any business can easily construe a common machine readable format that's also completely custom, and unimportable anywhere else, to satisfy the letter of the law.

If it's any sort of bigger service, competing services will implement their special format.

If they frequently change this format to break those implementations or clearly make it much more complex than it needs to be, that can also have legal repercussions.

That is a rather cynical take. More pragmatically, while it may be true that companies don't immediately jump at the opportunity to export in a common format (or there may be too many standards to export as), there is suddenly a market for data adapters if the companies do not import from their competitors, and a missing feature on a product sheet in any case.

I’ve been really stoked with Adobes offering around this, full access to all their products for $30 a month or so.

Except, if the only app you need is photoshop you can own Affiniy Photo for $30. Or, you can own Affinity Design for $30. And this one is the real kicker, Blackmagic is giving away most of what you need for free, and if you need more you can get a permanent license for a few hundred bucks. How long do you think it will be before Adobe starts really bleeding.

Adobe has had cheaper competitors for ever - basically the only move it’s competitors have is to be cheaper. Adobe still exists.

In the past Adobe’s competition wasn’t as good. Now, the competitors products are just as good. In the past PS competitors lacked features pros needed like PMS spot channels/CMYK/LAB. Not anymore. And Resolve is very compelling in the video realm.

Many of Adobe’s core products have languished at the same time as many of the features that set them apart in the past are become less differentiating. The photo editing/vector drawing space is becoming more commoditized.

Which is the "just as good" alternative to Lightroom? I've found its tools to be astonishingly high quality. Adjustments have exactly the effect you expect them to, instead of something closely adjacent to what you expect. The spot remover is basically magical. RAW handling is top notch.

I haven't seen anything at all that comes close on a cheaper model.

I should also add, Quark was the king until it wasn’t. Adobe will be the king until they aren’t.

I’ll have to checkout black magic.

Frankly, the writing has been on the wall for something like at least 10 years and surely came into life once Adobe announced the CS 5.5 subscription model in 2011.

https://opensource.com/article/17/12/open-source-business-mo...

The difference is you used to pay for a new version. Software 2 would be charged for another $50. Whereas now you pay for Software and it gets improved in perpetuity. It should be charged as such.

In general, software has zero marginal cost, so in theory, it's price should fall near zero due to competition. So why is the subscription market behave so different ?

And what can be done about that ?

It's almost like the industry is not perfectly competitive...

Even with zero marginal cost, you can still have substantial fixed costs. And actually costs don't define prices — what the market will bear defines prices.

> what the market will bear defines prices.

only if there's no competition, or there are substantial barriers to entry (such as regulation).

Well, one could argue that the price is near zero. It is very rare software product that does not have somehow usable open source competitor. Then what you are paying for is something else than the zero marginal cost basic functionality. Like support, branding, network effect rents or something else.

Basically user's time invested. They know how to use the current software "I NEED Photoshop to do my job!!!".

Even if some other software has the same or even slightly better abilities at a lower cost, the overall cost of switching is much greater. Especially if someone else is paying for the tools.

Until some software appear that make user's competition more efficient and they are forced to learn new ways to not get out competed.

Then the cycle starts all over again.

I was looking at a piece of hardware, that had some software that runs with it. Same deal. You buy the hardware but if you want to use the app.... it's a subscription.

Price gouging will create opportunities for new entrants.

One perfect example of a shitty conversion to subscription is the adobe suite.

One-time payments for software amount to a pyramid scheme where you benefit from the money of future users. That's not without its downsides either, where your userbase grows indefinitely yet doesn't pay for its own support-email rent.

There's obviously a middle ground, like being able to at least use the version you paid for indefinitely but having to pay a smaller price to upgrade. But which direction a pricing structure goes also depends a lot on the niche and type of software.

Citing the worst examples of "subscription abuse" isn't really an argument either. Of course you can find examples on each end of a spectrum. But as far as most software pricing goes, if it's so bad, then people wouldn't pay.

There's a difference between charging for updates and charging for continued use. What if Sublime Text started charging a monthly fee just to use it? There's no excuse.

> But as far as most software pricing goes, if it's so bad, then people wouldn't pay.

And yet people keep paying ransomware for access to their important files.

Like Webstorm?

I do feel that I get value for my money and with the yearly fee I get all the updates (which are important in the ever changing world of web development), but sometimes it still feels like protection racket.

no, not like webstorm. Webstorm is pretty similar to sublime text. You pay a price for a perpetual license. After one year of use, you can decide whether you keep on using the 'old' version, or if you pay again to have the newest one with all the updates. I think the comment you were referring to was thinking about stuff like adobe, where if you don't pay, you don't get anything.

Indeed, you're right. Don't Adobe give you some "cloud storage" thing though? Fortunately now that we're using Figma I don't have to deal with the hundred pound gorilla called Photoshop.

Jetbrains does not require subscriptions for continued use of any version. Purchasing a 12-month subscription (which is cheaper than their previous "perpetual license") gives you a perpetual license for the current version of software.

This is actually my favourite licensing technique I have come across for software.

https://sales.jetbrains.com/hc/en-gb/articles/207240845-What...

There’s a difference though. Previously the version you ended up with was the one at the end of the year of updates. In the current model you have to reinstall the old version of the time of purchase, undoing the updates you got. It’s actually quite clever since you immediately notice what you’re missing by not paying.

I could have sworn they offer specials or something where you can get the current and next version.. can't recall, I end up upgrading each year to keep my discount going.

You can actually keep your next current version by buying for a few months to extend to the next 12 month anniversary. See last timeline example on this page:

https://sales.jetbrains.com/hc/en-gb/articles/207240845-What...

Sorry but nobody is forcing you to use sublime.

If they started charging just for use they wouldn't need neither an excuse nor a justification.

I really don't see your point.

If you could explain further why you think sublime would need an excuse for charging a monthly fee I would really appreciate.

Yeah, I don't see the reason besides "I don't like it" either.

If the problem is access to files, then they could just have a free converter tool to some standard format.

>If they started charging just for use they wouldn't need neither an excuse nor a justification.

Actually they'd need both.

And companies are not so dumb as not to understand this. Any company that started charging or went into a subscription model, offered both excuses and justifications for doing so.

You might be think "it's a free country, it's their product, and they can do anything with it".

Which is true in a trivial way (they're allowed by the law to do anything they like in that area), but it's also wrong.

Wrong in that in practice they're not really free: to survive they are bound to the market and their customers responses.

So they're free only as much as they're OK with pissing their user base and potentially losing sales (or have most of them go for some other product, and kill their own user base -- e.g. Quark xPress vs Adobe Indesign, and tons of other similar examples).

If the sublime folk offered a “bug fix only” monthly subscription, I’d buy that. My grandfather went through a half day’s paycheck of tooling a month on his own dime. I figure bug fixes are the software equivalent of sharpening chisels, buying new bits, surfacing a mill, etc.

> If the sublime folk offered a “bug fix only” monthly subscription, I’d buy that. My grandfather went through a half day’s paycheck of tooling a month on his own dime. I figure bug fixes are the software equivalent of sharpening chisels, buying new bits, surfacing a mill, etc.

Except bits don't rot?

New OS versions (especially from Apple) regularly create new bugs or completely break compatibility with old software. The cost of upkeep is not entirely predictable and easy to "price-in" to the upfront cost.

Can't speak for Apple I guess, but this isn't exactly the normal case on Windows.

Microsoft jumps through hoops for backwards compatibility, often to their own detriment. Apple - comparatively, not absolutely - almost ignores backwards compatibility...also often to their own detriment.

They don't rot but they get rusty.

The middle ground doesn't work. See MS Windows update hell.

The way things should work is like this: Subscription for when you want to use the software and open format data files that you're always easily able to export and download.

If a company goes out of business or does something stupid then another one will pop up to handle your needs. This aligns incentives correctly while still allows a healthy dose of competition.

Actually, it works in software that many of us use every day. Once again, there's more nuance to making an argument than being able to point out a single absolute example. Why is this such popular rhetoric on HN? Especially in such a varied imprecise world of pricing structures?

I use various bits of successful software like Transmit and Alfred that let you update until the next major version in which you must pay again.

It "works" in the way that having insecure software on your computer "works" which is to say that it doesn't work, we just put up with computers where the non-technical don't know the mess they are in. Every major operating system (iOS, Android, Windows, QNX) or widely used software application (MS Office, any browser, any widely used app) issues emergency security patches.

The reality is that having everyone on the same version drastically simplifies not just security, but also interoperability. It isn't 1998 anymore. The default is internet connectivity and we should stop pretending our apps are secure when they are not.

There is a kind of perverse incentive not to work too hard to secure software when doing so removes a critical reason to buy the updates.

That being said I do think truly securing an application requires first having a secure platform. Otherwise users could indefinitely demand mitigations at the application layer until it contains most of its own OS.

I don't need a secure MSpaint.

Now you do you, and I'll use Goldwave v4.0 for basic audio edits - exactly the same tool I used in 2000, one that works well and I never had to re-learn.

I despise tools that capriciously change in my hands. Extra "security" is not worth the productivity hit.

Again, Internet connectivity isn't my default. Dropbox can handle that, and that software gets all the updates. The tools I use to make stuff I put in Dropbox better all be offline and never changing unless I say so.

That applies to the office suite, DAW, graphics editor, etc. I haven't seen anything their online features could give except a cloud lock-in.

You may value extra security, often necessitated by the very virtue of the software being online. And I value the ability to use without having to think about whether the signal is strong and my card payments went through.

The reality is that the interests of the users are often not aligned with the interests of software vendors.

> The middle ground doesn't work

The "middle ground" has been the most common model for desktop app pricing for decades. You buy version X of a product. When the next major version is released you pay an upgrade price.

In fact, the "middle ground" has been the norm for the majority of software apps over the history of desktop computing. It's the SaaS subscription model for cloud-hosted software that is the new deviation. But that cloud-based subscription model is now leaking into desktop apps (like the Adobe suite).

> The "middle ground" has been the most common model for desktop app pricing for decades. You buy version X of a product. When the next major version is released you pay an upgrade price.

Mobile app stores killed that model, when they made continuous, perpetual updates the norm. At the time, everybody was too busy pointing out the implications of going from 50$ to 50ct when the even bigger change was a farewell to major version upgrade sales.

But there, after a few cycles of companies ruining themselves by hiring in the short-lived money surge of growing an app to saturation, parts of the industry seem to have found their way to quite reasonably priced subscriptions.

>Mobile app stores killed that model, when they made continuous, perpetual updates the norm.

Continuous, perpetual updates are definitely NOT the norm in mobile apps.

Some get some bug-fixes and the occasional new feature, but any major update lands a version 2.0.

Far more (the "long tail") are just abandoned (like all those thousands of 32-bit apps being deprecated in the iOS store).

Both apple and google don't support 'pay for upgrade' model. Its either subscription or pay once. Of course, you can roll out your own payment gateway(which apple has further issues with) or re upload the app under a new v2 name, but those are not very satisfactory solutions.

There is a way to provide upgrade pricing for iOS. It's an ugly hack though....

2015 - you sold v1 for $5.00

2017 - you start selling v2.

You sell v1 for $5.00 but in the description tell everyone not to buy it.

You sell v2 for $5.00 but direct people who have already bought v1 to buy the bundle of v1 and v2 that you price at $8. Anyone who has already bought v1 can buy the bundle for $3.

and now, you have v3. The bundling will grow exponentially!

It’s quadratic, not exponential. For N versions, you have prices:

  N * (N + 1) / 2

It is likely that you could reduce to something approaching an upper bound of 2*N, as with version 8 out, you can drop support for a discounted upgrade from v2 to v4.

I stand corrected!

>or re upload the app under a new v2 name, but those are not very satisfactory solutions.

Not very satisfactory, but very prevalent.

And, I'd say, given the low price of apps, it might be just right to not offer any special discount for an upgrade. You could always reward older users some other way.

It was easier to price software like a physical good when the distribution method was identical to other physical goods. You'd go to the store and give them $50 to leave with a box that contained a disk. When you wanted vN+1, you'd repeat this process. It all made sense to the user. That model is collapsing as digital distribution takes over.

Also, Adobe's tools have always been among the most pirated software out there. People who needed to do occasional work wanted Photoshop but there was just no way they were going to drop $700 on it. Nowadays, Adobe just charges them an $11/mo fee. That makes it much easier and I know several instances where this newer model has resulted in real income for Adobe, from what would've otherwise been pirate installs.

People are much more willing to pay a small fee several times than a massive fee once. The subscription makes them feel like they have more control and they can cancel when they don't need the product anymore (not always true because some of Adobe's subscriptions have annual lock-ins). It reduces the friction from "figure out how to pay us more than your car payment" to "just send us the cost of a trip to Starbucks, it will only take a couple of minutes and you'll be merrily on your way".

This dramatically changes the incentives re: piracy. Instead of spending hours trying to find a cracked build from a site that won't pwn your PC so that you can save hundreds of dollars, you just pay a few bucks and speed through with a known-good version (plus a clean conscience).

It's almost a side benefit that subscriptions frequently represent additional lifetime value v. a "pay once" model.

An alternative is intentionally packaging releases into tiers (performance, stability, feature, etc) and sell them as such. Something closer to Windows XP service packs than individual updates. I'd expect small, necessary updates to continue to exist as they do.

A user can buy the base version, then a few months later when you create a release that allows processing widgets through the program twice as fast, they can update for a small fee. This would force the developer to create updates that would be easy to communicate the benefits of to the consumer.

It'd be a bit more difficult to figure out pricing for new users over time and which version they receive, but it would allow they old pricing model without forgoing the benefits of incremental development over time.

> An alternative is intentionally packaging releases into tiers (performance, stability, feature, etc) and sell them as such.

That could quickly devolve into supporting a large number of distinct versions of the software, which would be difficult and costly. Nobody wants to do that.

Not to mention that, if this versioned software is a platform (for example, WinXP and its service packs you mention), downstream vendors would also need to QA and support their software on all actively supported versions of said platform.

So, basically, DLC.

This is possibly the worst use of the term "pyramid scheme" I have ever seen used in print, and that's really saying something with how often it's abused nowadays.

Yet you left your entire argument up to our imagination.

I explained why I find the description fitting. You could at least entertain us with why "future users pay for your upgrades" does not fit in the context of one-time-paid "perpetual license" software beyond disagreeing with the phrase I chose to name the predicament.

It seems to me that there's one very obvious difference: once software is created, creating more copies of it is almost free. The reason pyramid schemes collapse is that they require an exponentially increasing amount of new deposits to keep up with the payments owed to existing depositors, and that's just not how software publishing works. So long as there's a sufficient stream of customers buying, it doesn't cost much more to keep everyone up to date than it does just the new customers, and if there isn't then forcing repurchases won't necessarily fix that.

But with the way that App Store economics work, you can't easily charge for upgrades but both iOS and Android have new versions yearly with new features that users expect you to implement.

Which is why most companies do a mixed model. Pay an amount for a version, get one year of updates (or that specific version + it's updates), then pay to upgrade (and/or reup your year of support).

There's a flaw in your logic: You assume that everyone keeps using every app they bought forever. If that was the case, pay-once would be unsustainable.

However, that is not the case. Everyone stops using your app at some point. So you don't need to support users forever, you just need to support them for some finite amount of time. Some users might use your app for a few years, some use your app just for a few weeks. Some users will email you half a dozen times, but most will never email you. Your app price should be priced so you can afford to support the average user.

And another thing that people forget: When your user base is growing, the amount of word of mouth recommendation you get also grows, and you automatically start selling more licenses, so your revenue also grows.

Pay-once and subscription aren't really that different in practice. Pay-once even has an advantage: You get all the money up front. And that's awesome if you are planning to grow your company.

It's not a flaw. That's just uncertainty and a gamble. Whether that's sustainable for a given piece of software varies per product. And that's kind of the point I'm making.

The extent at which some users subsidize the app for others is just one of the decisions you need to make when deciding on a pricing structure, not a rebuttal.

> There's a flaw in your logic: You assume that everyone keeps using every app they bought forever. If that was the case, pay-once would be unsustainable.

There's also a flaw in this logic: It assumes that all apps require some kind of server provided by the developer in order to run.

Once upon a time, standalone software was standalone and could run in perpetuity without requiring anything more of the original developer. It's still entirely possible to write software like this, there's just a trend away from it due to the prevalence of always-on internet connections and the ease of keeping some of your application server-side as a form of copy protection.

Until your standalone software answers customer support requests automatically, you‘re still going to have variable costs that grow with your user base.

It's trivial to only supply customer support for a limited time after sale of the software, or to offer a separate support contract for customers who require support.

“Hi, I lost my license key!”

“Thanks for contacting us! Our standard license includes support for 12 months after purchase. I’ve attached an offer for our extended support contract. I’m looking forward to your response!”

Atlassian has such a pricing scheme. Pay x amount per year for the cloud version, x-x/10 for the self hosted license and half that amount yearly for updates and continued support. Of course you can still use the license without updates and support. The <10 user license is almost always $10 so that the entry bar is set low, other licenses are much more expensive (>100x) with price going down with numbers until you hit the datacenter versions which are priced differently. I think it's very flexible and covers just about every use case.

> Of course you can find examples on each end of a spectrum.

A good example is the Adobe Creative Suite 6 Master Collection which was so expensive to buy in Australia that it was cheaper to fly to the US, buy a copy and fly back. Now, you can get a years subscription to Creative Cloud for 'All Apps' for about 1/5th of the price of what CS6 MC was in Australia. It is still expensive, and after 4 or 5 years you will have spent more but thats significantly longer than Office 365 which costs about 2/3rd the price per year of a copy of Office 2016 outright.

> One-time payments for software amount to a pyramid scheme where you benefit from the money of future users. That's not without its downsides either, where your userbase grows indefinitely yet doesn't pay for its own support-email rent.

This logic applies to literally any one-time purchase that includes a warranty, yet the consumer goods sector has managed it somehow. What's really happening is a deterioration of the idea of "owning[1]" software. I'm sure it's great for the publisher, but it's almost all downside for the user. I'm still using an email app that was acquired and shut down by Google in 2012 and it continues to work just fine. On PC I use WinRAR from 10 years ago with no issue. I can't imagine having to pay for all of these little apps on a subscription basis and then having to find alternatives when the creators get acquired/go bankrupt because the subscription is no longer supported.

Edit: To add to that, if the cost of the subscription is to provide me with superior support, will publishers selling me these subscriptions be removing the warranty disclaimer from their EULA? Probably not, right?

[1] Yes, you never "owned" it, but licensed a copy - but in practice it was similar to owning a physical good.

You're exposed to pretty serious security vulnerabilities if you're still using a 10-year-old version of WinRAR to unrar/unzip/uncompress files you get from the internet.

Did you know that every physical good sold is a pyramid scheme using your definition?

In the US, companies have product liability to correct defects for the lifetime of the product, subject to routine limitations. [1]

So companies create business models to compensate for this limitation.

Companies figured this out with software decades ago, so it’s certainly possible (eg, Microsoft).

[1] https://injury.findlaw.com/product-liability/what-is-product...

IMO except for highly priced products, paying once for software does not entitle you to support unless it's the user forum/SO type. John Doe giving a company $50 three years ago does not give him the right to email people at the company asking for tech support.

Many many companies still allow this, but I think if you want support you should be paying a support fee (e.g. monthly subscription).

> that price is vastly out of proportion from the cost of providing the software

This is not how prices are set. They are set on the willingness and ability to pay. Water costs nothing to produce but if I have it all you will pay me everything you have.

For example if their main competitor went bust and not a single atom changed inside this company, they could charge you more.

Do they have a monopoly or is it hard to migrate? These are the conversations the people who set the price are having. They are not talking about people being upset if they still pay.

Reply on news.ycombinator.com