Airmail Users Frustrated About Sudden Switch to Subscription-Based Pricing on iPhone and iPad [Updated]

[grmlin]

I bought many premium apps both for iOS and macOS in the past, no regrets. Things like magnet, paste or Apollo (on iOS) are just too good.
I also use 1Password, but I'll look into alternatives next year when the subscription has to be renewed... pretty sure there are decent alternatives to get. I just missed it in June.

 

[MacBH928]

I love that guy who is complaining that he paid 5 Euros a year ago...

I am not a developer. but by the power of the internet if you can make 200K users pay $10 for your app, thats $2M. Not sure if its enough income for the developer or not, and thats just for 1 of the apps he develops. Some make 2-5 apps.

I am more ok with subscription with more complex bigger apps like Office, but something like a mail viewer and a podcast player...meh.

bought full versions of games that automatically updated to needing in-app
purchases and lost what I had previously paid for.

If you want recurring payment, that's fine; do it from the start or grandfather existing users.

That should be illegal

 

[1144557]

I love that guy who is complaining that he paid 5 Euros a year ago...

It's the point that the developer set a price (not the consumer) and consumer paid it for the app. Maybe the developer should have planned better and charged 10? THEY set the price, that is their problem.

But now developer says, well we know you paid for the app but were taking away this basic app function (push notifications and multiple accounts) unless you pay us monthly now suddenly. Even though you already paid for the app with all of the features.


Going back, it would again be like buying a car and 3-4 years later the car manufacturer deciding you know what you cant use your key fob anymore unless you pay us $10/month, even though you bought the car with a key fob in the price already that is the standard in cars today to unlock and lock it remotely and used it for a few years.

Its completely wrong and likely illegal.

 

[grmlin]

I don't care how much I paid for the app, all that matters is that I did.

Just imagine this becomes an accepted business model. It shouldn't be allowed to transform a premium/paid app into a free2play/use one. Make it a new app and abandon the old one. But don't take it away from the customers who paid for it.

 

[illmatic41]

Going back, it would again be like buying a car and 3-4 years later the car manufacturer deciding you know what you cant use your key fob anymore unless you pay us $10/month, even though you bought the car with a key fob in the price already that is the standard in cars today to unlock and lock it remotely and used it for a few years.

While completely different from the dickmove that Bloop did for Airmail, your analogy reminded me of my same thoughts regarding Tesla removing the long range function of cars they sold that at the standard range price. Granted the owners were not paying for these upgraded functions, I think it's interesting how Tesla can remove range, heated seats, etc. with software and then ask the owner to pay for the upgrade later. Vehicles becoming gimped software with a trial period...what strange times we live in.

 

[Eric8199]

Developers need to eat. You can help feed them by paying for products and services.

Completely agree. However, I already paid them. I don't buy milk at the grocery store and then have the price go up the next day and have the farmer show up at my door telling me I need to pay them more money if I want to continue using the milk.

This would be like buying a car and three months later the windshield wipers don't work anymore, and you call the dealership and they say, yeah, we decided to charge you a subscription fee for that. I really enjoyed AirMail, but I won't be enjoying it anymore. Hope Apple tells them they can't do that and makes them get rid.

 

[Wildkraut]

And -1; Adobe, MsOffice, Ulyssis, 1Password and ScanBot please welcome Airmail yeahhhhhh

 

[JRobinsonJr]

I think that subscriptions are the best thing what happened for developers and the worst thing for customers. I hate subscriptions too, at least in cases where do you see developers only milking their customers. I have no problem with content subscriptions like Spotify / Apple Music etc. But paying monthly for an email app? No thanks!

While there are valid arguments for both sides, as a customer I'm OK with subscription software. The reality is that my needs change over time, and sometimes that means switching to a different product. With a subscription, I get to pay a reduced cost every month/year... then if I still like it and use it I'm OK with paying that reduced cost again.

Buying a perpetual license doesn't solve the main argument in this thread. It gives you a license to a particular version of software. Step outside the App Store and consider this. Buy Microsoft Office today. In 18 months they have a major version release. Your original purchase may or may not cover the upgrade. If it does... great! If not, you probably have to pay an upgrade fee. If not... you pay full price.

From my perspective, it's not a question of "are we going to pay" but "how much and when".

 

[Eric8199]

While there are valid arguments for both sides, as a customer I'm OK with subscription software. The reality is that my needs change over time, and sometimes that means switching to a different product. With a subscription, I get to pay a reduced cost every month/year... then if I still like it and use it I'm OK with paying that reduced cost again.

Buying a perpetual license doesn't solve the main argument in this thread. It gives you a license to a particular version of software. Step outside the App Store and consider this. Buy Microsoft Office today. In 18 months they have a major version release. Your original purchase may or may not cover the upgrade. If it does... great! If not, you probably have to pay an upgrade fee. If not... you pay full price.

From my perspective, it's not a question of "are we going to pay" but "how much and when".

I agree. I have no issue with subscriptions. But not for something I already paid for. Had AirMail released a new version of the app, hell, even if the updated version of the app didn't look any different, but my app still functioned as it had when I PAID for it, I'd be fine. Instead I got zero warning, suddenly my app is updated and I no longer have a feature I paid for already. That's BAD BUSINESS. If they would have even given us advance warning this was going to happen, I probably would have said eh, I like this app, I'm willing to pay a little extra. But the way they went about it, I will NOT pay them a dime in subscription fees.

 

[dsusanj]

The developer does not seem to understand the criticism around the lack of communication in this case, sadly...

https://www.reddit.com/r/Airmailreddit/comments/cdv5f0/app_doesnt_work_with_subscription/

 

[conshok26]

I agree. I have no issue with subscriptions. But not for something I already paid for. Had AirMail released a new version of the app, hell, even if the updated version of the app didn't look any different, but my app still functioned as it had when I PAID for it, I'd be fine. Instead I got zero warning, suddenly my app is updated and I no longer have a feature I paid for already. That's BAD BUSINESS. If they would have even given us advance warning this was going to happen, I probably would have said eh, I like this app, I'm willing to pay a little extra. But the way they went about it, I will NOT pay them a dime in subscription fees.

I would agree with this approach. I've bought multiple versions of Reeder and Tweetbot. In these cases they don't strip functionality from their app (I'm pretty sure) they just add new functionality and offer a new app. If you want this great, you buy the app. If you're good with the way things are they don't take anything away from you.

 

[mfm77]

This was a colossal screwup, and the worst part is the developers are shrugging their shoulders and saying, “Costs increased, oh well” to their customers. No apology, no “We want to win you back, and here’s how we’ll do it” - nada. Maybe they banked on losing a huge portion of customers while getting enough people to buy the app to further their goals. That won’t last, though - not if they don’t zap the myriad bugs in the app.

 

[mnylen]

You can send push notifications from a DigitalOcean server that costs $5/mo; it just involves sending an authenticated network request with a notification payload to APNS. As long as they’re convincing at least 2 people per month to buy the app at the old one-time price of $4.99, then they’ve made their money back on notifications with a couple bucks to spare.

Of course, that’s an overly simplistic view. But especially at their scale, push notifications should be a relatively negligible expense.

It isn't quite as cheap as you might think it is. If you have 10,000,000 email accounts you need to check every 10 minutes with each check taking on average 1 second, you end up doing 44,640,000,000 checks monthly and as many seconds of time. Likely the interval is shorter than 10 minutes too.

Because the checking is not compute intensive (it's mostly waiting on Gmail's API) let's say you decide to get AWS EC2 t3.medium instances with 2 cores each at $0.0416/hour. Let's say you can run 8 "checker threads" per core as mostly they spend their time waiting on API calls and not doing any computing.

You end up needing 1042 EC2 instances for running the checking at 24/7. This will cost you $31,200 per month and only includes the cost of _checking for mail_. It doesn't include for example the cost of databases (where the account information is stored), the components that schedule the checking to happen once every 10 minutes for each account, storage to store the checker application and the logs, any automated monitoring (you really don't want to manually go through 1000 instances when something goes awry), data transfer etc. Plus you would often have some extra capacity provisioned in case there's any slowdowns with the external APIs.

Sure, you can get by with $5/mo DigitalOcean machine for your personal use. But in scale, you need a lot more.

 

[Khedron]

This was a colossal screwup, and the worst part is the developers are shrugging their shoulders and saying, “Costs increased, oh well” to their customers. No apology, no “We want to win you back, and here’s how we’ll do it” - nada. Maybe they banked on losing a huge portion of customers while getting enough people to buy the app to further their goals. That won’t last, though - not if they don’t zap the myriad bugs in the app.

The customers they're losing are the ones who already paid the single cost. The customers they gain are on subscriptions. Fewer customers to support and the ones they do pay more. Win-win, apparently: "...love to focus on smaller but loyal userbase"

https://www.reddit.com/r/Airmailred...ion/etx1uqc?utm_source=share&utm_medium=web2x

 

[ginkobiloba]

Modern apps requires cloud infrastructure to operate and there are costs to that. You can’t expect perpetual support for a piece of software under $5. That being said, they should have grandfathered in users who already bought the app.

I'm perfectly ok paying for an upgrade ( or even full price if it's not expensive ) for new versions. I have some pro audio and video software that costs over 200$ for upgrades to new versions. I just don't like being hold hostage by software that has no jusfication for monthly rent.

 

[auxbuss]

It isn't quite as cheap as you might think it is. If you have 10,000,000 email accounts you need to check every 10 minutes with each check taking on average 1 second, you end up doing 44,640,000,000 checks monthly and as many seconds of time. Likely the interval is shorter than 10 minutes too.

Because the checking is not compute intensive (it's mostly waiting on Gmail's API) let's say you decide to get AWS EC2 t3.medium instances with 2 cores each at $0.0416/hour. Let's say you can run 8 "checker threads" per core as mostly they spend their time waiting on API calls and not doing any computing.

You end up needing 1042 EC2 instances for running the checking at 24/7. This will cost you $31,200 per month and only includes the cost of _checking for mail_. It doesn't include for example the cost of databases (where the account information is stored), the components that schedule the checking to happen once every 10 minutes for each account, storage to store the checker application and the logs, any automated monitoring (you really don't want to manually go through 1000 instances when something goes awry), data transfer etc. Plus you would often have some extra capacity provisioned in case there's any slowdowns with the external APIs.

Sure, you can get by with $5/mo DigitalOcean machine for your personal use. But in scale, you need a lot more.

This makes some weighty assumptions.

First, I think we can safely say that Airmail doesn't have 10M accounts to check. Most users are going to have one account, far fewer with two, and so on. (fwiw, I have three; but I know I'm an edge case.) But let's go with it.

Next, not every account will have push activated, nor will each have push available.

Next, if you run OS threads per processor, then you're doing it wrong. For example, even if we take the 10M accounts as correct, then you can spin up that many erlang processes simultaneously on a single 32Gb instance of some VM. (erlang processes are approximately 8Gb minimum per 3M (on a 64-bit system).) A dedicated CPU 32Gb Linode runs 120 USD per month.

Now, there's more to it than that, of course, including a database (as you mention, although scheduling is simple per process) though there are many ways to limit the checking. However, this is not something that requires large scale infrastructure along with its associated costs.

At bottom, you're just pinging endpoints and sending notifications based on results. Trivial stuff, even at massive scale, which this isn't (by today's standards).

 

[auxbuss]

For my part, I spent the afternoon migrating two macOS and two iOS with three email accounts back to Apple's Mail.

That's wasted time, far more costly that the software.

This is my second experience of Apple's enabling of bait and switch. The other was Ulysses.

While I hold both developers in absolute contempt – equivalent to thieves or, more accurately, pickpockets – I also blame Apple for enabling this practice. Apple controls the market absolutely. None of this would happen if their rules precluded this obnoxious behaviour. Where are they?

Henceforth, I will not buy anything from the MAS until Apple blocks this practice.

 

[Khedron]

This makes some weighty assumptions.

First, I think we can safely say that Airmail doesn't have 10M accounts to check. Most users are going to have one account, far fewer with two, and so on. (fwiw, I have three; but I know I'm an edge case.) But let's go with it.

Next, not every account will have push activated, nor will each have push available.

Next, if you run OS threads per processor, then you're doing it wrong. For example, even if we take the 10M accounts as correct, then you can spin up that many erlang processes simultaneously on a single 32Gb instance of some VM. (erlang processes are approximately 8Gb minimum per 3M (on a 64-bit system).) A dedicated CPU 32Gb Linode runs 120 USD per month.

Now, there's more to it than that, of course, including a database (as you mention, although scheduling is simple per process) though there are many ways to limit the checking. However, this is not something that requires large scale infrastructure along with its associated costs.

At bottom, you're just pinging endpoints and sending notifications based on results. Trivial stuff, even at massive scale, which this isn't (by today's standards).

Nice try but you ignore a couple of key points; we passed peak push a few years back and now the push wells are running dry. The cost of a barrel of pushes has grown exponentially over the last few months. The government has started rationing pushes to ensure the supply of military pushes and medical pushes is not compromised, as well as subsidising agricultural pushes and stockpiling pushes for future use. This has only increased the burden on consumer push suppliers and users face increased push costs or a loss of push services completely. Senator Ted Stevens forewarned us that the internet tubes would be eventually be clogged by unsustainable push use, but the propaganda campaign by big push drowned him out.

 

[Eric8199]

For my part, I spent the afternoon migrating two macOS and two iOS with three email accounts back to Apple's Mail.

That's wasted time, far more costly that the software.

This is my second experience of Apple's enabling of bait and switch. The other was Ulysses.

While I hold both developers in absolute contempt – equivalent to thieves or, more accurately, pickpockets – I also blame Apple for enabling this practice. Apple controls the market absolutely. None of this would happen if their rules precluded this obnoxious behaviour. Where are they?

Henceforth, I will not buy anything from the MAS until Apple blocks this practice.

I did the same. I like having my work email in a separate app from my personal mail, but I need push notifications for my email on both ends. I don't like Spark, Outlook doesn't work for me for some reason, so I'm back to Apple Mail, which has always been excellent for me other than having to have my personal and work accounts in the same app.

 

[yayarod]

No advance notice is a big part of the problem here. Another issue? The people getting grandfathered are those who bought the full-price app within the last four months. Those of us who’ve been using it for years are apparently out in the cold.

The developer’s excuse that push notifications aren’t a core function is ridiculous. If they’re now publishing an app that doesn’t have, at a minimum, the basic features that the iOS Mail app offers, they’re just making things up to avoid Apple kicking them off the App Store.

I guess it’s time to switch to another client. Canary’s twice the price Airmail was, but I have a feeling it’s unlikely to go subscription. One never knows, though. I didn’t much like Spark when I tried it out some years ago; it’s time to take another look. Haven’t ever looked at Polymail - will be doing that today.

I’m not generally one for punitive action, but I would like to see Bloop lose App Store access over this unless and until they change their policies. At a minimum, previous purchasers should get grandfathered subscription access - at least until there’s a complete redesign with improved functionality (Airmail 2, or whatever). If they feel their product is good enough to win new subscribers, fine. Forcing existing customers to pay more to continue to use pre-existing features is immoral.

Spark is fantastic but it only supports IMAP.
[doublepost=1563311787][/doublepost]

Try Spark, Outlook or Apple Mail.

Spark is fantastic but it only supports IMAP.

 

[jdclifford]

Spark is fantastic but it only supports IMAP.
[doublepost=1563311787][/doublepost]

Spark is fantastic but it only supports IMAP.

DO U mean as opposed to POP mail accounts?

 

[jonblatho]

It isn't quite as cheap as you might think it is. If you have 10,000,000 email accounts you need to check every 10 minutes with each check taking on average 1 second, you end up doing 44,640,000,000 checks monthly and as many seconds of time. Likely the interval is shorter than 10 minutes too.

Because the checking is not compute intensive (it's mostly waiting on Gmail's API) let's say you decide to get AWS EC2 t3.medium instances with 2 cores each at $0.0416/hour. Let's say you can run 8 "checker threads" per core as mostly they spend their time waiting on API calls and not doing any computing.

You end up needing 1042 EC2 instances for running the checking at 24/7. This will cost you $31,200 per month and only includes the cost of _checking for mail_. It doesn't include for example the cost of databases (where the account information is stored), the components that schedule the checking to happen once every 10 minutes for each account, storage to store the checker application and the logs, any automated monitoring (you really don't want to manually go through 1000 instances when something goes awry), data transfer etc. Plus you would often have some extra capacity provisioned in case there's any slowdowns with the external APIs.

Sure, you can get by with $5/mo DigitalOcean machine for your personal use. But in scale, you need a lot more.

Containers and microservices exist for a reason.

 

[yayarod]

DO U mean as opposed to POP mail accounts?

Yes, Spark doesn't support POP3 accounts.

 

[wirefire]

Simple solution... they should have released "Airmail Pro" and used the subscription model on that software. Airmail should have been discontinued (no longer sold) but allowed to exist in its current form. Until some iOS update destroyed it and forced everyone to move to airmail pro. I don't like it but I was screwed on this just like Plants Vs Zombies. I have the original paid for version on my iPad but it has bugs, those bugs do not exist on the current advertising supported version but the old paid for version will never be fixed.

 

[dsusanj]

... and now they have started deleting negative comments and locking threads with constructive criticism and advice on reddit:

https://www.reddit.com/r/Airmailreddit/comments/cdv5f0/app_doesnt_work_with_subscription/

Sad to think that this is all from a company who won a wwdc design award and an endorsement from Apple couple of years ago.