> with anything else meaning “upgrade to PRO” which, by the way, costs EUR 0.99 monthly or EUR 8.99 per annum.
This is exactly what I expected to read when I started reading. So many apps have done this to me that I'm now hesitant to install, and become dependent on, any new apps.
I'm genuinely sorry, but I'm not paying a subscription for an app that doesn't do a lot of back-end work (that's valuable to me). No, syncing doesn't count; I have iCloud and Nextcloud and the share sheet functionality for that. No, social features don't count either. This goes quadruple for apps that have a feature enabled and then sneak it behind a paywall many updates later.
And when I write that I'm genuinely sorry, I really am! I know that developers need to earn a living but I'm not willing to sustain it on the back of 10 or 20 or 30 "low annual price of just $9 or $15 or $25 or $49 per year!" subscriptions. If that means fewer apps or fewer people making their living at being app developers, that is bad and I hope that applies pressure on Apple and Google to add features to their app stores that make a better balance.
Apple clearly must have some strategic reason for continuing to resist this model year in and year out, but there is a proven middle ground between subscription-only and one-time-purchase licensing. Software in a handful of fields (Sketch in UX design, Bitwig Studio in music production, Jetbrains in software development) has adopted it, and it's simply where you gate updates rather than access by subscription.
It's the fairest model there is: Developers get a recurring income stream so long as they maintain the software, while customers aren't forced to rent software ad infinitum.
I wish there were some way Epic V. Apple could force Apple to implement this. It would solve so many problems on the App Store.
iOS's backwards compatibility for apps built with old SDKs is mostly pretty solid. It tends to be things more like architectural changes (e.g. dropping 32 bit) which are an issue
Also new features for the iPhone wouldn’t be supported across all apps unless people paid to upgrade - and Apple fundamentally makes most of their money by selling phones.
They release a new major version every X years and stop supporting the old one. The old one will keep working until it doesn't. If you want the new one, you buy it and upgrade.
Airmail used to do this and I believe Tweetbot still does it.
Airmail messed it up and I stopped using it altogether after years. I talked to friends that I know used it and they also moved away from it when they messed it up.
It's incredibly difficult to migrate even engaged users to the new, different app. Apple should really provide a better route for major version upgrades, especially on the Mac App Store.
I'm not sure how your experience with JetBrains products, which have a dedicated, good flow for easy upgrading thanks to not being limited to the app store, is a good counter to the point that this model could use better support on the app store?
It’s 6 flipping dollars a year. For an app that relies on a third party API and needs fairly regular updates. While I am not a big fan of subscriptions, feels like a good fit for Tweetbot.
> Apple clearly must have some strategic reason for continuing to resist this model year in and year out
Subscriptions make for more services revenue, and services is their growth driver now. That's all there is to it. If there was an option to stop paying and keep what you have they would make less money. Same reason you now get what are essentially ads in the Settings screen "reminding" you about AppleCare+, Apple Arcade etc. Services services services.
As a long time fan of Apple hardware and software it's pretty sad to see.
Wouldn’t it be possible to emulate this in a somewhat contrived way with in-app purchases? Make the app free, and then have a new in-app purchase each year: “2020 features”, “2021 features”, etc. Every year would include all the features of previous years, so new users on year N don’t have to pay N times as much to start using the app.
This seems a bit convoluted. Another is the drop to 15% for subscriptions after the first year. I believe this only affects developers doing $1M or more in revenue. But still
I confused my self. The 15% across the board is for under $1M. While 15% after first year of iap subscription would only benefit $1M+ Devs as the others are already at 15%.
In other words I complicated it. It’s 15% for all in app purchase subscriptions after a year for all developers.
The problem is that you find out that you don't have unfettered access after you update, when the app announces that the export feature is now crippled. (One way to fix this would be to export all data before every update, but that seems unreasonably cumbersome.)
To me, as long as it's available, export to an open standard is all I need; but I agree that, if it's revocable, then it's hardly unfettered. How would an app allow unfettered access in an irrevocable way? That is, if I were an app developer (I'm not), and wanted to provide this service to my users in a way that wouldn't require them to trust my good intentions, then how would I do it?
> This is exactly what I expected to read when I started reading. So many apps have done this to me that I'm now hesitant to install, and become dependent on, any new apps.
To provide a little context: Most developers have no choice but to do this. I've seen a lot of developers for the App Store getting broke because of Apple's refund policies. People are quick to click refund, but then the developer has to pay for Apple's 30% of the fees (and transfer costs).
Say, you sell an App for 10.00$. Apple gets 3.00$ of that when a user buys the App, and the remaining 7.00$ go to the developer's account. When the user is now unhappy with the app they can get a refund for the costs.
User gets 10.00$ back, but from the developer, which means the developer now has a net negative of -3.00$ (ignoring the accumulative bank transfer costs)
And no, there's no limit in how many people can scam an App developer on the App Store, which means that you could literally use millions in the case of a social justice thing that goes viral, which happened to quite some Indie game developers I personally have worked with and was a target of in the past because the target audience (kids, teens and the like) doesn't realize the damage they're causing with this. Kids think it's funny and it forces the Game developers to implement the features they so desperately want, but the result is that the developer is broke and legally so indebted that they won't be able to recover from this anyhow; certainly not when they are a private developer.
> I'm genuinely sorry, but I'm not paying a subscription for an app that doesn't do a lot of back-end work (that's valuable to me)
What does count as "a lot of back-end work" that's valuable to you?
> I know that developers need to earn a living but I'm not willing to sustain it on the back of 10 or 20 or 30 "low annual price of just $9 or $15 or $25 or $49 per year!" subscriptions
Why are subscriptions bad and what would your solution to this be?
1. If I ever stop paying, I get nothing, no matter how much I've spent on the software over the years.
2. It shifts a lot of the risk of developing the new version onto the customer. I'm paying in advance for the next version, instead of being told what the features are and being able to make an informed decision. If I hate the new version, see (1).
3. It's more winner-take-all. If I'm spending $Y/mo on an app, that's $Y/mo I'm not spreading around to other apps. All-in-all, it feels like a clear win for the developers that 'win', but it feels lose-lose to me as a customer. I get all the joy of (1) and (2), but also I get less variety, both in my own purchases, and probably in available choices in the long run.
[0] - To be clear, I'm talking applications which require subscriptions for basic functionality, 'extra' functionality that was clearly removed to push subscriptions, or include useless tacked-on functionality like cloud storage (also clearly to justify a subscription). I'm not talking about apps for which there are recurring costs associated with my continued use of the app.
Edit: An addendum: I also like that I don't have a contract that explicitly states what I'm getting for my money each month. Usually it's just access to the app. Then they'll have language about how the subscription allows them to work on new features, offer support, etc. But they're not contractually obligated to provide me with any of that. (I don't have any levers at all, except not giving them more money, which doesn't help much.)
> 1. If I ever stop paying, I get nothing, no matter how much I've spent on the software over the years.
The Zune music service actually had a great fix for this: every month they allowed you to download a couple DRM-free tracks that you could keep forever. So if you ever cancelled, at least you had those tracks.
That was nice, but at the same time I disagree that when you stop the service you are left with nothing. In fact, what you got was access to the service. If I ever stop paying for Spotify, I'm left with nothing... except for all the hours I spent listening to and enjoying music on Spotify without ads. Isn't that something?
I used JetBrain on another comment. I like their license.
If I pay for a year, I get to keep access to that version.
Regarding Spotify, for Spotify, while yes they have an app, you're mainly for the music streaming from the service. They mention that subscription to services are fine.
You’re talking about a service. Every time I use Spotify, it costs Spotify money. This cost is why I specifically excluded subscription services from my criticisms. It makes sense for services with recurring costs to have recurring billing.
Spotify the Application, is just the means of accessing Spotify the Service. It provides me with cheap access to a library of data that I don’t otherwise possess.
By way of analogy, I understand why I have to pay for a magazine subscription: they send me information on pieces of paper every month and that has a cost. But I wouldn’t pay a monthly fee to the manufacturer of my toaster. Whether I toast a thousand slices or let the toaster collect dust on the counter, the manufacturer incurs no additional cost. Leasing a toaster would be absurd, but that’s the expectation some developers of pretty basic applications expect.
One of the related posters mentioned JetBrains. I agree JetBrains get it right. You get to keep your toaster when you unsubscribe. If you stay subscribed you get the latest toaster if that’s what you want. And the incentives and risk are on them to make sure they make the next model toaster compelling.
You said “To be clear, I'm talking applications which require subscriptions for basic functionality”
That’s exactly apps likes Spotify. They restrict downloading, skipping, and bitrate unless you pay for premium, which I would say is basic functionality. You get access to the music whether you pay or not. I’m not paying for the music, I’m paying to skip ads and download to play offline.
> You said “To be clear, I'm talking applications which require subscriptions for basic functionality”
You're taking that statement out of the context of the post in which I said it. I began my post with "Subscriptions to services are fine." Everything in the post after that, including that quote, is specifically and explicitly not talking about services like Spotify.
You're not understanding my point. I'm saying that Spotify is actually closer to the model you decry than to a subscription service of which you approve. An example of a subscription service would be Netflix. When I pay for Netflix, I get to watch Netflix. When I stop paying for Netflix, I don't get to watch Netflix anymore.
Spotify is different. With Spotify, when I stop paying for Spotify I still get to listen to the music. But I lose the basic functionality of the app, like the ability to save songs to disk, or to skip tracks. To me, this seems like the exact model you're saying is bad. They have removed basic functionality in order to push a subscription.
No, I didn't misunderstand your point. I understand the distinction you're making between Spotify and Netflix in terms of how subscriptions are motivated. My post is about the distinction based on recurring versus fixed costs.
My footnote is not part of my argument. It's only meant as a clarification for which types of applications (not services) I'm discussing. You also mistake what I mean by 'basic functionality'. I maybe should have said required functionality. Spotify is perfectly usable without download, skip, or bitrate. Instead, I'm talking about applications like an RSS reader (again, an application only, not an online service) that will only show you sample feeds if you're not subscribed.
> What does count as "a lot of back-end work" that's valuable to you?
By way of examples, something like a weather application or access to content. Weather data is, at least until the end of humanity's existence on Earth, ever-changing and in need of update.
> Why are subscriptions bad and what would your solution to this be?
Because I do not want to wind up in a situation where an app developer suddenly turns on subscriptions and hides features of an app behind it, on the back of an opaque update. Plus, for most apps I am quite happy to pay for a particular version and be done. (I have no problem with paying for applications.) I want the finality of the transaction, that I've given my $5 or $10 or $20 or what-have-you a single time and our business is done. Application makers have done this since time immemorial, sometimes releasing complimentary bug fixes, other times not.
I just don't like the idea that in order to retain access to a program that does all of its work on my device, I have to keep paying rent for it. Like the app at issue here, where it was doing what the user wanted and then poof, a change that cannot be reversed.
I also understand that the current model on the app stores is not conducive to one-and-done application purchase transactions. Apple doesn't make it easy to do upgrade-by-version pricing, for example. In-app purchases are available, but they're not (by default, or nearly as widely) sharable on the family-level like outright purchases are.
My example is RSS feeds sync. No need for Feedly, Reeder 5 has just landed iCloud support so now it sync through iCloud whatever backend mechanism API it provides and boy was I glad to pay 7.99 AUD for that.
> > I'm genuinely sorry, but I'm not paying a subscription for an app that doesn't do a lot of back-end work (that's valuable to me)
> What does count as "a lot of back-end work" that's valuable to you?
Perhaps functionality that's not easily recreated using built in tools like shortcuts ?
In the case of the app in question, a simple shortcut in the shortcuts app or in i.e. Pythonista that adds a record to a spreadsheet, or even directly to a .csv file, would take <1 hour to write.
Another example is apps that provide some kind of backend. Of course keeping infrastructure running is not free. For the app in question it's not providing that.
> What does count as "a lot of back-end work" that's valuable to you?
Overcast handles a full index of pretty much all podcasts, with specific metadata. For example, it learns how often a podcast updates. Is it a weekly show? If so, what day does it come out on? etc.
Just the part where it syncs where I am in each episode is just a bonus, I'm happy to pay for the feed database.
I moved off Fantastical for this reason - I used to recommend it to my friends and coworkers but I kind of feel affronted that features disappeared behind a paywall that I'd paid a license for previously (several times actually).
I really wish Flexibits had created a different app without the cloud features and let this one languish. Clearly they needed a "teams" version for families/small enterprise that they felt they could monetize. That they bundled it into their main app annoys me.
Same with 1password - but in that case, I felt it was worth my subscription (family for several users, admin capabilities worth it alone).
You can't even see the date in the menu bar without paying. Those who don't pay see a menu item that has no date. That was egregious. However, previous versions also had the "travel time" feature that's already an macOS Calendar.app feature.
Finding out that macOS/iOS Calendar is perfectly fine for me.
I don't really see how these "there's a word for that in German" are much, if any, of an improvement over their English equivalents.
Verschlimmbesserung
Improvementagrrivation
It's only three characters shorter, and English being a Germanic language, English is often referred to as simplified German, the languages maps pretty much one to one in the general case.
English does the same sort of thing were you'll see something like use case become use-case and then usecase.
This is called portmanteau in English, and here's the Wikipedia article, with some common examples.
>I don't really see how these "there's a word for that in German" are much, if any, of an improvement over their English equivalents. Verschlimmbesserung Improvementagrrivation It's only three characters shorter
The different is that the former is a legit syntax/form of a composite word in German (where you can combine words more freely), whereas Improvementagrrivation is not. Also the former conveys a meaning, whereas the latter would have people scratching their heads.
The limitation with port-manteu and/or compound words in English is that you don't just create them and that's it: they also need to be adopted in order to be part of the language, accepted as words and understood as terms and intention (and not just an error of forgeting to add a space between two words), etc.
In programming terms, one might say that German has first-class support for compound words, whereas with English you need to wait for each new word to slowly be added to the vocabulary.
> In German you can combine more freely, and people will more ofen than not, get it immediately too.
It could be true but I'm sceptical, I reckon tht in most languages you can change things on the fly and people will get it. For example, the next comment down from yours has this:
> I suppose in English it's an updowngrade?
All you'd need is the context and pretty much everyone would get it. You may say that context is the difference but doesn't everything in English need a bit of context?[1] (I'm also a bit sceptical of the context thing too though!:)
It may be less-acceptable to combine words in English, but that doesn’t stop me from doing it when it feels right. I use hyphens, though, to help with clarity. If enough people use the same hyphenated word the hyphen might be dropped. From what little I know of German (Pimsleur CDs, then about three months around the country), it does seem to accommodate these combinations better than English does. Is it okay that languages change over time? If so, how do they change?
Look at English. US vs UK vs Australia vs South Africa.
These were all once the same language.
Cypriot Greek is another good example with some millennia of divergence.
It amuses me when we brits complain about US spelling when in some cases it’s we that changed and not North America.
English has compound nouns: strings of compounds that form a noun phrase. The main word, the head of the noun phrase, is rightmost. I think that is called head-final.
For instance "law school entrance exam administration".
This noun compounding process is extremely free and productive in English.
Neologisms are adopted just as quick in German. The difference really lies in the inherent ability by the language to convey meaning almost no matter how crazy the word is.
I am willing to bet that every German will likely understand Morgenkaffeesehnsucht, the desire for coffee in the morning (which I just made up on the fly).
There are certainly English phrases shorter than their German counterparts. "Car insurance" vs "Autoversicherung", for example. It's just that beyond a select few exceptions certain types of things are not a convention in Standard English, like nouns for very specific feelings.
It doesn’t transliterate to the same. That would have been “verschlimmerungverbesserung”
Instead, the “ver-“ prefix and the “-ung” suffix are factored out and basically, a new verb is invented and immediately can be understood by German speakers: “verschlimmbessern”
>It's effectively the same as saying "improvement aggregation"
No, it's not (I also assume you meant "improvement aggravation").
First, pragmatically, as it is at the monent:
(1) "verschlimmbesserung" IS a word in German, and is in use, people understand it, use it to express their update pain, etc. "improvement aggravation" on the other hand, is not a thing. It's just two words next to each other. People will scratch their heads as to what you mean, and you'll need to explain it with further terms in a circumlocutory way.
And that's not by accident, it's because (2):
(2) The second issue is that in german adding two words together (or more) "parses" in the language, and creates a valid new term (with either an obvious or a culturally transmitted meaning).
In English that's not the case. You can't just say "This is an improvement aggravation" (I mean, it's a free country, so you can, but people would just think you've made a syntactic mistake).
>I don't really agree much with your last point. Someome still has to come up with the word in German, and others have to use it.
Yes, but like in Lisp with macros, in German the speaker (programmer) is allowed to come up with such words.
It's not really a new word, nothing new has been added.
If we take the context away from the German as used here in this case, then it's just as meaningless as the English context-free utterance.
The German term is just the two words smooshed together.
Have you ever watched Lucy The Daughter Of The Devil?
There's an episode where Jesus and Satan are doing smooshes.
People think these German words and cool because they're novel and foreign, but English has plenty of very similar sorts of constructs, weirdness, and coolness too. Dang nab it!
- I used an app (gas cubby) to keep track of my car mileage. You could enter your vehicles. When you filled up, you would select a vehicle and enter mileage and gallons and cost, and it would keep track of everything. You could export the data. You could also enter lots of other things, like VIN, insurance, service intervals, etc.
It was a decent app and offline.
And one update - everything changed. It made everything cloud based, uploaded all your very private information and added a login. jerks.
- I used an app (camscanner+) that would let you take a picture of a document, it would find the edges and turn it into a .pdf file. This was sold to tencent, which had no privacy policy (broken link) and uploaded all your data to the cloud. When the privacy policy link eventually worked it was in super ambiguous broken english and basically said they use all your personal information.
- I used an app called adblock ios that created a VPN at 127.0.0.1 and allowed you to filter your phone traffic. Apple made them change (cripple) it. Happily I read the 1-star reviews and didn't update.
I think I'm a pretty astute user, the general population just has to get used to being worked over in this fashion.
Apple is 100% in the wrong here. You should:
- be able to know what your phone is doing, what sites apps are contacting
- be able to firewall your phone - even to apple
- know of changes - especially change of policies/behavior/ownership before installing an app
- be able to revert apps
I think the GPL is becoming more and more important as this stuff has taken root.
I'm sure they will, given how bad they screwed up their main app. "Hey let's force all our users to move to a slower version with less features, built on Electron, it'll be great!"
Earlier this month I got tired fumbling through all the little steps to use notes app for scanning and then separating out the PDF so I blogged[0] up the steps (including screencaps).
I don't know for the other commenters, my main use of a document scanner is to backup in PDF form on my NAS and in Google Drive so I can trash the physical copy.
The scanning part is only the beginning, the phone -> network pipeline needs to be automatic and reliable, and Apple will do nothing for third party synching.
Perhaps a Shortcut pipeline could do it ? If it still gave enough flexibility about batch scanning etc. ?
Thanks, but those are just hooks in the app (and stuff like DSFile disconnect every here and there)
You still need to manually save each document where you want it, instead of the app doing it automically in batch at every scan. With dedicated scan apps I never need to think about it, just scan and I know it's been done.
Yeah I was happy when they added scanning to the notes app, which I believe was the first time apple added scanning (the time period was before files).
It's a shame this guy can't easily roll back to an earlier version of the app. This is one of the reasons why Android users back up APK files of the apps they like. That way, you're still free to use the software you already use, license withstanding.
The thing is, it used to be possible on iOS but along the way Apple crippled it. Back in the days you'd connect iOS devices to your computer and sync them, the syncing can involve transferring *.ipa files from the device to the computer, and can also install new apps. If you wanted to roll back, just delete the installed app on the device and ask iTunes to reinstall it using the file from the computer.
Eventually other features (app thinning?) got prioritized and this is no longer possible.
Starting in August, Google is requiring new apps in the Play Store to start using app bundles instead of APKs; I have never tried to work with these .apkm files, but I would be concerned Android is moving in the same direction here.
I believe what the parent comment is saying is that since you can sideload apps on Android, if you want to roll back an update, all you need to do is grab the apk from a website such as apkmirror and then install it. The Play Store moving to app bundles shouldn't affect this.
...but the APK file never existed if the app was shipped as an app bundle. The APKMirror people seem to claim you will at least need a special installation app to install the bundle components, which is fine for now as that app can be an APK, but... this is feeling sketch.
While having to use either a custom installer, or if you don't trust any third party app [1] alternatively using ADB (or possibly the package manager directly if you have root) to install split APKs, is somewhat of a hassle [2], the more annoying part is that grabbing an APK from a random device is no longer enough to cover all screen densities, languages and whatever other criteria Google uses to split APKs [3], so good luck finding an older APK version if you're using some exotic screen density or (more likely) language.
Curiously enough by the way, Google themselves only recently added a feature to the Play Store which officially allows directly sharing (free) apps with nearby devices. I wonder how that is supposed to work going forward when app bundles will become more and more widespread – can you then only share apps with somebody having the same screen density and language setting?
Or can apps at least still fall back to dynamically using the assets for whatever language/screen density was available just like they would today with a monolithic APK?
[1] As a perhaps more trustworthy alternative, there's SAI on FDroid (https://f-droid.org/de/packages/com.aefyr.sai.fdroid/), which a quick search indicates that it might be able to read and install APKMirror's bundle format, too.
[2] Also the fact that it makes Titanium Backup semi-useless for those apps, even though in other respects I still like it better than any of the other alternative backup apps I've tried.
[3] CPU architecture at least, although there's less variation there because almost everything will be either 32- or 64-bit ARM, and even before App Bundles some apps would already ship separate APKs for different CPU architectures.
That doesn't make any sense at all, even merely cryptographically (as you would need the signing key to merge the bundle back into a normal apk file and sign that new different file), and doesn't match the behaviors people talk about on various blogs and stack overflow.
Well, the app I tested — YouTube Vanced — is installed as a bundle, you can see Vanced Manager download its parts. I did check with jadx, that apk is validly signed with v1, v2, and v3 signatures.
Now, as far as my own knowledge of apk signing goes, v1 signature should definitely be fine because it operates on files inside the archive, so naturally merging multiple apks signed with the same certificate would keep it valid. But v2 signs the archive itself, which means merely repacking it would render the signature invalid. I don't have much of an idea about v3. Maybe there's some algorithm to merge multiple apks such that signatures remain valid.
Related question that needs research: how are bundle apps stored after installation? I know that installing a regular apk copies it into /data/app, extracts native libraries from it, and adds its metadata to some XML files I forgot names of. Again, it's been too much time since I did something serious for Android, so I'm not quite up to date on this stuff.
(are you the guy behind Cydia btw?)
edit again: apparently it doesn't back up the additional parts
The apk I backed up earlier is base.apk, so that mystery is solved. It probably just adds them all to the classpath in runtime like regular JVM does with a bunch of jars.
there seem to be edge cases where the bundle isn't extractable back into an apks but despite packaging my apps as bundles and downloading from the play store they still end up as useable apks in the end.
as far as I've gathered bundles are a way of adding multiple ABI's into a smaller shared apk that then is split into per abi by google but I may be entirely off base here
In theory, but in my experience when restoring a backup most apps don’t actually get fully restored, it’s more like a metadata pointer which downloads it from the App Store. But sometimes you’ll actually get your older version of the app from the backup this way. I successfully used this technique to restore an old copy of Slack on my phone for a year after they changed the app logo, purely out of spite. But it doesn’t seem to work consistently, I think it’s dependent on something about the way the app is published which I don’t fully understand.
Steam has the steamdb listing all versions of packages that are available for download. As long as you have a steam account that "owns" a license for the software you can use that info to download any listed version through steam.
Really? I've seen developers add old versions as "beta"s as a workaround to continue saves after a big patch so I assumed downloading old versions wasn't normally possible.
It's not possible through the Steam interface, but there are third party apps on github that let you download old bundles. I used one recently, but I forget the name, sorry.
I am not sure if every old version stays available. However I had to binary search through versions of their VR tools once because only a few versions ran on my dev. system. The normal steam UI doesn't really expose the versions or provide an option to prevent updates, so the beta workaround might be necessary to tag these versions for end users.
Wouldn’t that put an undue burden on app developers who have a server-side API if their users were running a bunch of different old versions, en masse?
Because of things like this I stopped updating apps a long time ago. But now many apps have a “phone home“ aspect, where it will literally lock you out of using the app if you don’t upgrade to the latest version every so often. Which is especially annoying if you need to approve a transaction that was blocked on your credit card but then you can’t do it until you update the app but your signal isn’t great, or you need to reply to a Facebook message but even though you can see the message the app refuses to function until you update. And a million other frustrating examples.
I miss the days of things not changing out from under me without me having any say about it. I remember a time when I looked forward to updates because they brought interesting new functionality or, you know, actually fixed bugs. And when I didn’t like a new version of something, I could simply go back and reinstall the old version and keep using that.
Absolutely agree. If an app stops working just to show a message like "your App is out of date, download the latest version to continue", it's an instant one-star review from me.
I have used the iPad for over 10 years now, and I can now say that most of the apps that I have had in the past that were actually useful are now unusable or worse than they were.
It is actually amazing that Apple’s ability to prevent people from downgrading aligns perfectly with developers wanting monthly subscriptions for everything.
The only application that has gotten better consistently over the last 10 years has been iThoughts, which is one of the best tools ever for high-level abstract thinkers, and it is the only reason I use the iPad now.
> aligns perfectly with developers wanting monthly subscriptions for everything
Let's not pretend this isn't also something Apple wants. There's a reason they never implemented paid upgrades despite developers begging for it since the beginning: every new app subscription is more services revenue, and services is driving their growth right now.
I do wonder about the legality of an update taking features away and charging for it, as is the case in the article. It sounds like it's illegal from previous examples I've heard. I guess no one is going to sue over some relatively cheap niche app though
Is it going to be any use though? Certainly for apps that store everything locally, but the way things are going, when the cloud backend stops supporting the API version the older app uses, you're screwed...
And even if apps do use offline storage, the problem you get is that eventually OS support moves on. I have this issue with Adobe Creative Cloud - I still use CS6 (pre-cloud) because it doesn't require a subscription, but now I have to keep an old machine dedicated for that on Mac OS X 10.14 (currently two versions behind - I expect it won't be supported for that much longer with updates)...
What would get around that is requiring being able to point apps at your own server, and requiring open API documentation, but I don't see how that could really be workable in the real world...
> when the cloud backend stops supporting the API version the older app uses, you're screwed...
I think a standardized software "Nutrition facts" or "Truth in software" disclaimer would work for this: a simple table displayed in app stores that makes the important facts visible, such as the effective-price-per-year and data privacy declarations we have today, but add a line for how long-term support works, and what parts of the system are dependent on first-party hosted services and to what extent it can operate offline or with self-hosted services).
I'm not opposed to SaaS (after all, it's how I earn my living right now), but I am strongly opposed to software companies pulling a Darth Vader and altering the deal of software through app updates that most people don't think twice about, such as when Angry Birds changed from being a straightforward $0.99 app without ads to being another trashy freemium game[1], to Adobe dicking around with Creative Cloud
> What would get around that is requiring being able to point apps at your own server, and requiring open API documentation, but I don't see how that could really be workable in the real world...
Again, my issue isn't the fact that Cloud-hosted SaaS exists, but that customers get sold one thing, but which slowly morphs over time into something that no-longer works in their best interests.
I remember decades ago, being a sprog in lower secondary school, in IT class, when we were discussing the nature of "business software" compared to retail software and computer games, and the teacher was telling us about how site-licenses work compared to individual EULAs we'd all seen - but also that when the school or local education authority, bought software, especially custom or modified software, that they'd seek a clause in their support contracts with the vendor that the source-code of the software be placed in escrow in the event the vendor goes bust - so the school can keep on running. That's sensible. What I'm proposing is that SaaS vendors that reach some kind of minimum critical mass of customers, should be legally accountable to their customers to provide a working self-hosted system - or at least straightforward steps towards building a working self-hosted system, in the event of insolvency or their bus-factor going below unity - this doesn't need to be arduous: given we're all hosting code on GitHub, so just a law that states "reasonable precautions" must be taken should be sufficient. This would need to be a criminal statute thrown at the company's execs though, given that if an LLC company goes bust then anyone pursuing a civil law case doesn't have any company to chase after.
-----
[1] I'm not angry (hah) that /Angry Birds/ became an exploitative freemium game: I'm angry that the game that I paid for was compromised: I think the iOS App Store should have rejected their proposed update when it went freemium and instead required Rovio to launch it as a separate SKU in the store. Unrelated, but I'm also angry that games and apps I paid-for on the iOS app store can be pulled by their publishers with absolutely no-way to re-install the game or app if we do a clean-reset of our devices unless you made a private backup of the IPA files (Apple will give you a refund if you complain, at least...). EA is bad for this: they'll routinely remove great games of theirs from the iOS App Store for the most banal of business reasons: the won't even simply remove games from being advertised, but block paying customers from downloading previously paid-for games - which is ridiculous.
It's really unfortunate that there are no commercial incentives to provide strict sandboxing for outdated versions of applications, rather than only allowing the latest version to be installed. There's a strong public good in preventing versions of applications with known vulnerabilities from accessing the network (or perhapse restricted to an (possibly empty) allowlist of domains). However, the inability to save old installers and run old applications in some cases leads to lost functionality, which is sometimes effectively lost data.
I got my wife to start using a password manager, but I really should have given her some guidance. She picked one that was popular in her native language. I also encouraged her to upgrade iOS on her phone. We backed up her phone and upgraded, but the new iOS version would drain her fully charged phone in an hour or so. So, we restored iOS from backup. Unfortunately, her password manager had been discontinued, and iOS backups just contain effectively remote symlinks to installed apps rather than containing the actual binaries. The discontinued password manager then showed up as something like a stub displaying an error message that the app could not be installed. Luckily, I was able to get into the backup and determine the password manager used an unencrypted sqlite DB with English column names and I was able to give my wife her passwords back.
I would have much preferred a warning that the backed-up version was no longer supported, and be given 3 options: (1) install the last released version in a locked-down sandbox (2) install the backed-up version in a locked-down sandox or (3) uninstall the app.
> I was able to get into the backup and determine the password manager used an unencrypted sqlite DB with English column names and I was able to give my wife her passwords back.
WTF?
No wonder it got pulled!
I am curious as to why you don’t just use the native password manager?
I don't use iOS Keychain because I've been an Android user since Google gave all employees a free HTC G1 for the holidays. I had an original iPhone, but I don't recall that having Keychain.
As for why my wife doesn't use the iOS Keychain, I'm not sure. I try to avoid pressuring her to use one solution or another. If I pressure her to use some solution and there's a bug, then that's on me.
Apple uses their App Store rules to ensure a good experience for customers. The descriptions on apps must accurately reflect what the app is about.
What if Apple required the changelog on app updates to accurately reflect what changed? This would create a better experience for users and is exactly the sort of thing I would want in my walled garden.
> 2.3.12 Apps must clearly describe new features and product changes in their “What’s New” text. Simple bug fixes, security updates, and performance improvements may rely on a generic description, but more significant changes must be listed in the notes.
I was super annoyed when the really fun Galaxy on Fire 2 HD that I paid $9 or so for got bought by a different company and suddenly had ads and micro transactions. I haven’t played it since.
Also annoyed when Apple removed Binding of Isaac which I paid $15 for, my money and the app never to be seen again.
iOS apps are listed in your purchase history and available for re-download. Even those removed from sale on the store. So unless you’ve hidden the app (an unnecessarily complex process) I’m not sure why you’re not able to continue playing The Binding or Isaac. Unless of course the app is old enough that it doesn’t work on newer iOS versions. But that’s not what your post was implying.
> iOS apps are listed in your purchase history and available for re-download. Even those removed from sale on the store.
Nopety nope.
The Secret of Monkey Island vanished from my purchase history after Disney bought LucasArts. Apple support essentially said "Fuck off, dear customer". The end.
I bought a game (Lumines) that downloaded all the necessary game files upon launching from a server. The server is now down. No way to play the game even after downloading it from the list. screw the app store.
Wait, isn't this the fault of the company that made the app? How does the App Store have anything to do with a company shutting down their server or not?
I wish the App Store guidelines had a provision for disallowing that behaviour: if there's no online component (e.g something like WoW or Fortnite, that have regular content updates) then the app should be self-contained in the App Store.
Sadly it used to be the case (not sure now) that the app size limit had Apple itself recommend that behaviour be implemented when it was too big.
It might have been a mitigation for the 4G download limits where people couldn't download games over X size on 4G so they started downloading content on boot to get around it.
Apple should forbid this practice then. I have plenty of apps that I paid for that don’t work for various reasons. I don’t get to keep my purchases but Apple gets to keep their hefty 30% fee.
+1 for Galaxy on Fire 2. It was proof that the "pay a real price once, enjoy a real game" can work in the appstores. The best game I ever played on mobile, ruined in a single update. Galaxy on Fire 3 was a sad joke compared to even the first installment.
I’m in the minority on this one, I’m sure. But the subscription apps don’t bother me at all. I don’t even know how many I have..I don’t need to since the App Store manages it for me. I know that I pay for Concepts, Halide, and Overcast. But I don’t know what else. I don’t even know how much those apps cost..but I like using them and I want them to keep working well. Which means the developers need to be able to earn a living. Besides, what’s the use buying a phone that costs two thousand dollars if I can’t splurge here and there and drop a few bucks on an app that I like?
I would assume that this Pro version is intended to be completely unlocked all the time, while the free version is being monetized and directed towards people who want to try something but aren't dedicated users
(Although I don't understand why they don't simply have the "Pro" version be a $20 once-off in-app-purchase in the free app bundle)
> Although I don't understand why they don't simply have the "Pro" version be a $20 once-off in-app-purchase in the free app bundle
Most in-app purchases can't be shared between family members. I will never buy an in-app purchase again. I learned this the hard way after spending $20 on a children's game and my son could only use my phone to play it -- not my partner's. It was really annoying to always have to give up my phone when on the plane/train.
OmniFocus has a similar thing, with a standalone “Enterprise” edition (https://apps.apple.com/us/app/omnifocus-3-enterprise/id14847...), where IAP is replaced with an upfront price. I believe it’s because schools can get a volume discount on paid apps but not IAP, and Omni says it “simplifies deployment” for enterprise users.
Thank Apple for deciding for you that you can never install any version of an app (on your own device!) that isn't current.
More in the stream of neverending censorship bullshit from the App Store.
Apps aren't even included in "backups", so restoring a device from backup won't actually put it into the state it was before:
you get the current versions of any apps (if they are even still available, sometimes they aren't) re-downloaded at time of restore. This is, of course, contrary to the entire concept of what a backup is for.
Of course, you can't download any apps at all without an Apple ID, and also providing your unchangeable hardware serial number to Apple (transmitted when you launch the App Store like a supercookie). You also can't get the Apple ID without providing a telephone number that can receive SMS, so you basically have to dox yourself to restore a working backup.
I've had every single iPhone that has ever been produced (except the 3G - I went from 1 to 3GS), but I'm fairly confident I have purchased my last one.
As for smartphone, I have a few Pixel 3 XLs I'd like to test graphene and lineage on. In general I'm kind of over smartphones because browsing the web on them (the main thing I do with them) is suboptimal compared to a computer/laptop that can run uBlock and NoScript, so I will probably mostly just leave an iPhone-sized hole in my life.
The Nokia 106 is also good for receiving SMS, which is the only thing I really have a non-data
SIM for at this point.
I carry a battery powered gl.inet wireguard/lte wifi openwrt router with me with a Google Fi data SIM in it for internet. I have root on it and can write my own iptables rules, and it only sends traffic via the vpn.
FYI, Firefox on Android supports desktop extensions, including uBlock and NoScript.
They're pretty painless to use, too; there's an 'Add-ons' option in the main ⋮ menu with an entry for each extension. Tapping an extension's entry opens the usual extension dialogue in a full-screen view, and tapping 'back' in that view has the same effect as closing the extension dialogue on a desktop browser.
So you can use NoScript to enable a script and reload the page in...5 taps. It could be worse.
Linux phones are rapidly becoming more and more usable. The pinephone is already usable as a daily driver if you're willing to put up with a few hacks. For some segment of the population I think the trade offs they offer will be worth it and you'll probably start to see a decline in both Android and iOS market share.
I would love a proper linux phone. Seems like they're still a couple of iterations away from really being "ready" from video reviews I've seen, but I'm excited about something happening at all in that space
Honestly almost all market-guessing apps are crap from the start. And those that aren’t only fit you accidentally. For apps to be useful, users have to create them themselves or join into groups that discuss features face to face and quickly implement these, without apps being a private property (eula) or an unhealthy pride (foss). The main stopper for that is an enormous complexity of the app programming and the semantic abyss it turned into, on all platforms. Today it feels like 90’s programming was actually simpler (vba, delphi, various game makers), and every modern programming tool is just a gatekeeper of huge budgets.
If in 1990s I’d have to connect(mapStateToProps) or something similar to draw few spaceships on the screen, I’d become a lawyer or a salesman. And that’s still relevant. If you want cool apps, reduce salesmen and increase makers. Summon them from regular smart people of all ages right now and stop maximizing young geek culture.
Many apps follow a similar curve. They peak at a certain value/annoyance level and then the developers realize they're as high as they're going to go in terms of revenue. Then they start looking for ways to further monetize their users. For some of them that's adding dumb trendy features like social integration. For some that's locking previously available features behind a paywall. For a few it means selling your whole app to a third party to fill your users' phones with adware.
It's disgusting and it's not getting better. Probably 50% of my screen time is spent on apps I no longer update because the new versions are unacceptable.
You might be able to do a full encrypted iOS backup to your computer, which usually includes the Documents of most applications. It has to be "encrypted" though, or the backup excludes a lot of things.
Unfortunately Apple has made advertising in free apps unviable with their universally loved by users iOS14.5 update. "Stick it to the man!!!" Unfortunately, this means advertisers who funded free apps cannot pay for apps any longer. End users must start paying for apps now.
I have converted my apps to subscription now away from advertising. I am making a more steady income. I will convert my last app in the next month or two.
Apple sell more devices with this privacy change, and they get big profit in taking a 30% cut of subscriptions. They took 0% cut of advertising.
So many end users choose to abuse app developers who funded apps through advertising. Instead this is a self inflicted wound, end users support Apple making this privacy change, so now end users have to pay for apps instead of advertisers.
If you continue to see users as the enemy, it is likely you will not have much business success. In my experience, organizations that follow on that path don't bring marketable value for much longer.
So, what is the alternative? The same as it was with software ever since the dawn of computing: Sell major versions. Provide bugfixes in minor versions. Be more like GoodNotes, who follow that exact model.
I don’t see users as the enemy! Where did you pull that absolutely stupid idea from? ** DON'T ATTACK DEVS **. YOUR BEEF IS WITH A FRUIT COMPANY. My post states the facts - advertisers used to pay for the apps - now this has been stopped so users now pay for the apps!
My business is more successful because of this change. Users love the content in my apps. This change has not and will not hurt me financially. I am trying to explain why the dev made that change to you! Attacking the dev or offering junior level inexperienced advice will make no difference. Please take note of the reason why and who made the change that did this to you.
>Data export has changed from “export all data via CSV” to “export CSV of the last 7 days” with anything else meaning “upgrade to PRO” which, by the way, costs EUR 0.99 monthly or EUR 8.99 per annum.
As a German, you can luckily still request your data using the GDPR. I'd suggest finding an alternative (preferably Open Source) app and giving them your export file to ask if they can add support to import this other format.
Some humans do. Increasingly more people around me are switching to ISO date, inspired by the fact major organizations do the switch for easier sorting in lists and directories.
Changing behaviour like that in an export file in a format that traditionally is used for data exchange between computer systems, especially when not mentioned in the release notes, is a borderline bug.
This is exactly what I expected to read when I started reading. So many apps have done this to me that I'm now hesitant to install, and become dependent on, any new apps.
I'm genuinely sorry, but I'm not paying a subscription for an app that doesn't do a lot of back-end work (that's valuable to me). No, syncing doesn't count; I have iCloud and Nextcloud and the share sheet functionality for that. No, social features don't count either. This goes quadruple for apps that have a feature enabled and then sneak it behind a paywall many updates later.
And when I write that I'm genuinely sorry, I really am! I know that developers need to earn a living but I'm not willing to sustain it on the back of 10 or 20 or 30 "low annual price of just $9 or $15 or $25 or $49 per year!" subscriptions. If that means fewer apps or fewer people making their living at being app developers, that is bad and I hope that applies pressure on Apple and Google to add features to their app stores that make a better balance.