Record opened 2024 · Kingston, Ontario · entries read 2026-07-20
Mobile games / upkeep / en-CA
Is anyone still looking after this game?
A game does not stop working the day it stops being updated. But the two stores it lives in keep two separate dates, and when those dates drift apart you are usually looking at two different products wearing the same name.
Five entries. Every date below was read off the two public store listings on 2026-07-20, prices in CAD from the Canadian storefronts. Where a source would not give us an answer, the line says not confirmed instead of a guess.
Google Play · updated5 months ago
App Store · updated52 months ago
BADLANDone name, two datesgap 47 months
iOSPlay
2013-04-04 first releasetoday
This is the shape of every entry on the site. Hatching marks the stretch of calendar where one store had a maintained build and the other did not.
Who keeps the record01
I started it because a receipt did not match a product.
In March 2024 I paid for a puzzle game I had loved on an old Nexus, put it on a borrowed iPhone, and got a black screen with music playing underneath it. The App Store page said nothing was wrong. The Play listing for the same title had been updated eleven weeks earlier. Same name, same icon, same studio, two completely different states of repair.
I am Dana Fournier. I keep this record at a kitchen table on Barrie Street in Kingston, which is a student town, which means half the people I know are running four-year-old phones on purpose and buying old games because they are cheap and they are finished. Those are exactly the buyers who get burned by a frozen listing.
So the record does one thing. For each game it puts the two store dates next to each other and says how far apart they are.
One honest limit on me: I am not a developer. I read store listings, not source code, and I test on two devices, not twenty. If a store page changes tomorrow, this page is out of date and I will not know until I check again.
Upkeep is the only thing this record measures. It is a maintenance reading, not a review score.
A game with a dead update history can still be the best thing on your phone. Prune has not been touched on iOS since early 2024 and it is still lovely. Upkeep tells you whether someone is likely to fix it when the next OS update breaks something. Those are different questions and this site only answers the second one.
Four readings are used, and the boundaries are arbitrary on purpose so you can argue with them:
In service
Both store listings updated inside the last 12 months, and less than 12 months between them.
Split
More than 12 months between the two store dates. One platform is being maintained, the other is not.
Dormant
Neither store has seen an update in 24 months. Nobody announces this. It just goes quiet.
Out of print
The listing is gone from a platform entirely. You cannot buy it there at any price.
Sources: the "Updated on" line on the Google Play web listing (read with the Canadian storefront), and currentVersionReleaseDate from the public Apple lookup endpoint with country=ca. Minimum OS version and app size come from the same Apple response. App size is iOS only, because Google Play does not publish a size on the web listing and we are not going to invent one.
Where the method stops: recent App Store reviews come from a public RSS feed that frequently returns nothing at all. On four of the six titles here it returned nothing. Where it returned nothing, complaints about a game being broken are marked not confirmed, no matter how many forum posts say otherwise.
Midjiwan AB · turn-based strategy · first released 2016-02-09
In service
Google Play · updated14 days ago
App Store · updated14 days ago
10 years, 5 months oldboth stores shipped on the same daygap 0 months
Play + iOS
2016-02-09 first releasetoday
This is the entry everything else is measured against. A decade after release, both listings carry the same update date, which is not a coincidence: it means one build pipeline, one release, both platforms shipped together. That is the single strongest upkeep signal available from public data.
It also passes the check this record does on monetization, because the check matters to what gets maintained. The tribes you can buy are a fixed-price list, priced item by item, and neither store carries a loot box or randomized-purchase advisory. You can see what you are getting before you pay.
The honest part
The two stores disagree about how good it is: 4.3 in Google Play against 4.61 in the Canadian App Store. That gap is real and it shows in the listings. The current build also requires iOS 15 or newer, so the price of being maintained is that older iPhones stop getting the new version.
Google Play
4.3 stars, 236K reviews, 10M+ installs. Free, in-app purchases listed at $1.39 to $5.99 per item.
App Store (CA)
4.61 stars from 2,313 Canadian ratings. Free. Version 2.17.2.
Download (iOS)
224.7 MB
Age rating
9+ (infrequent or mild cartoon or fantasy violence)
The tribe screen: a plain list of named items, each with its own price. Nothing here is drawn at random, which is why neither store attaches a randomized-purchase advisory.
Entry 02 · one name, two products
BADLAND
HypeHype Oy, originally Frogmind · side-scrolling action · first released 2013-04-04
Split
Google Play · updated5 months ago
App Store · updated52 months ago
13 years, 3 months oldiOS build frozen since Feb 2022gap 47 months
iOSPlay
2013-04-04 first releasetoday
Read the two rows above as two separate apps, because that is what they are. On Android, BADLAND is free, carries ads and in-app purchases, has 50 million-plus installs and was rebuilt within the last five months. On iOS it is a 99 cent premium purchase that has not been touched since February 2022 and still declares a minimum of iOS 8, an operating system from 2014.
That split also explains the thing people argue about online: the ratings do not match. Google Play sits at 3.5 across 1.61 million reviews. The Canadian App Store sits at 4.68 across 338. Those are not two opinions of one game. The paid, ad-free, frozen version is the one people liked; the free, ad-supported, actively maintained version is the one people complain about. Both statements are true at once.
What we could not confirm
The public App Store review feed for this title returned no entries at all, on 2026-07-17 and again on 2026-07-20. So reports that the iOS build misbehaves on current iPhones are not confirmed here. What is confirmed is only this: the last iOS release was 2022-02-24, and it targets a minimum of iOS 8.0.
The same silhouette art ships on both platforms. What differs is everything around it: ads and a live build on one side, a frozen 2022 binary on the other.
Entry 03 · the mirror case
Superbrothers: Sword & Sworcery EP
Capybara Games, Toronto · adventure · first released 2011-03-24
Split
Google Play · updated1 month ago
App Store · updated32 months ago
15 years, 3 months oldAndroid revived, iOS left behindgap 30 months
iOSPlay
2011-03-24 first releasetoday
BADLAND read backwards. Here the Android listing is the one that got attention last month, and the iOS build, which was the original and the celebrated one, has been sitting still since late 2023. The studio is Capybara Games, out of Toronto, which is the closest thing this record has to a hometown entry.
This is also the only title in the record where the complaints are documented rather than assumed. Reading the Canadian App Store review feed on 2026-07-17 returned two one-star reviews, both about the app failing to start rather than about the game:
"Does not work on iPhone 11 Pro Max iOS 16. When launched, distorts when put into landscape mode. Cannot proceed past opening screen."
"Does not work. This is featured by Apple but when you pay and load it, it doesn't work."
No developer replies appear in that feed. Re-reading the same feed on 2026-07-20 returned nothing at all, which is the feed being unreliable, not the reviews being withdrawn.
The honest part
Two reviews are two reviews. The Canadian App Store score, 4.26, rests on 70 ratings in total, which is far too small a sample to mean anything. Treat those quotes as evidence that it failed for two specific people on two specific phones, not as a measured failure rate. Note also the rating: 12+, with advisories for mild fantasy violence, mild references to alcohol, tobacco or drugs, and mild horror themes.
Landscape orientation, which is exactly what both Canadian one-star reviews say breaks on a newer iPhone. We could not reproduce it and we are not claiming it as measured.
Entry 04 · the quiet fade
Prune
Joel McDonald, Polyculture · puzzle · first released 2015-07-23
Split
Google Play · updated9 months ago
App Store · updated29 months ago
Just under 11 years oldboth dates drifting, no announcementgap 20 months
iOSPlay
2015-07-23 first releasetoday
Nothing happened to Prune. That is the entry. There was no shutdown notice, no farewell post, no final update note. One store date slid past a year, then the other, and the app is still on sale at full price on both platforms while nobody is obviously minding it.
It is the cleanest listing in the record by a distance: zero content advisories on either store, rated 4+, no ads, no in-app purchases, one flat price. Read the dates and it is a split by our own definition, twenty months apart, and if the Play side goes quiet for another fifteen months it becomes dormant without anyone saying a word.
This is the situation the record exists for. A frozen listing does not look any different from a maintained one when you are standing in the store page about to pay $6.99. The only tell is the date, and it is in small type near the bottom.
What we could not confirm
The Canadian App Store review feed returned nothing for this title, so we have no recent reader complaints to weigh and none are claimed. The 4.12 App Store score rests on 51 Canadian ratings, which is close to meaningless as a number. Whether it still runs correctly on a 2026 iPhone is not confirmed.
Still on sale, still beautiful, still $6.99 on both stores. The only thing that has changed is the date underneath it.
Entry 05 · the counterexample
Osmos
Hemisphere Games · physics puzzle · first released 2010-08-05
In service
Google Play · updated3 months ago
App Store · updated9 months ago
15 years, 11 months oldthe oldest thing here, and still patchedgap 5 months
iOSPlay
2010-08-05 first releasetoday
The argument that old means abandoned dies here. Osmos went on sale in August 2010, four months after the first iPad existed, and both of its store listings have been touched inside the last nine months. Nothing about it is new. Somebody is simply still doing the work.
It is also, at 46.7 MB, by far the smallest download in the record. Polytopia is roughly five times its size, Sword & Sworcery nine times. Whatever maintenance costs, it is not being spent on making the file bigger.
The honest part
The price is not the same on both stores: $6.99 on Google Play, $4.99 in the Canadian App Store. Same game, different tills. And the App Store score of 4.77 comes from 145 Canadian ratings, which is small; the 4.6 on Google Play sits on 92K reviews and is the number worth trusting.
A 2010 release, rebuilt on Android three months ago and on iOS nine months ago, in 46.7 MB.
The service board04
Six lifetimes on one calendar.
Each line starts at the game's first release and ends today. The blue mark is the last Google Play update, the dark mark is the last App Store update, and the pink hatched stretch between them is the part where the two stores disagreed about whether this game was still being worked on.
Read left to right for age, and read the hatching for neglect. They are not the same thing, which is the entire point.
20102014201820222026
The Battle of Polytopiain service · gap 0 mo
Osmosin service · gap 5 mo
Prunesplit · gap 20 mo
Sword & Sworcery EPsplit · gap 30 mo
BADLANDsplit · gap 47 mo
Super Hexagonout of print · no Play row
First releaseLast Google Play updateLast App Store updateDistance between the two
Super Hexagon's line simply stops in 2018 and never reaches today, and it has no Google Play mark at all, because there is no Google Play listing to read. An empty row is data. We are not going to fill it in with an estimate.
Still in service05
If you are buying one of these today, buy Polytopia.
Not because it is the best game in the record. Because it is the only one where both stores shipped on the same day, and that is the closest thing to a promise that public data can give you.
Osmos has the better story and Sworcery has the better art. Polytopia has the thing this record actually measures: a decade of parallel releases, right up to two weeks ago. Buy Sworcery on iOS and you are buying a 2023 binary. Buy BADLAND on iOS and you are buying a 2022 one. Buy Polytopia on either platform and you are buying the version the studio shipped this month.
The trade is that it is the free-to-start entry here, and the crowd on Google Play is the least happy crowd in the record at 4.3. Sustained maintenance and universal affection are apparently not the same thing.
Every entry above assumes a game still has two listings to compare. Sometimes it loses one, and then the question is not how well it is maintained but whether you can buy it at all.
Super Hexagon
Distractionware, Terry Cavanagh, music by Chipzel · first released 2012-08-31
Google Play · updatedpage returns 404, checked 2026-07-20
App Store · updated98 months ago
13 years, 10 months oldone platform gone, the other asleepno second date to compare
iOS
2012-08-31 first releasetoday
The Android version was pulled from Google Play by the developer himself, because it had stopped working properly on newer Android devices and he judged it better to remove it than to keep selling something broken. It is now sold directly through itch.io instead. That is a decision made by an author who cared, forced by a platform that moved.
And here is the part that makes it the right closing entry: the iOS version, which nobody removed, has not been updated since May 2018. It still declares a minimum of iOS 8. So one platform lost the game to a store decision and the other kept a copy that has been sitting untouched for over eight years. Neither outcome had anything to do with whether the game is good. It is superb.
Why there is only one button below
This record links official store listings and nothing else, so there is no button for the Android version. The itch.io route exists and is the developer's own, but it is a direct download rather than a store listing, and we are not going to blur that line.
Thirty-four megabytes of geometry that outlived its own Android store page. The iOS build shown here has not been rebuilt since 2018.
What pays for this07
Nobody pays for it, which is why it is short.
The site is free to read and there is no paywall, no membership and no subscription tier planned. The store links above are ordinary links to the official Google Play and App Store pages. They carry no affiliate tag, no referral code and no tracking parameter, and they earn nothing. If that ever changes, the disclosure goes here first and every affected link gets marked rel="sponsored nofollow" so you can see it in the page source.
No developer has paid to appear in the record, and no developer can. Games are chosen because their update histories say something, and two of the five entries are here because their histories look bad. Nobody buys that.
Running costs are a domain and hosting, paid out of pocket. There is no company, no incorporation and no advertising on these pages.
Games are also excluded on fixed grounds, and it costs us good entries. Stardew Valley has one of the finest update histories on mobile and cannot appear here, because its store listing carries a simulated gambling advisory and this record does not send readers toward that, however good the game is.
Follow the record08
Dates go stale. That is the whole problem.
Everything on this page was true on 2026-07-20 and some of it will not be true in six weeks. When an entry changes state, when a gap closes or a listing disappears, we send one short notice.
Roughly one message a month. Usually less. Never a sales message, because there is nothing to sell.
Your name and address go to OneSignal, Inc. in the United States, which handles email and push delivery for this site on our behalf. Consent is the basis for it and you can withdraw it at any time using the unsubscribe link or by writing to [email protected]. The OneSignal script also registers a service worker and stores identifiers in your browser. Details, including cross-border transfer and how to complain, are in the privacy notice and the cookies and storage notice. We do not ask for a phone number.
Questions09
Does old mean broken?
No, and the record is arranged to prove it. Osmos is sixteen years old and both its stores were updated inside the last nine months. Prune stopped being updated on iOS in early 2024 and still works fine as far as we know. Age tells you nothing on its own. What the dates tell you is whether anyone is likely to be there when the next OS release breaks something, which matters most if you are about to pay for a game you plan to keep.
Why is one store years behind the other?
Usually because they are not the same product. BADLAND is the clearest case in the record: the Android version is free with ads and gets rebuilt, the iOS version is a 99 cent premium app that was finished in 2022. Different revenue, different maintenance budget, different fate. Sometimes it is simpler than that, and one platform's build just costs more to keep certified than it earns back.
Where exactly do the dates come from?
The Android date is the "Updated on" line on the Google Play web listing, read with the Canadian storefront. The iOS date is currentVersionReleaseDate from Apple's public lookup endpoint with country=ca. Both were read on 2026-07-20 and both are shown as ISO dates on this page so there is no ambiguity about day and month. Store pages change without notice; if you find a date here that no longer matches, tell us and it gets corrected.
Do developers pay you, or can they buy an entry?
No to both. Nobody has offered, and if someone does the answer will still be no. Two of the five entries here describe maintenance situations that no studio would pay to have described, which is roughly the point of keeping the record independent. The store links carry no affiliate codes and earn nothing.
You quote two angry reviews for one game and none for the others. Why?
Because the public App Store review feed is unreliable and we will not fill the gap with forum posts. It returned entries for Sword & Sworcery on 2026-07-17 and returned nothing for BADLAND, Prune, or for Sworcery again three days later. Where it returned nothing, the entry says so and claims nothing. It reads as inconsistent because the source is inconsistent, and hiding that would be worse.
Is Upkeep a score out of ten?
It is not a score at all, and it is deliberately not numeric so it cannot be read as one. There are four readings: in service, split, dormant, out of print. They describe maintenance and nothing else. A game can be dormant and still be the best forty minutes you will spend this month.
