Ask HN: What Happened to iOS Widgets?
There seems to be a fairly low uptake in developers creating widgets for iOS since the update that let us place widgets on our home screens. Is anybody in the know regarding why are developers not jumping to develop them?
64 comments
[ 6.2 ms ] story [ 78.9 ms ] threadAnd it turns out nobody really cares. Everyone’s used to opening the app when they need anything. And I say this with 5 widgets covering my only Home Screen - not bashing them in any way.
I've got a calendar widget and a messaging widget. The calendar is in 'agenda' mode, so it shows me the next two or three things I'm supposed to do, and I can scroll to see more if I want. Almost everything I need from calendaring without loading the whole app. (Sounds like iOS might not allow scrolling, so then you need to use more screen, which is iffy)
For messaging, I've currently got it showing all messages (from one app), in case I dismissed a push and forgot about it. I used to have a widget showing just messages from my spouse, because I had a lot more messaging going on (used for work) and it was easy for things to get lost. Maybe I could be more organized, but that ship has sailed.
Weather seems like a nice option, but I'm ok with being surprised by unexpected weather.
I rather enjoyed the Windows Phone live tiles, where if you went to the home screen it'd show you all the icons, then if you sat there, things would flip and you'd get tidbits from everything. Widgets aren't quite the same.
I think taps are supported, I don't even know if you can scroll in a widget?
The reason for this is supposedly battery life. For the bigger widgets on an iPhone that feels a little over the top, but the widgets system underpins the lock screen widgets which are now possibly always on, and the watchOS "complications"/widgets, which can also be always on. Because of that I can understand it.
The Android one I can: Check tasks I've done, scroll the list of tasks, select individual tasks, change the view (which opens a selection dialog), open the configured view in Todoist, and add a new task to the configured view (which opens a small UI over my home screen).
The iOS one allows you to long press to change some settings, select individual tasks (which opens Todoist completely), open the configured view in Todoist, and add a new task to the configured view (which opens Todoist completely). It also has __very__ limited options for where you can put it on top of the already-limited sizing options. I can either put it at the very top of the screen, the middle of the screen (2 rows of apps above, 2 rows below), of the very bottom of the screen. While the Android one can be resized to basically whatever and put anywhere. You can also only see 3 total tasks in the widget since you can't change the size, the text size, and you can't scroll. I might as well just click the app because it offers __very__ little that just tapping the app doesn't, or notifications don't.
No wonder not many widgets are being adopted for iOS. They're extremely limited. And that's all on top of the point another comment was making about people making websites into apps with React Native or a wrapper which makes it really hard to create widgets, especially with things like heavy memory/CPU time limitations placed on widgets in iOS making it hard to be able to boot up the app and get enough data to create a widget with inside those limitations. Let alone making the iOS-specific APIs and whatnot available to your app.
Personally I find that the apps I want widgets for, have them. Personal Finance (Copilot), Food Tracking (LifeSum), Task Tracking (Things and TickTick), plus the built in ones cover all my use cases.
I also just went and looked to see what other apps I have installed that have one and I see other ones like Twitch, Youtube, Ritual, Reddit, Mela.
I think most apps just don't need widgets, really just things that I want to have a quick glance to and the vast majority of my apps don't fall in that category.
Edit: To clarify, I have yet to go looking for a widget for a specific app had not find one that does what I want.
Outside of that, I really like the app. I wish there was an iPad app, I think they should have done that before a Mac app. But the Mac app is nice.
It does what I need for it to, support is responsive but when its a syncing issue they are limited in what they can actually do.
Also if you use an Apple Card it isnt real time syncing like Mint you have to import it after the month is over... which doesnt really help with budgeting.
I have a feeling they used to be more popular, maybe around 4.0 when people seemed more into arguing about iOS vs Android like it mattered what other people used, but I don't use any or see people use any now. (Although installing Lineage on my new-to-me phone it came with a clock widget by default, which I might keep.)
So I would guess the developers of major apps have metrics on how few use them on their Android version, and don't see the point/it's lower priority than whatever else they're doing.
I find it super handy to be able to access the contact QR code without fiddling with lock screens and finding the app.
[0] https://blinq.me/
I can get most weather information I need from the MyRadar widget; rarely do I need to open the app. If I do, it’s mostly to check weather in other areas like my parent’s home during hurricane season.
Likewise with the Kasa widget. The only time I open the app these days is to setup a new device or an automation/scene.
Useful widgets exist, but I imagine their use cases are fairly limited to mostly trivial things. For more complex applications, it probably simply isn’t worth the dev time to create and maintain a widget.
I don't think you can fix 2. by fixing 1. - that is a common topic of discussion that I believe to be pointless, seeing as I very seldom (if ever) used them in Android as well.
I have about half of my iPad's home screen taken up with weather and calendar widgets, and that is about the most extensive use I make of them. On the iPhone I only use a few on the lock screen (for similar purposes).
From analytics (mine [0] and friends) on various iOS apps, around 3-8% of active users use the widget.
So this answers the question partially = only a small percent of people use widgets. Hence low priority.
iOS widget use depends on the app category also. E.g
- productivity apps have higher widget usage because they have power users, and often users already think in workflows & quick access
- more general users don't care or don't know about widgets at all
That being said, regardless of the category, I always encourage companies to add it. Why?
- the cost is much lower than they usually think
- improves retention
- higher potential of forming a habit
[0] - neuracache.com
There are even dedicated custom Widget apps like _David Smith’s WidgetSmith which does fairly well for itself once people realized they could have a more unique Home Screen with custom widgets.
iOS Widgets just don’t do a whole lot right now. The current widget API basically skewed away from Dashboard-style widgets and made them more like the Windows 8 Tiles but better: fancy app icons that can display dynamically changing information with some deep-linking capabilities; with the non-dynamic fallback being to just use the App’s existing icon.
When the whole point of the widget or watch app is too show the status of something and it doesn’t update consistently, it’s basically useless.
So in short the problem is that most of them don’t update consistently which makes the whole category feel useless.
But other things come into play: if the app was not launched for a while, iOS schedules the tasks less and less often. Same thing happens if low power mode is used or if memory is needed for other apps.
Anyways…discoverability?
1. If you are on the normal Home Screen on an iPhone, where is the affordance to even suggest widgets exist? There really isn’t, unless you enter the modal jiggly mode to rearrange apps, where you get a subtle + button to add widgets.
2. If you are on the home view and swipe all the way left to get to your “stack” of widgets (which I do use), you get a weeeee little Edit button down at the bottom to add new widgets.
3. If you add a new app, is there a universal affordance to know there’s a widget available? Not really.
So, for me, I just kinda forget they are there, or at least I am forgetful enough to check if there is new functionality available for me.
(And if I am missing an obvious affordance, please, make me look dumb here. I’d be glad to learn more.)
-
I will say I am getting more use out of the…oh wow we really are keeping that name…the Dynamic Island with Live Activities feature.
https://www.macrumors.com/guide/dynamic-island/
Barely any Live Activities out there, but that omni-present and slightly animated subview lets me know there are fresh things to interact with.
Screenshot here: https://f.alinpanaitiu.com/mt37aX/Image.png
While they're easy to write because of the new SwiftUI APIs, they're indeed very limited.
All I am able to do is add buttons which when pressed, they open an URL.
I managed to make that URL useful in Volum with a handler that extracts the device id from the URL and toggles the device power. Not always working because the app might not be in sync with the current state of the lights/speakers, as apps are not allowed to do much work in the background.
In the Sub Sol case, the URL only navigates to the party that you tapped on in the widget.
But because widgets live in a separate extension, completely isolated from your app's code, you can't do anything inside them. Even the rudimentary stuff needs code duplication between the app and the widget extension.
The good thing is that it leads to a very small widget extension binary that doesn't hog up memory, and I think that's what Apple wanted in the end. But I would love to have a bit more power so I don't have to open an app for stuff that I do often in 1-3 taps.
I keep a four-icon sized one in the top right of my home screen. It regularly delivers special photos and ”Memories” that I really like.
Other than that, widgets seem like Apple Watch complications. Good if set up just right, but fidgety to configure to taste and lacking in depth.