It's not the iPhone, it's just a recent, fancy phone versus an out-of-date or cheap one.
I refuse to pay $1K (Canadabucks) for something that could be destroyed in an instant if I happen to drop it or sit on it. I get whatever low-end phone Costco is pushing at the moment. The current one is a Moto G Play 2023 that cost me all of CAD$140 on special. It works fine, but bloat can catch up to anything.
I use an app called Flashfood. It was fine. Then they restyled it. And whatever else they did, on my phone it is now so glacially slow that I don't use it as much any more. They'll never know, because they test it on $1K phones where you can waste so much compute on a glorified screen refresh. The actual functionality could have been done with a 2010 vintage phone, if coded efficiently.
They should be testing on low-end phones. Because a significant percentage of the user base has them. Somewhere in a fiction novel I read about a music industry executive who would listen to new material on a crappy little record player, not on super fancy studio monitor equipment. Because that's what it had to sound good on.
> Somewhere in a fiction novel I read about a music industry executive who would listen to new material on a crappy little record player, not on super fancy studio monitor equipment
Tangent, but the Yamaha NS10, the de facto studio monitor, is used precisely because if you can make it sound good on those it will sound good on anything. They're awful.
This is utterly fascinating, and something I absolutely would never have encountered or suspected without your comment. Thanks!
And, your comment is without exaggeration, as per the wikipedia article on the speaker: "it may lead to listener fatigue with prolonged use in the domestic setting"
Another common mixing/mastering speaker back in the day, not sure about now, was the Auratone Sound Cube and its descendants -- tiny enclosure with a single driver, not full range, but "unforgiving midrange".
Totally understand your position. For what it's worth, a lot of the higher end phones these days are much more durable. I have a handful of years out of date iphone and am not careful with it. It's been through a lot. I have always had a hard silicon case on it (seems dumb to not use a case), but the phone itself genuinely still looks brand new.
Also if you buy a few generations old refurbished, it's only a few hundred, not $1k.
All this being said, of course apps should run on low-mid end hardware. They're still wildly powerful devices.
Fair enough, but disasters still happen. Someone in my family dropped a fairly late model iphone into a lake. They did recover it the next day, but by then it had leaked, i.e. it was deeper than the pressure it was waterproofed for.
A friend had hers snatched by thieves. Gone in an instant. I don't in general need the power (or camera) of an expensive phone (except for the app mentioned). And I'd rather have a recent, lower-end one than an older higher-end one, having been burned a few times by apps "not compatible with your device" due to the OS being too old, admittedly only on older iDevices that the kids play with or truly ancient Android tablets.
Oh, and two iPhone 7 vintage devices that were given to us as kids' playthings a few years ago (both with forgotten activation locks, so useless) were absolutely thrashed. They are tough but not infinitely so.
I witnessed exactly this at work at a prior company.
Odds were good that each year Apple would reserve space for or specifically invite (one year onstage) folks from the company to WWDC. And each year the founder of the company and usually at least 2-3 developers would buy the latest iPhone or Mac, and that would become the new development flagship. Sometimes they’d get advance access to new hardware, and they’d always take advantage of face to face time with Apple’s platform developers.
Now it made sense for new features like Touch ID, then Face ID. There was good reason to test that support on real hardware. But we’re talking every year.
I try to use my devices until they’re no longer in support at least. Sometimes I’ll upgrade outside that cadence, but rarely.
I learned the impact that has on the end user real quick. In 2018 I was rocking a 2015 MacBook Pro that was not maxed out on RAM and CPU, and our flagship product ran like shit on it because all the developers were running new MacBook Pros with as much RAM and CPU as they could buy.
Interestingly, Google uses and emulated Moto G (Power) for site speed testing in their pagesight/lighthouse tool.
> Captured at Jan 9, 2025 at 11:23 AM EST
Emulated Moto G Power with Lighthouse 12.2.3
Single page sessionThis data is taken from a single page session, as opposed to field data summarizing many sessions.
Initial page load
Slow 4G throttling
Using HeadlessChromium 131.0.6778.139 with lr
>Somewhere in a fiction novel I read about a music industry executive who would listen to new material on a crappy little record player, not on super fancy studio monitor equipment. Because that's what it had to sound good on.
Marissa Mayer, while at Google, kept using dialup at home so she would experience what someone without broadband sees.
I'm not justifying badly-optimized sites/apps but frankly I don't expect an entry level 4 cylinder gas-powered car to perform the same as a high-end EV.
Me neither, and if I wanted to play the fanciest mobile games, or something else that truly needs the compute, I'd get the tool to do it. A simple app that provably worked fine on a low-end device being ruined by added bloat just because they restyled the visuals still sucks. Especially because the app (Flashfood) is for budget conscious shoppers, who may disproportionately have lower-end devices.
Bad connections deserve more attention. Slower devices too, but at least they get some attention even if too many people ignore them. The amount of code that can’t handle brief transient network errors is crazy and I hardly see it discussed. You don’t have to be in the middle of nowhere, either. I’m in dense suburbs, with gigabit fiber and pretty fast 5G, and apps will still fall apart if I’m using them while I walk away from my house as the device tries to decide if the weakening WiFi signal justifies switching to cellular yet.
I don't understand, isn't one of the main points of the article that even older iPhones (like the four year old iPhone 12) are still vastly more powerful than what most Android users have? There are numbers and graphs that show that web pages load significantly faster for iPhone users than for Android users.
Ticketmaster now requires a smartphone for some tickets/venues:
"Delivery - Mobile - Free -
Your phone's your ticket. Locate your tickets in your account - or in your app. When you go mobile, your tickets will not be emailed to you or available for print."
I occasionally help a coworker buy tickets as he has a flip phone and no computer. Yesterday, the above was the only delivery option, so I couldn't help him. He was able to buy the tickets at the venue, but that is slowly going away as well.
Those of us who learned computing mostly trough it's limitations are building software considering for them, even if mostly not necessary. Other just use electron for a command line wrapper.
And don't forget, Lighthouse metrics are just a starting point. Add specific metrics to your application. Here's an easy one: how long does it take for your mobile navigation hamburger button to be clickable?
> Are you going beyond measuring a high level p75 and looking at the detail beneath?
>If an engineer is frustrated at the responsiveness of an ‘Add to cart’ button click during development, then they’re probably more likely to investigate why and resolve it.
IMO this should be in QA/SDET's realm (ideally via automated testing). If you're rich maybe you have one or a few SWEs devoted to performance. This will force other people within the company to also think about these metrics. Don't be a siloed developer quietly fixing all of these things for free just because it's slow on your machine.
True but it is very difficult for app developers to make apps for such wide range of devices. With iphone you can just test on the oldest supported device and you are good with android there is so much variation that you can't be sure even if you are targeting latest version.
What is worse is that apple has handicapped safari from taking full advantage of its HW and does not allows other browser-makers to do it either this results in inferior experience for ios owners using webapps and web app developers are forced to release their apps on iphone.
The article states that browsing on the top of the line Android phones are about the speed of a four year old phone.
If other browser makers can’t make a performant browser on Android, what makes you think either Google or Firefox is going to make a faster one for IOS?
The article states that the inferior experience is on Android. If anything you need to make an app for Android because of how bad browsers perform even on top end phones
> If other browser makers can’t make a performant browser on Android, what makes you think either Google or Firefox is going to make a faster one for IOS?
I interpreted this as you saying that Chrome wouldn’t be faster on iOS whereas I’m pretty sure it will
Im sure some sites could use performance loading as a filter of sorts: if you want to keep a certain economic class out, make the website require a fairly high performant (read: modern and expensive) device.
Irony here too is that the linked site features a weird display of the author’s image and bio that takes up the entire screen and more on iPhone. It took me a couple seconds to realize it wasn’t a focus of the page.
After years of being iPhone-only, my site recently released an Android app. One month in, and we received a review that simply says "asking money"—unfortunately, the stereotypes seem to hold true. At this rate, we might break even on Android if we're lucky.
We framed some 1 star reviews that were essays of how terrible we were for asking for money for the app. All caps “this is a deceitful bait and switch how dare you ask for money! From the bottom of my heart I stab at thee...” for two pages type of ranting.
How lightweight are the Real User Monitoring solutions the article suggests? I'd worry it might be somewhat counterproductive to fight software bloat by adding even more fingerprinting and tracking...
Many of the RUM solutions don’t collect personal data or use fingerprinting
Yes they see the visitors IP address (but often don’t store it), they’ll often drop a cookie so pages can be tied together as a session but generally they’re bot interested in tracking people across sites or long time periods
Okay, first of all, distill this down to a 2 minute preso that's heavy on the ROI, because many organizations aren't going to let go of the eye candy and—usually the worse offenders—the legion of third-party assets. Not to mention decision makers are too often simply not empathetic to such real world details.
Second, minimum reqs are/were a thing. I'm not necessarily arguing for or against anything, but at some point... ¯\_(ツ)_/¯
Lastly, people using badly performing devices (not always inherently a hardware issue) are probably accustomed to things not working smoothly. You can only do so much.
I think focussing on Android vs iOS might belie the true point which is not everyone has a 'good' phone.
Almost any currently in use iPhone has a 'pretty good' processor and probably has up to date software. Even the iPhone 8 still holds up pretty well.
On the Android side you get a wider spread of devices. I imagine a Pixel 8 or S23 compare pretty well to 'iOS'. The problem is old, low powered budget phones. I wonder how much of the problem one those phones is specific to the tech we use to build sites, and how much is the crappy hardware.
I also don't imagine the expected value of a potential customers on a flagship iPhone or Pixel device is the same as a potential customer using a Huawei P20.
“Everybody worth transacting with” is more accurate. Makes sense for business, but a shame when government (or an essential business that should be a government utility) does it.
Well as software engineers there’s not as much we can do for people who don’t have computing devices, but it’s still important that information be accessible without them.
Well, people can have laptops, or "unsupported" phones (i.e. non-iPhone/Android).
Locking all sorts of basic stuff to two for-profit closed platforms sounds rather non-ideal to me. It's also quite a shift from stuff being web-based, where in principle you or I could sit down and write a compatible browser and/or port an existing browser (some work, but very doable).
It's not all that different from locking everything to Windows and/or Internet Explorer like it was 20 years ago, except worse because it's so much more pervasive now. The old "mandatory Microsoft tax" got replaced with a "mandatory Google/Apple tax". Any startup has basically zero chance of entering the market.
Probably a well supported postal service as public infrastructure!
Yes, you can make an app or sell a widget without supporting transactions via mail, but many companies providing critical services are starting to require cellphones to participate.
My country has a large religious demographic with dumbphones. I'm glad to have them, as I have a smartphone but I'm not about to install an app for banking, an app for charging the car, an app for the supermarket, etc etc.
The problem isn’t older phones FTA they show that there is a 60% difference in flagships. I agree with you that the latest pixel has a huge disadvantage to a Samsung phone .
The bigger problem is how much money you can budget for supporting every phone and which platform is more profitable. This is a video game but, Warframe was ported to iOS and then to Android. If this isn’t an indicator that you will make more money than what is ?
> I imagine a Pixel 8 or S23 compare pretty well to 'iOS’
In fact from the article it doesn’t. Modern Android phones have the performance of a 4 year old iPhone.
When the new iPhone SE is introduced this year - Apples low end $400 phone - I wouldn’t be surprised if it is faster than high end Android phones. It will probably be based on the iPhone 16 SOC.
> In fact, your looking at the iPhone 12 until you find a previous version of the iPhone that scores less than todays fastest Samsung Android device. To add a little more perspective, this device was first sold over 4 years ago!
This is a fair point, even an optimistic read for Android has their best phones sitting somewhere on the lower end of the iPhone distribution.
I wonder if the relationship between Geekbench score and TTI etc. are linear though. My guess is the areas that slow Android down are crappy phones that are freezing up from lack of RAM lack of headroom in processing power due to system overhead. ie any phone above a certain threshold score is probably fine.
He is comparing high end phones though and traditionally Android phones come with more RAM than iPhones.
If a low end hypothetical iPhone SE comes out this year with the iPhone 16 SOC, there will be no iPhone that is as slow as the fastest Android phone.
The iPhone 14 is the slowest iPhone Apple sells and soon won’t be selling that or the SE in the EU at least because they both use Lightning instead of USB.
My point is that in the graph showing Geekbench score and INP [1] you can draw a straight line, but what I see in the graph are three clusters – phones with a Geekbench score:
- above 1,500 (look very good)
- between 500 and 1,500 (look decent still)
- below 500 (really terrible)
Both an iPhone and a flagship Android will have a much better INP than a phone with a Geekbench score of 500. I'm not so sure if an iPhone has a much better INP score than a flagship Android though (maybe it does, I'm not seeing that in the data though).
It’s lack of / small size of processor caches on the chips that are used Android phones that’s the big issue — Alex Russell has a good post about this somewhere
Yeah, Apple started pulling away from Android with the 64 bit transition that caught everyone by surprise, and the Android performance has simply never caught up again.
The halcyon days of the Qualcomm Krait (which certainly had some problems) are very long gone by now.
Oh really? It takes a few more seconds to add a can of beans to your cart - who cares?
The thing that really winds me up is forcing you to use a mobile for 2FA, when it would be easy to support a landline as well, which thankfully my bank does.
Personally, I don't understand why people buy iPhones. If you're in the US, I can understand the texting and video calls lock-in, but even then, I'd buy an Android just because I'd refuse to be a hostage to Apple's services, and I'd wear that green badge with pride.
Furthermore, paying the price of the latest iPhone Pro (or Samsung S Ultra for that matter) is ridiculous and is only acceptable to people with high prices on their contracts. Much like what US people are going through — i.e., if you're paying $100+ on a mobile plan, might as well throw an iPhone in the mix to sweeten the deal. But on my 5 EUR unlimited data plan, me and others do not want an expensive phone included and prefer to pay for that phone up front.
I also prefer to up front whenever possible and never had a phone paid by contract.
I buy an iPhone preferably used and use it until major OS updates are no longer available. I currently use an iPhone XR from 2018.
I think that long term with iPhone it’s cheaper than doing the same with Android unless you go with really budget ones like 150 €, but that’s to bad of an experience for me.
What I also really like with iPhone is that you can get really cheap displays and batteries which is not the case for Android phones in my experience.
Think about what you're saying here: the signal of someone being an Android user instead of an iOS one means they're willing to sacrifice their own user experience for (not that much) money and being tied to Google.
As a company do you want people like that as your customer?
Tesco is a low-mid end supermarket chain in the UK, so this is kind of irrelevant as they need to hit everyone, but in such circumstances people should not complain if the UX sucks on their device because they were excessively cheap, and yet, of course, they do. The truth is the numbers in the article are unacceptably slow across the board.
You can have alternative app stores, you can download torrents on it and then play them with VLC, you can have the actual Firefox browser and even Chrome clones (e.g., Brave) are more capable than iOS Safari web views. You can have an email client that works locally (K-9 Mail). Android has had usable PWA support for years. The most popular chat app, WhatsApp, is not Google's app. Samsung phones don't even push people into many of Google's apps, shipping with Samsung's own launcher, Calendar, Notes, Camera, Gallery, etc., and it even has their own apps store in addition to Google Play. And AOSP, the open-source distribution, is actually usable even without Google's Play Services, which has spawned distributions such as LineageOS breathing life into old devices.
I'm an Android user, and except for the Play App Store and Google Maps, I don't use any other apps from Google and that hasn't happened due to some attempt for "degoogling". Even the app that syncs my CalDAV/CardDAV accounts is not developed by Google (on Android this isn't native, but that's completely fine).
I don't understand why many iPhone users resort to such mental gymnastics to justify an expensive device that's being rented from Apple.
And yes, as an app developer, you want Android users because Android (and its allies, like WhatsApp) are actually winning in the marketplace. Providing clients for all people is how WhatsApp came to dominate the landscape. This is why Apple is so reluctant to give up on their nefarious practices related to their App Store or the restrictions they place on developers, continuously trying to undermine the EU's DMA because they know they only have a short window during which they can milk the user base.
Is this really true? In terms of monetary value, I am not sure what the deal is, but the author of this article works for Tesco, and the Tesco app is free on both platforms. In the UK, the market share of Android devices is much higher than in the US. I am an Android user who would almost certainly file useful bug reports.
It wasn't meant to be a moral judgement on iPhone vs. Android users. But Apple clearly targets the upper income demographic. None of their products are cheap. It's pretty frequently advised here to "raise your rates" or "charge more" in part because once you cut off the customers who are bottom-feeders, a whole bunch of problems go away.
A whole bunch of problems go away, but so do a whole bunch of solutions for people who might need them.
Disadvantaged people deserve a chance to have solutions to their problems too. If we want a society that’s worth being a part of, we’d do well to consider those people.
The fact that you equate people who can’t afford an iPhone to “bottom feeders” is quite telling.
Eh. If one wanted to do charity for the less fortunate, it might be better to just work for or give to bona fide charitable causes - rather than operating a marginal, possibly unprofitable Android app.
I own a Samsung S24 Ultra, by choice. Would not consider myself a charity case.
I got my first smartphone years and years ago, and it was an android. Then they came out with the iPhone 3GS, and I briefly had that. Didn't like it, traded for an Android. Was ready to experiment back then, even briefly ran Windows Phone flashed on a HTC device. But inertia had its way with me, and I've been sticking with Android ever since. I have all the apps I need. I have my ages-old "weird" workflows set up, like keepass synced through OneDrive. I run an andblocker on firefox. Moving to an iPhone would be just so much pain for no real gain.
If you don't provide an Android app, well, provide me with a burner iPhone if you want me to use your service.
What happens when your general approach (ignore the margins) means the "bona fide charitable causes" can't operate because they all require top-of-the-line equipment to even be functional?
Software is already shite, stop trying to justify making it more shite.
This is missing the point. Shouldn't software generally work well for the people who use it? Or are we really just saying out loud that we build software for our own enrichment and the users are just a means to that end?
I view maintaining android compatibility as a part of a vendor mitigation strategy, avoiding the creep of tight coupling that can make adjusting to market shifts an intractable problem.
If you are only focused on the US market and have a paid app, maybe.
Outside the US, Android is the dominant platform by a wide margin and iOS support is lower priority. And for many services you can't just exclude users anyway, be it for legal reasons or because network effects.
Sadly this rings true. Of all the apps and games we've ported to Android the cost has never been worth it. The games were pirated more than purchased and across both apps and games our support tickets went way up.
It's the Tesco app we're talking about, so none of that even applies.
Saying that half the population somehow doesn't matter for one of the country's largest supermarket chains is completely mental on every level and is not a serious viewpoint.
It's also just harder to support because of the sheer fragmentation.
I give my engineers different Android phones as their primary development devices and yet the weird Android issues that keep cropping up are near constant.
This is difficult, because the mess that is mobile(?) (Android?) development could be better in general, but that also sounds like your software just is poorly designed...
When it's a "works for me" on one platform but falls over with different hardware, that usually points to some serious issues (e.g. threading, contention, false assumptions) in your code/app.
Not to say that there probably aren't buggy areas in Android itself, however with an appropriate test budget you could determine which are likely hardware (different phones) vs OS (common across devices, with simple repro apps) vs your software (you have bugzorz).
In our case, the weird Android issues we encountered a lot lately has been due to the Jetpack Compose’s bugs. It’s amazing how buggy that framework is.
And exactly this is the reason behind Apple users are being more profitable on average.
iPhone users contains a bigger percentage of people who don’t have the attitude „I only spent enough to get it usable“ and are willing to pay for getting better quality.
I refuse to pay $1K (Canadabucks) for something that could be destroyed in an instant if I happen to drop it or sit on it. I get whatever low-end phone Costco is pushing at the moment. The current one is a Moto G Play 2023 that cost me all of CAD$140 on special. It works fine, but bloat can catch up to anything.
I use an app called Flashfood. It was fine. Then they restyled it. And whatever else they did, on my phone it is now so glacially slow that I don't use it as much any more. They'll never know, because they test it on $1K phones where you can waste so much compute on a glorified screen refresh. The actual functionality could have been done with a 2010 vintage phone, if coded efficiently.
They should be testing on low-end phones. Because a significant percentage of the user base has them. Somewhere in a fiction novel I read about a music industry executive who would listen to new material on a crappy little record player, not on super fancy studio monitor equipment. Because that's what it had to sound good on.