[oferzelig]

The short business answer is that it's currently not a metric that influences people's decision to install or uninstall an app. If it were one, we would've seen an engineering effort to reduce bloatware and duplications as seen in the post.

[danpalmer]

On the contrary, lack of storage is a huge limitation on installations. This may not be worth optimising 50MB to 30MB, but optimising 500MB to 300MB would be very impactful to installation rate.

People love taking photos on their phones, they love big games, and Apple are notoriously stingy with storage space and iCloud storage to ship off device. It is a very significant portion of the population who are borderline out of storage all the time. I work quite closely to this sort of stuff on a regular basis.

[jasode]

>, but optimising 500MB to 300MB would be very impactful to installation rate. [...] It is a very significant portion of the population who are borderline out of storage all the time. I work quite closely to this sort of stuff on a regular basis.

Does Apple send telemetry information back to developers about failed app installs due to users' device being out of space?

I don't have a current Apple Developer account to test this but the documentation doesn't have any obvious statistic concerning failed installs: https://developer.apple.com/help/app-store-connect/view-app-....

EDIT reply to: >I don't think there's any special permissions required to get free disk space, so presumably he's getting it via telemetry from his app.

I was thinking of new users who didn't have the app at all.

E.g. hypothetical... "We're a brand new YC tech startup. Our app is 500 MB. This bloated size prevents 20% of potential new users from installing the app." <-- Does Apple provide enough stats to make that type of confident correlation?

[ninkendo]

I’m imagining a lot of users find themselves low on space, go to Settings, find what apps are using a lot of it, and delete them. “I’ll re-download LinkedIn if I need it again” they’ll think to themselves. This is terrifying to LinkedIn because it means push notifications no longer deliver, and how are you gonna increase engagement now?

For all I complain about OKR’s, this is a super simple one for LinkedIn to understand: Every MB your app takes up can probably be shown to decrease user stickiness by some percentage. Fix it!

Ah who am I kidding, they’ll probably switch to moving the assets out of the app bundle and making them live-download on first run and stored in the Caches folder somewhere, which will make the app “smaller”. Problem solved on their end, app-size OKR accomplished. The fact that it makes the app slower due to asset fetching is next quarter’s problem.

[danpalmer]

I'm not sure if Apple does, but the Play Store provides plenty of information about size of app, size relative to similar apps, size change over time, installation rates, and basically all the pieces necessary to figure out that size is preventing installs. I have to be careful about what I say here given my role, and this is all opinion or public information, but I'm not sure that Apple's telemetry is particularly insightful in this regard.

In my experience publishing on the Apple App Store, and as a user of iOS, there's likely no material difference between iOS and Android with storage issues, or if there is it's probably in Android's favour with the (mostly historical) availability of expandable storage and Google Photos providing a lot more off-device photo storage.

[paxys]

Apple doesn't incentivize developers to build smaller apps. Instead they push users to throw away their phone and buy one with more storage (at a cost of $100 per 128 GB) and/or buy iCloud subscriptions.

[gcr]

When I’m low on space I don’t even install some apps that I see on the App Store or, more commonly, I install them just to try them out and remove shortly after.

[gruez]

I don't think there's any special permissions required to get free disk space, so presumably he's getting it via telemetry from his app.

[RetroTechie]

^^ This. Companies often don't care about app size, because the (potential) users that avoid their app for size and/or performance reasons, are not the ones with the fat wallets (or valueable data).

Read: company doesn't care about including a large % of potential audience. It only cares about including the users that are most easily turned into profit.

P.S.: maybe those companies are right. But it also means they're fishing in a small & crowded pond. There's a lot of untapped market out there if only a company would care to cater to it.

P.S.2: Things change, and people's circumstances change. Today's 'useless' user could be your company's most valueable asset tomorrow. Get 'm while you can.

[jesterson]

that's right - however the anger is not channeled towards LinkedIn, therefore they can't give two f*cks about the app size

[ksec]

I have ~30 of either Banks or Payments Apps and Social Media Apps. Every single one of them are 500MB+ in size. That is 15GB of Apps. My HSBC Apps is 800MBs and I dont have a fucking idea why. And then there are Gmail, Google, Outlook all 300MB+.

To me it is just disgusting.

[JeremyNT]

Although I think you're exactly right, the irony is that they'll never actually be able to determine why somebody doesn't install an app due to this kind of junk. It's not like they can a/b test "bloated app" versus "good app" installation metrics, and they have no way to know why people are not installing something.

The trend towards huge installations is one of the many reasons I install as few apps as possible, especially in this class of "should have a fully functional web version" type social media things.

The damage is also cumulative: I have a perception that "apps" in general are likely to be huge piles of junk, so I'm reluctant to install any of them, regardless of their own merits. Even if LinkedIn were to spend the time improving their app it would suffer from the overall reputational damage to the entire ecosystem, and I still wouldn't even think to install it.

[eddd-ddde]

When I was in an internship, I needed a new library for some feature in one of the sites. The _first_ thing their PM objected on was bundle size. Only after confirming that adding my feature would bundle size remain sensible could I continue.

They are very sensitive over this kind of bloat over there, and it is definitely a key metric. You will even get a warning in your email if one of your builds increases bundle size too much!

[SkyPuncher]

I'd conjecture even further. The most valuable users of LinkedIn, those with in-demand skills, are more likely to be running a high-spec'd device where they don't care about these issues.

On my phone, 500MB would put you as ~30 largest app on my phone. Those largest 30 include, messages (large attachments), photos app, music/media (downloaded for offline play) and a variety of junk games I never play.

I'd have to delete LinkedIn 40x to save the same amount of space as the dozen or so junk games I have on my phone.

[actionfromafar]

How would they know, though? If the download times out, does LinkedIn care or notice?

[WJW]

Surely they'd at least notice a decrease in installations of the new "too big to download" version?

[screamingninja]

> Surely they'd at least notice a decrease in installations of the new "too big to download" version?

You are referring to a hypothetical scenario where A/B testing took place.

[actionfromafar]

And that department will take immediate action, communicate efficiently across corporate governance teams, all in perfect harmony, to address at the next sprint, as ever. /s

[legacynl]

On android there is a way for apps to initially provide a barebones package, that downloads and installs quickly, but that then fully downloaded itself afterwards. So I guess there already has been some engineering effort, just not in making sure apps are small

[vitorgrs]

Actually, it's a huge metric. Not sure on iOS, but on Android this is so fundamental that usually companies even launch after a "Lite" version exactly because of this.