I remember, like, three years ago looking at the LI app and wondering, why was it almost 200MB. Bloat breeds bloat.
Why is the LinkedIn app almost half a gig?
51–60 of 205 posts
Re: Why is the LinkedIn app almost half a gig?
#52Earlier quoted context omitted.
Uber actually had a pretty reasonable explanation for their app being huge when their rather large app size made the rounds several years ago. Turns out localization takes a lot of space. I wonder how much of LinkedIn’s insane size can be attributed to that vs just lazy coding
> Turns out localization takes a lot of space. Does localization involve more than a few hundred strings per locale, all of which should compress really well? Maybe add in some localized icon sets, but much fewer than the number of supported locales. I'm not saying you're (or they're) wrong, I'm genuinely curious.
Re: Why is the LinkedIn app almost half a gig?
#53Earlier quoted context omitted.
> There are likely 100s-1000s of app developers involved in the LinkedIn app I don't understand why you'd need 1000 or even 100 developers to produce such app. It's a CRUD frontend for a backend that already exists. What am I missing? Are you counting the developers of all the third party libs they integrate?
Furthermore, I would like to see some stats how much of their traffic is via the app and how much is via the website. Of course many people have LinkedIn profiles, but are there really so many "hardcore" users who install the app? So maybe the app just isn't that important for them in the grand scheme of things, and the bloat is more the result of neglect than of hundreds of developers working (specifically) on the a…
Re: Why is the LinkedIn app almost half a gig?
#54Earlier quoted context omitted.
If I only ever use a single locale on my phone, why isn't there a way to avoid needing to download all of the extra info that my phone will never use?
I think App Store's app thinning doesn't care about locales, so any such attempt should download an additional localization file from Uber's servers. While not technically impossible, it will mean that downloading an app is no longer sufficient to use the app; it should also go through the initial (or incremental) download, which was significant in the case of Uber's case (1 MB per locale according to the blog AFAIK)…
Re: Why is the LinkedIn app almost half a gig?
#55Earlier quoted context omitted.
Hacker News has a lot of people who swear up and down that phrases nobody has said before ever are actually standard lingo
Technically that's how you start trends - you claim that others are doing it, so your target group does, and now you have an example to point to!
Re: Why is the LinkedIn app almost half a gig?
#56I remember, like, three years ago looking at the LI app and wondering, why was it almost 200MB. Bloat breeds bloat.
Re: Why is the LinkedIn app almost half a gig?
#57I would start by looking at what junk is linked in.
Re: Why is the LinkedIn app almost half a gig?
#58TL;DR: There're few global incentives for a LinkedIn app developer to make an overall good app, as opposed to making their PM happy. In detail: This is how death by a thousand papercuts looks like. I've had the (dubious) pleasure of observing a roughly similar BigTech app project. There are likely 100s-1000s of app developers involved in the LinkedIn app, and they are likely under immense pressure to Just Ship Alread…
> There are likely 100s-1000s of app developers involved in the LinkedIn app I don't understand why you'd need 1000 or even 100 developers to produce such app. It's a CRUD frontend for a backend that already exists. What am I missing? Are you counting the developers of all the third party libs they integrate?
Re: Why is the LinkedIn app almost half a gig?
#59Re: Why is the LinkedIn app almost half a gig?
#60Earlier quoted context omitted.
> the grumpy ones You know, grumpiness really is a thing. I wonder if it should be treated as a symptom like left arm tingling before a heart attack. A call to make some serious changes.
Meh, developers are people too. They are right and they are wrong. I've seen some guys who are proud of having the grumpy persona, which is really just a nice word for "I know better than the other people I work with" and it's not a great attitude. Engineering choices all have trade-offs yet the grumpy guys always frame it as "my way or everything will collapse later".
There are long-tail risks to the stability of codebases, organisations, institutions, governments.. the soviet union collapsed.
It seems likely that some engineers can intuit such long-tail risks.
If linkedin is subject to a competitive threat which requires rapid innovation, the org will lose because it's development practices preclude it.