Here’s why the Facebook iOS app is so bad (UIWebViews and no Nitro)
blog.mobtest.com
blog.mobtest.com
I'm sure FB could increase their engagement by having a better mobile app. Often I want to see what my friends are up to, but just give up in frustration after waiting literally minutes for the timeline to display. I guess with such a large market share, they see no need to improve.
[1] - http://blog.mobtest.com/wp-content/uploads/2012/05/facebook-...
I have a Galaxy S, which is not the fastest phone in the world, but if I get a notification that someone has commented on my post or something, it can literally take minutes for the FB app to:
- Start when I click the notification
- Replace the pointless blank screen with my timeline (not that I wanted that, I wanted the notification) and the now-red notification bar
- After tapping the notification icon in the FB app (the second time I've now indicated I want to read the notification) watching the little bubble appear and wait for it to be populated with notifications
- Once it's populated, tap the notification I want (3rd indication now), and wait for that page to load.
Each of these on a 3G NZ mobile network can take 10s of seconds to complete. All the while I can't use my phone for anything else, or indeed let the screen go black and lock, because that will often drop me out of the particular step I'm up to (this may be more my phone than fb, I'm not sure of the internals).
Thinking purely as a user and ignoring the technical issues, there is something very wrong somewhere when it takes a few seconds to start a youtube video streaming but it takes a minute to view my friends 6 word sardonic reply to my kitten picture.
The FB app on WebOS was bloody perfect. So many things WebOS did right... I miss that platform.
The interface is attractive enough and doesn't have any glaring usability issues.
Performance (on my Nexus S w/ 2G connection) could improve but isn't so bad that it causes frustration.
And it's available for free and without ads.
I get the style sheet bug up to a few times a week, but it's no big deal. Just pull the page down to refresh and it's fixed.
Maybe my standards are low but I suspect 95% + of people are the same, so why would Fb bother doing it any differently?
* - https://play.google.com/store/apps/details?id=org.mots.haxsy...
The problem with the mobile web interface is that it doesn't support HTTPS (unless they have changed that recently). Not a problem when you are on your own wireless or your carriers cellular network, but I wouldn't trust it on a public wireless AP (I've seen people be caught out by session stealing tricks like that illustrated by firesheep - nothing malicious but you know if someone out there is using it to joke about someone else will be using it to cause grief).
Then again, the app may well use HTTP too - I've never installed it to try so I've not profiled it to see what access method is being used.
If I'm right and most people don't care, then it's a huge win to mix in HTML when convenient for the same reasons mentioned in the very article this thread is linked to that's lambasting Facebook's app.
The app UI is great, but I just don't use it because it the payoff is not worth the wait.
the story you're commenting on presents evidence which contradicts what you're saying. unless you count a brief introductory paragraph, it is literally the first point they make.
the article says that the app has nearly 12K one-star reviews and an average of 2 stars, indicating that people don't like the app. the writing is not 100% clear, but it appears to be saying that the app only has a little over 21K reviews total, which means more than half of all reviews for the app are one-star reviews.
how could you possibly square this information with your belief that most people don't care? they counted the number of people who care and not only are there thousands of them, but more than 50% of the reviews are as negative as it is possible for a review to be.
seriously, dude, "people hate it" is the first point this article makes, and it does so with plenty of compelling evidence.
The Android app has 3.2 million reviews (averaging 3.6 stars - 600,000 or so are 1 star).
So it doesn't seem as bad when you consider "all versions" instead of just "this version". I wonder if it's a matter of more tech-savvy people reviewing the newer versions when they come out? Or it could just be that the app really is getting much, much worse with every release...
Personally, I'm waiting for them to fix contact sync on Android ICS. That's the only reason I have the app installed...
The sync worked before because Google allowed them to use a special API to allow them to show contacts on your phone without actually syncing them to your account (removing the Facebook app removed the contacts)
"Facebook for iOS"? More like "Loading... for iOS"!
1. Tap highlighted icon 2. Suppress the instinct to tap the first entry (high failure rate here :)) 3. Pull to refresh 4. Wait for refresh 5. Hit the new first entry
Either this happens to everyone, then it contradicts your point, because not every normal user knows how to pull-to-refresh. Especially since there's no visual indicator for this at all.
Or this bug only affects some users, and then it explains why you are happy. My girlfriend on the original iPad certainly has the same problem.
As a counterpoint, my wife often finds the Facebook Mobile app to be incredibly sluggish.
Many users often don't complain about performance, but if you make things work faster they tend to appreciate it anyway.
Skip forward 3 years and remove Joe Hewitt, in an attempt to simply their development portfolio the Facebook app is everything it shouldn't be. It's slow, it's glitchy, occasionally when you tap something it'll decide that you've requested something completely different. At the end of the day a company needs to simplify development eventually, but not at the expense of crippling their application.
The code itself is poorly designed, the API's are all over the place, there are inconsistencies everywhere and the performance is clearly lacking. On the flip side, take a loop at the LinkedIn app for iPhone. It's super slick, easy to use but still has a ton of features and the performance is far better. Yes, it's not as slick as Path but its miles ahead of the Facebook app.
So if I were Facebook, I'd rely a bit less on the 100% platform agnostic approach, take it back a bit and build the things that make sense cross-platform with HTML and build the parts separately that make sense to build using native code. This approach to me accomplishes much of the same cross platform success without creating such a crappy and laggy app.
Probably just to cross a bullet point somewhere. Still, with their warchest, they could easily dump money on a small company and get it over with. At least something basic, but usable.
Say, timeline plus chat.
The reason it is fast though, is it is not loading that webview from a website, but building it from data locally. That gives it a native app speed.
On the other hand, the original Google+ release for the iPhone was bad because it was pretty much a webview downloaded on demand too.
Facebook is just loading it on demand basically as a webpage.
So it's possible Twitter is using a webview, but downloading the data separately for speed.
http://www.reddit.com/r/Android/comments/h7nx1/we_are_the_de...
http://www.reddit.com/r/Android/comments/n5my6/we_are_the_de...
http://www.theprintlabs.com/with-ios5-the-faster-nitro-javas...
This was how app was designed. With so many technologies involved things are bound to go wrong