Show me the data that says users want bigger phones. Some people point to overall Android market share and conclude that because it’s “winning”, and because many popular Android phones1 have comically huge screens, then people want comically huge screens. That doesn’t mean shit. Not all Android phones have four-inch displays; some have displays smaller than the iPhone’s. Are people buying Android phones because they want a bigger screen? Are they buying them because they prefer Android as a platform, and actually dislike the large screens? Or are they simply buying whatever phones their carrier is pushing them towards?
It’s pretty fucking ridiculous that you can hardly find any decent Android phones with a screen size of less than 4.3 inches. ↩
iCloud’s launch probably wasn’t as big of an impact as Apple would have wanted. Everyone had really high expectations for it and that it would radically change how iOS users interact with apps on multiple devices. It didn’t quite turn out that way.
The obvious first things to complain about are sync issues. As mentioned in my earlier post, mine seem to be fairly compartmentalized. Contacts and calendar sync perfectly across all my devices. Photo Stream works marvellously. “Documents and data” works for some apps and doesn’t for others. The process is so opaque that it’s hard to know whether iCloud or the apps in question are to blame.
Bookmarks have me tearing my hair out; I have three separate bookmark hierarchies on my iPhone, iPad, and iMac. It’s like the root bookmark hierarchy it’s applying deltas too wasn’t the same on all devices, and now there’s no real way to get them to converge. Sometimes iCloud will just automatically revert changes to my bookmarks as soon as I make them. Needless to say, I stopped using in-browser bookmarks for anything but bookmarklets and moved everything over to Hatena Bookmarks.
As far as sync reliability is concerned, I don’t see how iCloud is any different from Mobile Me. Both had wildly unpredictable sync issues and the “solution” on both was to nuke everything and start over. The big difference is that where Sync Services would make a conflict resolution dialog appear if a conflict occured in the Mobile Me days1, iCloud attempts to automatically resolve it behind the scenes. You better trust it makes the right decision.
Many people are reporting duplicates all over the place; luckily, I didn’t get any of those.
Next up on the complaint list: the lack of documentation or examples on how iCloud should be used on the Mac.
When it was announced that iCloud could be used to sync documents and data between various devices, everyone showed it working between iWork on the iPhone and iWork on the iPad, and we all expected iCloud syncing would appear on the Mac in time for launch. It did not.
Yet, if you asked just about anyone prior to the iCloud announcement what the most annoying thing about the iPad was, it was how to get documents on there for editing. You had to either connect your iPad to your computer via iTunes and drag them there, email them to your device, or set up a WebDAV server with all your documents on it.
The iCloud launch added one more cumbersome way to exchange iWork files between your computer and your iPad: a Web app where you can drag files in and out of iCloud. Even with iCloud, you are still managing copies of the document instead of directly manipulating the “ubiquitous file”2 stored in the cloud.
Everyone who expected updates to the iWork suite sat there going “what the fuck” on the day iCloud launched, and there still has been no word from Apple about what’s going on.
That might seem like it’s a problem limited to iWork users, but it’s much bigger than that. Since iWork updates are MIA, there is no concrete example of what the iCloud experience should look and feel like on the Mac for traditional document-based applications.
iWork on iOS is self-contained. Any documents that are available on one iOS device are available on the other. You don’t carry around every single document you’ve made, but rather a subset of your documents that are relevant right now. That model is pretty easy to understand.
iWork on the Mac isn’t self-contained; Pages/Numbers/Keynote files are hidden away in folders across your hard drives, and you might want to keep things that are years old around for archival purposes without necessarily wanting them synced to all your devices.
Knowing how the two environments differ, how do you design the user interface for both getting documents on and off of iCloud on the Mac? Do you just make a “shared on iCloud” checkbox? If so, how do you open documents from iCloud? Remember that with sandboxing on the way, developers won’t be able to modify the Open dialog anymore3, so you’d have to have two different Open dialogs; one for iCloud docs and one for local files.
Developers that want to use iCloud in their apps want the experience to be consistent with other apps, but right now, there is no first-party app to be consistent with. Each developer who wants to get an iCloud solution out the door now needs to come up with their answer to those questions and hope it doesn’t end up being too different from what Apple will eventually deploy in iWork.
There are also use cases iCloud can’t really deal with. What if you prefer iWork on the desktop but another office suite on the iPad? Only iWork is allowed to access iWork’s container, so those files are inaccessible to any other app unless you export them and get them into the other office suite somehow. People have come to expect greater inter-app compatibility with the loads of Dropbox-enabled apps on the App Store today; iCloud is a step back as far as that’s concerned.
So what will Apple do with iWork?
This will piss many people off if it ends up happening, but I can’t really see it happening any other way; Apple will end up making iWork on the Mac more like iWork on iOS and have an “active” subset of documents which are synced via iCloud, and an “inactive” archive of older documents, all stored within the app’s sandbox. It would feel inconsistent if iWork didn’t abstract away the file system when Apple is strongly suggesting third-party app developers do so. That would also solve the user interface problem in the cleanest and most consistent way possible.
But how long do we have to wait for those iWork updates? Will Mac developers grow tired of waiting and we’ll end up with a mess of inconsistent iCloud experiences? Who knows.
Now, the only place where you can do manual conflict resolution is syncing contacts/calendar to your iOS devices via iTunes instead of over the cloud. ↩
"Ubiquitous files" are how documents stored on iCloud are called behind the scenes. Sometimes. In the spirit of iCloud’s unpredictability and inconsistency, it’s also called "mobile documents" in certain places. ↩
You could make the Open menu item and keyboard shortcut open a dialog which isn’t the Open dialog, but has an “Open Local File…” button, and maybe make “Open Local File…” Opt-Cmd-O or something like that. ↩
"If you’re deleting the Instacast.icast file and the system is telling you that some changes are pending, this means that the iCloud service is stuck somehow. I couldn’t yet figure out why, but I have reports that deleting everything and starting over solves the problem sometimes."
When it works1, iCloud is awesome. Problem is, when it doesn’t work, there’s no real way to fix it other than nuking everything. Apple really isn’t that great at cloud services.
Calendar and contact sync has been flawless. Documents & Data is magical with djay, but has worked once with Instacast. Don’t even get me started on fucking Safari bookmark sync.↩
I hadn’t even heard of Kachingle prior to this, but most of this could apply to Readability (which has resurfaced in the media this week) as well. It’s also worth reading the four tweets referenced in this post by Ben Brooks.
The Kindle Fire embargo was lifted last night and reviews have been coming out left and right, more or less saying exactly what I thought they’d say. (It’s the best Android tablet to date by a long shot and definitely the best tablet you can get for that price.)
But… wow. What the hell is up with those third-party developers and their icons? Some are squares, some are round rects, but the round rects don’t all have consistent corner radii. My eyes are bleeding.
A friend of mine worked on Batch, a photo sharing app for the iPhone, and the latest in the horde of apps requiring you have a Facebook account. The app is pretty damn good; it’s pretty, it uploads crazy fast, even on 3G, and it’s better suited for sharing albums (batches) than single pictures of a moment (which Instagram is better at). I told my friend when I first used it that it was what Color would have been if they launched something actually usable.
I went out to the arcade with friends earlier, and decided to take a couple pictures. I wanted to test out the service with something from one of my few real-life outings, uploaded them to Batch, and shared it to Facebook friends.
What do you think Facebook friends are faced with when they click my “Biermans Night 11/03” batch? A Facebook Connect button.
I thought I had fucked something up and had some crazy privacy settings enabled. I didn’t. The two privacy settings within Batch are Private, which lets you (and anyone tagged in the photos see them), and Friends, which says “friend on batch can see this”. I had chosen the most permissive one, Friends.
It’s not like Batch is entirely usable from the Web either. The only way you can get to a batch outside of the app is if you are linked to it on Facebook or Twitter. You can’t upload photos or create batches from the web. All you can do is comment or like photos. And Batch’s app is currently iPhone-only.
So why couldn’t the Facebook Connect button only be forced upon the user when they choose to engage with the photos they’re seeing? Why do I feel like by sharing my photos, I’m no better than those stupid “like this page to see some stupid video” people? Why does it seem like my photo sharing is being restrained by artificial inflation of their user count?
This is making me extremely unlikely to share photos in Batch with anyone else, and honestly, unless a substantial subset of my friends start using it, I’m unlikely to ever use it again.
Having decided the hook line should be “something-something-dancefloor”, they managed to complete the phrase by Googling for “nice words and weird combinations of words”. They then built the story of the song around it.
In case you weren’t aware the briefly-available Gmail app was just the mobile Web app shoehorned into native navigation bars and toolbars, here’s a set of screenshots comparing the two, courtesy of Matt Gemmell.
The mobile Web app is amazing for a Web app. Shoehorning it into a native app, where expectations are much higher, is setting yourself up for failure.
I honestly have no idea what Twitter for Mac’s cache is like internally, but it seems to be a mess. Not only does it often take several restarts of the app for it to pick up on avatar changes, but it just behaves weirdly.
Twitter for Mac has two different avatars for halcy in its cache, but neither of them are his current one. One of them (in the timeline) is his previous avatar, and the other (used in username completion) is at least two avatar changes old. If I click his (cached, old) avatar in his profile to get a bigger version, it downloads the big version of his new avatar, but doesn’t cache it or replace the old one.
In a dream world, Tweetbot for Mac would exist. Until then, I think I might switch to Twitterrific again just for Tweet Marker support.