Google has released an unfinished SDK for the upcoming Honeycomb incarnation of Android, a release intended for touch tablets as well as smartphones. According to Xavier Ducrohet, Google's Android SDK tech lead, Honeycomb is "specifically designed to give developers the tools and capabilities they need to create great …
Their big improvement in user interface is to copy the iOS UI?
@DJ ....is to copy the iOS UI
Nah, it's way nicer than Cisco's Command interface
Doesn't look anything like the iOS UI
Android's 3.0 is nothing like the iOS UI except in superficial ways.
The UI is far more 3D enabled - e.g. Add a widget and all your views are arranged in a semi circle to make it easy to drag & drop the widget onto one of them. Flipping between views also invokes a 3D swipe. It's also clear that Google have put a lot more thought into multitasking than iOS with recently used apps etc.
That said the preview emulator performance is terrible. The emulator looks like it is struggling to do most of the 3D in software and it chugs. There are also only a handful of apps to look at and the default theme is pretty ugly. I hope and expect that when the final release turns up most of these issues will be resolved, but it's a bit worrying to see how long Android 3.0 is taking to come to fruition.
Get some perspective
"Android's 3.0 is nothing like the iOS UI except in superficial ways."
Awe come-on. I develop for both Android and iOS and I love both platforms. But the level of worship is blinding acolytes to the bleedin obvious. Just stand back from Android for a minute and think about if you had to write a history of computing and it should hit you like a baseball bat this statement is from someone lost in the trees of development, loving all the detail but wholly unable to see the forest. Androids strength is to absorb and improve. There's nothing wrong with admitting it has copied most of the interaction paradigm from iOS (and also where it has diverged it is often - not always, but often - the worse for it. I know many will disagree on that). Also regarding multi-tasking. I'm afraid the proof of the pudding on that one is turning out to be in the eating. We developers dis the Apple approach because they won't let us do just what we want to do. But the Android approach, whilst being more open, is somewhat akin to putting a bucket full of cash in the middle of the street and saying "now now everyone, we are making this available to you because we don't want to be restrictive and all, but please only use what you need." So again, standing back, yes I love the flexibility and power, but I'm really unable to say it is the best approach or better thought out than iOS and I can't think of any use cases that are in the user interest where iOS's more measured approach makes me deliver a functionally compromised app, but I can certainly think of cases where Androids approach can lead to a compromised user experience.
Get some perspective yourself
Have you actually run Android 3.0 preview? If you had you would realise how superficial the resemblance is to iOS. The home page doesn't look anything like an iPad aside from showing some app launch icons. It's not even that similar to Android 2.x with most launcher completely gutted and replaced, and major changes to the browser & other apps.
I don't even understand why you think I lack perspective. My original post was hardly glowing of what I found in Android 3.0 preview. I think it's a bit lacklustre so far and I'm hoping things improve substantially before release.
"but it's a bit worrying to see how long Android 3.0 is taking to come to fruition"???
do you mean something other than what you seem to mean with that statement? Are you serious?! What other OS has revisions pumped out as often as android?
Just remember how many YEARS microsoft take to release fairly minor revisions to windows!
Yes I'm serious
<i>do you mean something other than what you seem to mean with that statement? Are you serious?! What other OS has revisions pumped out as often as android?</i>
Honeycomb has been rumoured for 4 or 5 months now, certainly in the works longer than that, previewed before Christmas and now when we see it it is virtually broken. It barely runs at all in the emulator and even the browser app crashes virtually instantly. It's barely fit even for development work.
So yes it's worrying. Dozens of tablets are announce to be running on this thing and it isn't ready. I hope and expect things improve but what is there is seriously underwhelming.
"... copied most of the interaction paradigm from iOS ..."
And where did Jobs copy most his work from?
Dating from early days Apple has plagiarised others work so they are not innovators as much as a photocopier is not an artist or author.
Fix the emulator!
The SDK is still using the same god awful ARM emulator. It runs like an asthmatic dog pulling Canada up a mountain, even on a quad core system. How are you meant to develop anything when the test environment is so piss poor?
Re: Fix the emulator!
"How are you meant to develop anything ..."
Actually (whilst I'll concede the emulator should be improved) in general it's good discipline for developers to test their code on slow systems. That way they'll put more effort into getting them running fast in adverse circumstances, so that when the code gets released - onto unknown systems, often trying to run loads of other programs at the same time, the code will still work.
The same applies for Internet bandwidth: if your code (website, whetever) relies on having bandwidth available, develop it under adverse conditions, to force your developers to think about compression etc.
Yes, it's a sometimes painful process, but the end result will be much better.
Emulator is fine in 2.x
I use the emulator on 2.x apps and it works fine. Slow but acceptable.
I think the emulator is chugging on 3.x simply because of the new 3D heavy UI. The 3D is being done in software (emulated software no less) so it heavily impacts on performance. If Google were to hardware accelerate the graphics, I think performance would dramatically improve.
The emulator is actually QEMU based, so I suppose that a custome guest driver could generate traps that QEMU could translate into host GL / DirectX instructions. Thus the rendering would be moved out into the host where it would be accelerated. This is more or less how other VMs work.
- Vid Hubble 'scope snaps 200,000-ton chunky crumble conundrum
- Bugger the jetpack, where's my 21st-century Psion?
- Google offers up its own Googlers in cloud channel chumship trawl
- Windows 8.1 Update 1 spewed online a MONTH early – by Microsoft
- Interview Global Warming IS REAL, argues sceptic mathematician - it just isn't THERMAGEDDON