Very cool.

Very cool. Except it said it was raining and breezy on my street at work earlier when in fact it was quite sunny and warm.

Apparently it's 32°. I presume that's Fahrenheit as it blinking freezing here!

It's able to determine my location, so should be able to use local units.

There's a switch in the corner.

Just as long at it's not Kelvin.

It has me down as being about 30 miles south of where I actually am...

BTW, we've had -4°C every morning this week, but the air is so dry, I haven't had to scrape the ice off the car once.

It would make a great phone app.

"Apple needs to finish writing this syncing ship"
http://www.theverge.com/2013/3/26/41486 ... -just-work



There are many cloud based database systems which do work. I wonder why Apple built one that seems fundamentally flawed..?

Actually, there aren't many that work the way iCloud is supposed to. What they're talking about is quite hard. It's not the same as a cloud based file system like Dropbox or SkyDrive. Nor is it the same as a single schema sync system like CalDav. iCloud sync is ostensibly about syncing any kind of data at all that any app might require, for multiple apps with different requirements. With the results pushed out instantly to all connected clients. Effectively we're talking about an arbitrary number of database schema per instance, for a very large number of instances, with instantaneous commits. That's 'non-trivial'.

It's technically doable with enough work and Apple certainly have enough money to throw resources at it until it does work. So they really don't have any excuses to be honest. But it's not the same as an app using a bespoke web service back end for it's own syncing and it's not the same as file list syncing the way shared drive services do things. So it's not really valid to compare it to those. The requirements and complexities are different. Not insurmountable and it should work, but it seems to me right now this is a bit of an internet pile-on and people are using the fact that systems than aren't anywhere near as complex as it work as a stick to beat it with, which isn't correct.

It SHOULD work, but whether Dropbox or Beanstalk work is irrelevant to whether iCloud should work or not.

I realise it doesn't compare to simple file syncing services. The difference is covered in the article, which also notes that: "When Apple does use iCloud, it most often doesn’t even use Core Data to sync information. Apps like Keynote, for example, use the far simpler Document-based syncing method."

The "other cloud based database systems which do work" I was referring to were things like Amazon's RDS and even the Google App Engine which enables apps to sync without users needing an account - it uses the App's account, which is free up to a certain (rather small) size.

I don't know what features Core Data attempts to deliver that these other databases can't? And since it doesn't seem to work well, perhaps they'd be better off with something simpler.

One for bratty:

I remember the story of the previous Apple attempt at this - .Mac - which Steve Jobs asked the developers what it was meant to do. They told him, and his follow up question was "why the [LIFTED] doesn't it then?"

Sadly, they removed something that I used a lot - the iDisk (a kind of Apple Dropbox) - when kicking off iCloud. Now what happens is if I start a Pages document, t tosses it into the Cliud, and when I save by default that's where it goes. This is great for shunting stuff around between iOS and Macs, but the assumption that I want the cloud as a default storage space is wrong. I've had sync problems with documents - iOS and Mac OSX not agreeing on which is the most recent version.

I much prefer the Dropbox method.

O.M.F.G

I knew it.

I see triangles everywhere.

I should probably upgrade my polygon count.