Slashdot Mirror
John Gruber On Third-party Apple Watch Apps: They Suck and Are Really Slow
An anonymous reader writes During this week's episode of John Gruber's podcast, The Talk Show, Gruber sat down with Joanna Stern of the Wall Street Journal to talk all things Apple Watch. About two hours and 9 minutes into the podcast, both Gruber and Stern began lamenting the poor performance they saw with third-party Apple Watch apps. 'It makes me question whether there should be third party apps for it at all yet,' Gruber noted. The pair also took umbrage with what they perceived to be a poor design choice for the Apple Watch app screen, with both noting that the app icons were far too small to be practical.
It's a new field. Vendors are going to trip and stumble until the lessons of the street straiten things out. The first PC apps sucked, the first Mac apps sucked, the first Linux apps were...all in EMACS, anyhow, you get the point.
Table-ized A.I.
That makes no sense. They don't achieve anything if their apps look better than the other apps on the device, they just make the entire experience worse. It would be like cutting off their nose to spite their face.
Bogtha Bogtha Bogtha
What's worse than the apps on an Apple watch?
A 2 hour podcast about the Apple watch.
The real question, of course, is whether the apps are the problem or the device itself?
After all, Apple no longer has perfectionist management at the top. It seems to me that they are more likely to release a product before it's fully baked. When the iPad was release, Apple had gone through hundreds of prototypes. I wonder if they put the same amount of design effort into the Apple Watch.
Smart watches ARE a good idea, they just need to have better characteristics then they have now. They need a much longer battery life, a LOT cheaper and they must universally work with any device, not just proprietary ones. There is nothing wrong with the idea of a programmable display on your wrist, but so far the implementations suck.
Good-bye
Keep in mind that you're looking at people who spent hours upon hours writing blog posts speculating about the leather and alloys Apple would be using in their watch bands.
A 2 hour podcast about an actual shipping device seems comparatively reasonable.
Log in or piss off.
Currently the application are hosted remotely on the I-Phone. Apple has promised that the will release a native api in the near future. What they are seeing right now are NOT native apps.
love is just extroverted narcissism
Unless it's in China, who the heck wears a watch nowadays, other than old people?
I mean, come on!
And why would I want to be constantly interrupted by stuff I don't want to do anything about?
Maybe an Apple Monocle. That I could see. Give it a wider spectra range so I can see IR and UV and display stuff, but pop out of the eye when I don't want to be bothered, like a real monocle. Totally retro steampunk. That's the ticket!
-- Tigger warning: This post may contain tiggers! --
Ohh FFS -- that was at the initial launch and not done as a fuck you but simply because they were more interested in just getting the new product and OS out the door. Initially, Apps were JS based and highly sandboxed. But they realized that Devs wanted something better and Apple set themselves to creating an entire Development Platform and App Store to support that. So any claims that Apple is somehow hostile to developers is utter BS. If anything, Apple stands to make millions off 3rd party Apple Watch Apps in regards to App Store and In-App Purchases.
Ohh FFS -- that was at the initial launch and not done as a fuck you but simply because they were more interested in just getting the new product and OS out the door.
It was definitely a "fuck you, this is a phone; this is not another fucking Newton".
Full disclosure: I was an Apple Core OS kernel team member at the time. I wrote 7% of the kernel that runs on the things.
The UI definition is held in a Plist format (like, but not, XML) but that's not what the device gets. It gets a very compact binary form of your UI, that is loaded onto the watch before the user even opens your application.
The Apple Watch API is actually EXTREMELY conservative with what gets sent over to the watch, to the extent that even attempting to set the same label value twice in a row is rejected with a warning. and UI elements on the screen are wits-only (you cannot query the watch see what currently displayed values are).
"There is more worth loving than we have strength to love." - Brian Jay Stanley
The Pebble is 1/3 the price, has a 1-week battery, and works with iOS, Android, and Sailfish. I will be interested to see how the Apple Watch actually does after release, since every smart watch review I've read for the past 2 years has measured against the hypothetical iWatch rather than the real competition.
"Because Science" is one step from "Because old book". Try "Because of my experiment testing my falsifiable assertion".
uhm, actually plist files are xml
ACTUALLY plist files can be either textual or binary, which is very much not XML
I should have said not necessarily though, instead of just "not"... but it was kind of irrelevant to the main point.
They certainly aren't very compact as far as formats go, even on the watch.
Sigh, didn't read much of that original message, did you?
They don't NEED TO BE EXTREMELY COMPACT because they are sent over only once, when the app is loaded on the watch - that said, it is in the binary format which is much more compact than the textual format.
In use the watch pulls files from that bundle at runtime. And if you were any kind of programmer you'd know there is a tradeoff between compression and computation (which the watch has little of) in terms of file formats, so a fairly but not maximally compact file format is better for performance than whatever you are thinking of.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Full disclosure: I was an Apple Core OS kernel team member at the time. I stole 7% of the kernel that runs on the things from Mach and BSD.
FTFY
Given that I also wrote much of the init.c in FreeBSD, you probably failed to fix that for me. :p
Eat my shorts.
http://code.metager.de/source/... /*-
* Copyright (c) 1995 Terrence R. Lambert
* All rights reserved.
*
* Copyright (c) 1982, 1986, 1989, 1991, 1992, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*