1696 posts • joined 18 Jun 2009
- amongst others, Spotify can continuously stream music; streaming and playing audio from the background is one of the specific types of multitasking Apple explicitly wanted to allow. A direct quote from their documentation: "Applications that play audio can include the UIBackgroundModes key (with the audio value) in their Info.plist file to register as a background-audio application. This key is intended for use by applications that provide audible content to the user while in the background, such as music player or streaming-audio applications. When this key is present, the system’s audio frameworks automatically prevent your application from being suspended when it moves to the background."
- "Any time before it is suspended, an application can call the beginBackgroundTaskWithExpirationHandler: method to ask the system for extra time to complete some long-running task in the background. If the request is granted, and if the application goes into the background while the task is in progress, the system lets the application run for an additional amount of time instead of suspending it." That's how a well-written browser can continue downloading files when in the background.
- there's an execution mode, designed with VoIP clients directly in mind, that allows a piece of software to connect itself to a socket and subsequently wake on demand. That's how a file store-type application could accept and transmit files without having to be kept in the foreground.
- I'm counting this as the same problem as the file stores on a technical level, though I appreciate it much better demonstrates how the alleged deficiency can directly a task that's an upfront promoted feature and a reason that non-technical people will possibly buy the device, whether rightly or wrongly.
So: yes, the request to continue arbitrary task processing may be denied. Yes, the best support is for event based tasks, and it's a simple fact that the number of permissible event hooks is small.
No, that doesn't mean that multitasking isn't supported, or that it isn't 'real' multitasking. Multiple tasks can run, simultaneously. That's multitasking. Your complaint, taking it at its strongest, is that Apple will sanction only certain types of multitasking, which cuts to issues of approval and control rather than technical software issues.
It would therefore appear you plugged in an average of five times a year, that being the iOS release average for the two years immediately following the release of the 3G. You couldn't perhaps be exaggerating, could you?
I'm not one to claim that Apple's kit "just works". But I do think the Mac is the only device that properly talks to the outside world in terms of file formats. The iPod, iPad and iPhone seem to be very iTunes dependent. Suffice to say, if you'd kept all your media in the formats Apple supports (which are all standards from other industry bodies, so not proprietary — they've just been selectively chosen), in iTunes, then the autosyncing is a great boon. It turns a common manual task into an automatic one and makes moving between devices immeasurably easier. However, that comes at the cost of, as you found, about a million hours more work if you've decided e.g. that WMV and OGG are the things for you.
There's not really an extra step because the assumption is that you're already using iTunes. So you already have your favourites in playlists.
So I'll say that it works very, very well if you stick to the rules. I won't say that it "just works".
The camera accessory kit includes an SD card reader. You can use it to transfer photographs directly onto the iPad without a computer in between. So it's not crazy to think they may use that route for importing video too, especially as they're seeming to want to pitch the device as something creatives can use out in the field.
You've obviously got broken software. Multitasking is extended to third party applications, in much the same way as it is in WebOS, which appears to have been the model.
While it's always possible to pick apart the specs, and find something to claim a distinction on, no objective definition of multitasking would exclude what third party software can do on the iPad.
The developer programme — giving access to the App Store, amongst other things — does indeed cost US$100/year and is a direct charge to the developer. So it's probably a safe assumption that they've collected US$1bn though that route, that being just 10,000,000 developer subscriptions worldwide.
I don't agree with you otherwise though. The user gives their money to Apple. Apple gives some money to developers. So it's definitely not the developers paying Apple per application in any legal sense. Apple handle hosting, payment processing and provide a limited amount of publicity. That's the traditional role of a shop, and claiming that developers are paying Apple because of the 30% cut is like saying that anybody with any product in any shop is paying that shop. So, even if true by the test you're applying, it doesn't sound like there's any grounds on which to single Apple out as doing something wrong.
Probably worth checking your maths on that one
If you want to phrase it that way around, it should read "App Store users have now paid around $857m to Apple".
That would make sense. The mechanism I tend to use, which I think is a defacto standard, is to get the relevant screen that the window I want to go fullscreen from is on, create a frameless window of that size and position and promote it to frontmost. That should be multi-monitor aware but doesn't always create the nicest switching behaviour. Giving us a proper way of doing it is probably sensible, but it'll be a bit odd if that's an on-the-box sort of feature.
I took fullscreen apps to mean frameless, game style. Which obviously are already trivially easy to produce, but so is any possible interpretation of a fullscreen application.
Obviously you're right about the zoom button.
Trying to play devil's advocate...
Auto Save (at the OS level) + Versions (presumably) join together as an evolutionary advancement on Time Machine, further reducing the ability for people to lose work. Auto Save is a bit more interesting than the stuff provided since time immemorial by Word, etc, because it has an aspect of versioning to it and should become uniform across the OS (or, more probably, like the built in grammar and spell checker, the Keychain, etc, will just make it even more obvious which are the particularly shoddy Windows ports).
File Vault and Air Drop seem to be designed to expose what many people would consider latent functionality, even though your or I could easily achieve both things now, on any moderately recent major OS.
Bundling the server features into the mainstream OS arguably makes this more of a premier OS than previously.
I'm unable to come up with any reasons why full screen applications are being sold as something new or even a particularly good idea, or to support the other cosmetic changes as worth paying for.
There are those that believe the LLVM/Clang push is part of an attempt to transition to a virtual machine (or, at least, to provide a path to one) without bundling it with a new language.
Quite the contrary
Microsoft's dedication to backwards compatibility costs them substantially in bloat, security and ongoing quality of code and seven years' backwards compatibility (albeit that I count only five in this case — the first Intel Macs, and hence the first Intel OS X, shipped in 2006) is about the Mac average.
In 1991, System 7 introduced the concept of being 32bit clean and made the MultiFinder non-optional, causing compatibility to be withdrawn for many older apps over the years immediately following. The mid-90s are awash with developed, heavily promoted and then nearly immediately dropped Apple technologies — see OpenDoc, Quickdraw GX, Rave, etc. Following which there's the transition into OS X, and now the final steps of the move away from PowerPC. And it's not just the emulator they get to drop, it's a whole bunch of legacy support stuff in the system libraries, since limiting to Intel binaries limits them to 10.4 behaviours.
They could just turn it into a breakout box
USB PCIe cards cost, what, about £20? So — subject to licensing fees as yet unclear — you're basically talking about being able to do a USB hub that connects via Thunderbolt rather than USB for close enough to a normal USB hub that those who want one can buy.
I guess it'd be the same position Mac people who bought the original iMac found themselves in with respect to their ADB, serial and SCSI interface devices. Except without the forced hand.
Good, except that
Some pages just stretch to the width of the screen rather than having a fixed width column of text, putting you in a very uncomfortable reading position. That's one of the reasons that standard paper sizes remain tall and thin and the main reason that newspapers use columns.
It's not just that, it's the obligatory equal pricing, so you've no way to pass Apple's 30% on to the consumer. It's a cost that businesses must be able to swallow if they want to play inside Apple's ecosystem. They now have to exclude themselves from the biggest selling mobile OS (yes, I'm counting iPod Touches and I do know that Android phones outsell the iPhone) or find a way to supply 30% to Apple while still being competitive with those that have excluded themselves from iOS.
For someone like Netflix, Spotify, etc consider US$99/year to supply a free iOS app to their subscribers versus US$99/year + 30% of take from anyone that creates or renews their subscription inside the app, with no option but to add that functionality to the app. And it's for subscription services, so you're probably not getting additional revenue from the on-average more affluent iOS customers versus customers in general, and Apple's rules prohibit a different price for iOS customers.
Likely outcomes are (i) dump iOS; (ii) introduce a higher subscription rate for all mobile customers. And the latter just adds weight to any consideration of the former further down the line.
On the flip side there are people, like my own company, with no previous subscription infrastructure in place and aspirations. This is great news for us, but for the industry as a whole it's likely to turn iOS into amateur hour. So even we're seriously considering whether to do anything.
The backlash started in 2003
The difference is that Apple have now made it impossible for others to defend them convincingly. The new charge on subscription services means that the costs of being in the ecosystem now outweigh the benefits for many major companies that you've heard of.
If competition regulators are able to do anything here (which seems unlikely) then it could be the first time they've intervened effectively to save a company's business model.
Not quite the same though, is it?
Apple's transitions have all come along with forward paths and backward compatibility. For the shift from 68000 to PowerPC, a 68000 emulator was bundled (and, reputedly, used internally by much of the initial versions of the PowerPC OS) that allowed old applications to continue running. The APIs stayed much the same, so there was no discontinuity in development skills. If you wanted to target Macs, you continued with QuickDraw, etc.
The switch to OS X came with the idea of carbonised apps — limit yourself to a subset of the existing APIs and build correctly and your app works not just now on OS 9 but natively on OS X when it comes. And OS X had an ability to run OS 9 as an app to run older apps. Although Carbon was a concession Apple didn't seem happy about, it was a first class member of the software stack up until 10.5.
The PowerPC to Intel switch was like the PowerPC switch. An emulator for old apps, all APIs to remain the same.
Conversely, the switch from Symbian/MeeGo to Windows Phone <version number as yet undisclosed> is announced to be a complete discontinuity of APIs and software support. The old Nokia obviously thought that sort of thing was a bad idea, hence QT to ease from Symbian to MeeGo or wherever they might end up. With the Microsoft tie-up, even QT is out of the window, per Microsoft's decision not to allow unmanaged code. Per announcements so far, we're not even going to get an emulator for older Nokia apps as Palm supply (or supplied?) on WebOS.
I really think the main reason people are concerned about an OS being dead is that they can't take their software with them and, more than that, if the APIs are about to go then the developers vacate immediately.
You're completely right about free stuff though. Giving away free can mean either that people don't want a thing (ie, it isn't selling) or that they do want it (ie, it's a good inducement).
I suggest you check again
Compared to the minimum Windows Phone 7 specifications, the Nokia N8 has a significantly slower CPU built on an older generation of the ARM architecture. The Broadcom GPU it uses is also likely to be slower than the PowerVR parts found on most high end mobiles.
As a consumer, I don't care and probably can't even tell, but I thought it was worth saying anyway.
London is quite central...
... if you work it out as average commute time for actual, living people. There are almost eight times as many people living in London as in second place by population Birmingham and the transport links are excellent. You know, in relative terms, compared to the country as a whole.
I'm not sure that supports everything happening there though.
The Apple statement was comparative. Unlimited data for £35/month in 2007 (so, 30% less than you seem to think) was a good deal and the risks are compared to picking some other smartphone. Obviously that's more risk than subscribing to a magazine, less than buying a house. And the iPhone is on the same price plans as everyone else now.
It's possible that all Apple want is to cream the 30% off whoever stays around for as long as they stay around, with more liberal terms to come when the market dictates them.
It's not really the same as the Apple/Microsoft shakedown. The Mac came to the market two years after Compaq had introduced the first complete IBM PC compatible and Microsoft had been trying to sell DOS as the platform before that, trying to create an 8086 market that didn't otherwise rest on IBM compatibility — like the CPM world before it. So Apple were already aware that proprietary systems were starting to lose.
They were very close to Microsoft in those days and licensed a lot of their GUI technology to Microsoft in return for software support. An error in the contract gave Microsoft the right to use the licensed bits in perpetuity rather than simply for Windows 1.0. So they lost the desktop not because they wanted to own the entire platform but because they were too trusting of third parties and were out manoeuvred, legally and on price (and technology, during the 90s, once the collapse phase began).
The Sony push will help
The Playstation phone is Android based, with Playstation games to be sold via the Google Marketplace. That should do wonders.
That's not the way most people see it
Their machines are already bloated monstrosities because their manufacturer helpfully included McAffee, etc. So the main experience of iTunes is that when they plug the device in, it just sorts itself out. No need to do anything manual whatsoever. Easy.
Nevertheless, a version that didn't seem primarily built of treacle is something Apple really should invest in.
Two significant areas without Edge or 3G...
Aeroplanes and subterranean railways. No tube trip is complete without seeing somebody listening to music or playing a game on an iPhone. It's difficult to believe Apple is happy to cut off those customers.
A 4gb budget device I could possibly believe, 0gb I don't.
I don't see how you reach that conclusion
Find me someone who had even noticed that iBooks is available. They've probably been barring jailbreakers since day one, it's just that nobody noticed yet.
iBooks and iAds are to Apple as phones and tablets are to Microsoft: evidence of fallibility.
I took A-Levels a bit more than a decade ago and even then the advice was to take General Studies just because in the final year the exam was before half term whereas pretty much every other exam was after half term. So it was used as a sharp shock to the student body. There were no formal lessons and the advocated preparation was "to read the newspaper". My university at least explicitly wouldn't accept it to count towards an offer.
That all being said, isn't the c.2000 division into AS and A2 meant to address the problem of people picking the wrong three? The average student takes five AS levels in their first year and whittles them down to three to study to A2 level in the second. So, especially for people like me that went to a separate college, you get a chance to experiment with interesting topics and then hopefully some decent advice and a rethink shortly before UCAS kicks off. That has a January deadline, so tends to become prominent at the start of the second year.
Not quite right on Safari
Just the pedant in me needing to shout up; it doesn't affect the tone of the article or the conclusions generally. Web pages in secondary tabs remain active in Mobile Safari unless and until the system starts to run out of RAM. Then they're purged, with only the little screen shot left to help you navigate from one tab to another. So it's just a way of attempting to do more with less, and is why you very often see tabs having to reload on the iPad (a small amount of RAM compared to the speed and storage required for native resolution graphics) but much less often on the iPhone (with more RAM and fewer pixels).
I guess the correct comparison would be: what does the BlackBerry do when memory starts to run low? But you shouldn't really be able to find out in normal usage, assuming they've put a decent amount in.
Any thoughts on the other 27 patent holders in MPEG LA?
I guess at least Samsung are okay by you, being Android licensees. But what about those innovation sucking whores at Columbia University?
You're thinking of nerds. Normal people pick phones based on a combination of their features, handset cost and available price plans. The vast majority of the phone market don't read technical publications like this and don't care in the slightest who the manufacturer isn't.
They're not really championing Flash
El Reg is doing one of the things it always does; irreverently poking fun at the gap between reality and the way various companies might wish things were. In this case, Jobs has stood up and said that Flash isn't suitable for mobile. It nevertheless seems to have obtained a foothold. So there's a story there, which for comment leavers seems primarily to be about Apple and control but in the story seems to be more about Adobe putting in some real work and a market full of players that take some positions just to differentiate themselves from Apple.
IE was about lock in
When Microsoft controlled the de facto browser, Microsoft controlled the de facto web standards. By doing so, they ensured that Windows was the only place you could see the web as the designers worked on it. Which didn't really do anything to prop up Windows on the desktop because Windows doesn't need propping up on the desktop — it won the battle for volume long ago — but did quite a lot to prop up the desktop as the only place to see full web content.
The transition back to an open market with multiple vendors has pushed the centre of gravity for real, practical, day-to-day work back to published, cross-platform standards. The big winner has been WebKit; once the web has to work properly outside of IE, suddenly Apple, Google, etc can put it onto a mobile phone. Or a tablet. Or whatever.
In the days when Microsoft had the only code capable of rendering what the web was filled with, you could expect them to have had a huge advantage in any emerging market involving the display of web content. They don't have that due to Firefox (plus others), and partly as a result their bottom line remains tied to the future of the full-size PC.
Much more so
And Nokia's flagship handsets don't currently meet them, so they can't just do an OS swap. They ship not only with the immediate bar of an incorrect screen size (more pixels than the minimum, but Microsoft give specific acceptable values), but all the Symbian^3 handsets remain on ARMv6 based processors whereas Microsoft have stipulated ARMv7, which is a modified and more recent instruction set. So it's not just a change, it's a need to move to more expensive components.
Not really a fair comparison
Bada is a clean up and opening of the OS they'd developed internally and deployed in various anonymous versions over the previous several years. Android (the company) was founded in 2003, bought by Google in 2005 and finally shipped its OS in 2007. So it took them about four years.
Hence, if Nokia had chosen to start again from scratch in an entirely reactive way to the original iPhone, they'd just about be finishing now. Instead — as you say — management seems to have scrambled around for quick fixes for far too long, squandering the talent and quite probably killing the company, at least as anything other than a forgettable me-too hardware shop.
Trolltech (as were) will be spun out and will survive happily on its own terms, just as it did for almost 15 years before the Nokia purchase in 2008.
Why do they bother developing anything at all?
Developing good software is difficult. So if 'difficult' is the threshold test for doing something then they shouldn't be developing at all.
If you just mean that Apple make it more difficult than the providers of competing tablets then I'd imagine it's because Apple have more users than the providers of competing tablets.
I suggest you check your facts. The Netflix app, provided by Netflix, Inc and available from the US iTunes App Store, is quite popular on the existing iOS devices and there's no reason whatsoever to think Apple would make the AppleTV the exception.
Not really, no
But people who will allow themselves to be seen only covered in white are quite excited that they'll soon be able to buy an iPhone.
But why not the z80?
Launched in 1976, built specifically for DRAM (with a built-in refresh counter and external signals) but smart enough not to require a memory access every single cycle, unlike the 6502. Though the 6502 sort of predicts the RISC era by having a fast storage area (the zero page) for load/store and only very simple processing abilities (barely any registers, none general purpose, all 8bit arithmetic) such that it uses most of those memory accesses you have to compare the performance effects on low end designs, once the RAM is being shared between multiple components.
The Electron used a 2Mhz 6502 with 2Mhz RAM. The video circuits have access to all of RAM. As a result, the 6502 spends most of its life underclocked at 1Mhz or ~600Khz, depending on the graphics mode. The BBC Micro used 4Mhz RAM and interleaved CPU and video accesses, but to achieve similar overall performance to the floating bus in the ZX Spectrum, that retailed at Electron prices.
All academic, of course.
But Sinclair and Acorn took their cues from across the Atlantic
It's not like the original home computers — things that plug into a TV with a built-in keyboard, usually built around a z80 or 6502, that boot into BASIC from ROM and are usually used with a TV for a monitor — sprung up near-simultaneously on both sides of the Atlantic. The Sinclair MK14 came two years after the Altair 8800, the ZX80 two years after the Apple 2, both because the Cambridge lot knew roughly what was going on in America and sought to make the same sorts of product here.
It cuts both ways
Have you seen all the headlines about Android being the number one OS on phones by quarterly sales recently? In that case, distinguishing devices with a 3G radio (like the iPhone) from those without (like the iPod Touch) has put Apple in a substantially worse position than it would have been had its aggrandisement been the only objective when dividing device sales into categories.
What about NDK apps?
That is, the native development kit, that I believe allows you to use C/C++ to write some code that executes directly on the CPU, outside of the Dalvik virtual machine? It's commonly used to port games, since they often have large C/C++ engines at their heart. It has limited API access though, I think to ensure you route most tasks back through the VM, so it's not necessarily a problem.
If this is a problem then I'll bet it remains middleware only; I'm not sure someone trying to explain the distinction between Dalvik and Android to consumers who are just asking why they can't download any of the popular games is a winning market strategy.
I stand corrected. I find what you're saying to be surprising, but that's hardly the point. Other devices with ARMv6s ship with some version of Flash that seems to do video content, such as the Nokia N8. That's why it hadn't occurred to me that it might be an issue.
I assume the other apps dipped into the H.264 stream that the BBC supply for iOS devices. Everyone else is now locked out of that since certain client certificates shipped only with those devices are required. That said, they were going around finding ways to locking people out on a case by case basis before they came up with that.
I'm with you in that case. If you're going to invest time in an app, ship it for a platform that can't already view the content anyway. The charter doesn't require them to find some way to get rid of the status bar or whatever else the negligible improvement of a custom app would be.
It already works on Android
Though they seem to have transitioned it to Flash since Flash became supported. Going to the iPlayer site without Flash installed gives you the message "To play this programme you need to download the Adobe Flash Player from the Android Market" and supplies a link (albeit labelled "Click here" suggesting that somebody on the web team can't think beyond their mouse) that goes directly to the relevant Google marketplace page directly in the browser, exposing a download button. The rest of the iPlayer site is the mobile version for small screens. So it's clear that they have somebody there who has thought about Android and ensured that the site works.
Compare that to the iPad where, also, the site works, and you get to pondering why anybody needs an app.
That would make sense
It's what the iPlayer app that already ships on Nokia phones (well, at least the N8) does.
I have absolutely no idea how the existence of a Nokia app plays into the wider picture — for all I know, Nokia developed it or otherwise paid for it.
The BBC are bound by their charter not to support the XBox...
... because Microsoft won't allow them to release for free and the charter won't allow them to charge. That's the officially given reason why they're on the Playstation and the Wii, but not the XBox.
Android devices aren't outselling iOS devices
Android phones are outselling iPhones — you're thinking of statistics directly from the mobile phone industry. If you add the iPod Touch and iPad to the iPhone, versus all Android phones + tablets then iOS remains ahead. This year's Android tablet push by multiple manufacturers should go a long way to redressing this disparity.
Surely the question should be the other way around?
Given that it already works just fine on all shipping iOS and Android devices, why bother with an iPlayer app for anyone?
Rating books more likely to be for Amazon's benefit?
You know, for the 'we've got some recommendations for you' part of their business? Which I guess is also to your benefit if you like to stick within your genres.
No Ovi Store?
Having recently checked it out, with support from at least EA, Gameloft and Rovio many of the same big name games that tend to dominate the other app stores seemed to be present on the Nokia. That said, I'm still acting under the impression that increased developer support is a good thing.
Thinking harder since last I posted, I think the risk to the Android path would be that they'd be back up to two Linux platforms even once the Maemo to MeeGo transition is complete. Probably a better path would be to put Dalvik on MeeGo, which probably wouldn't be too hard a port with the similar stuff underneath. That buys them entry to the Android ecosystem with a significant differentiating factor.
On an iPhone I copy and paste by (i) selecting the text; (ii) selecting 'copy' on the little sign board that pops up; (iii) putting the cursor somewhere else; and (iv) selecting 'paste'.
I multitask by (i) launching an application; (ii) going back to the home screen; (iii) launching another application.
For the record, I do pretty much identical things on my Nexus One. If you have no trouble doing basic things on a Nokia 5800 but can't figure those out, I guess you're used to looking for complexity that isn't there?