The problem is (from an app developer standpoint) is that there are too many variables in the android world to code an app once to run successfully across the ecosystem.
Yet strangely many people are successfully doing this.
You have to design a version for on-screen keyboards (because it'll use part of the screen real estate) separately from a version that uses a hardware keyboard. They don't need to be separate apps, but you need to design (visually at least) for both scenarios, or you end up locking out a good portion of the people who use android devices.
Completely wrong. Where are you getting this from?
Sure, there are 100,000+ android devices out there
Over 100,000 Android devices activated per day.
so in order to sell your app on all 100,000 of those phones you've got to tweak your app for each device
No you don't. You have no idea what you're talking about. And you're at Score:4 right now, which is shameful.
One way to call the Camera API
Ugh. You shouldn't have posted because almost everything you have said is just completely wrong.
The Android developer platform is extraordinarily universal. There's a density independent pixel format (which is how an app looks almost the same on a 320x480 screen as it looks on a 480x800 screen), support for varying screen ratios, a single way to inter-operate with the camera and send emails and read the GPS signal and get orientation signals, or even do advance OpenGL graphics.
One app to rule them all.
There are of course differences and occasionally "quirks". If you make a rich graphics game it's going to run terribly on a G1. Flash is only available on some devices. And of course if you have to target a newer API, presumably because it has a feature that you can't live without, you limit your app to that version and above (just as if I use Transactional Filesystem calls my Windows app would be Vista or newer).
Actually, he clarifies this in his comments on the blog post: Australians can't sell paid apps on the android marketplace yet.
Google is a terrible retailer. I love my Nexus One, but buying it from Google demonstrated that they aren't good at selling things, or being a conduit for selling things.
The marketplace is a good example. Us Canadians only recently got the ability to actually buy pay apps, and of course we (like most of the world) still don't have the ability to sell apps, at least with Google handling the transaction.
Which is why many apps have gone to either ad support, or some sort of activation key that you buy from a more world-capable transaction enabler like PayPal.
but when did they actually start work on a full face touch screen interface with a real browser?
Well the "real browser" was a part of the original implementation. A part of the reason Google started the Chrome project was because they wanted a browser everywhere. It was actually a critical part of their strategy since day one.
As according to wikipedia, the touch keyboard wasn't added to Android till 1.5 which was displayed till two years after the iPhone, makes me wonder what Android would have looked like if the iPhone had never come out.
It would have, absolutely, been much different. The original prototypes were very blackberry-esque. But here's the funny thing -- in the coming months we'll see Android strongly move back to that, having over-exerted on the "keyboardless" model. Personally I would rather have a good physical keyboard, coupled with appropriate on-screen touch elements.
On screen keyboards are not fun to type on, and they consume much of your limited screen real-estate unnecessarily.
But if your point is that Android ripped off what iPhone pioneered, to a degree that is absolutely true. It's also true that the favor is being returned. It's truer that Apple had a long litany of prior devices that they borrowed from.
As Steve Jobs himself said, tech is about stealing the ideas of others. Lately people have gone a little Kim Jong-il about him, however, and seem to have conveniently forgotten that.
Slashdot used to sneer at Apple and Apple devices. It is a relatively recent phenomena that Slashdot became so Apple-centric.
It's actually a bit bizarre because Slashdot's very pro-Apple slant seemed to take off just as disappeared from other sites like Hacker News. It's like all the Apple faithful migrated here.
So how, exactly, do you rationalize seriously saying that it came about because of the iPhone?
No doubt Google has stolen some elements from the iPhone, just as the iPhone stole from many other devices. Yet it is infuriating seeing history rewritten, North Korea-style, until Steve Jobs invented the Internet, the stars and heavens, and so on.
BTW - You should send Google some thanks, as a wide range of features queued up for the iPhone 4 OS were cribbed directly from Android.
And every time I come away feeling like Android would be a great OS on a larger form factor where the increased power it gives makes more sense, but not so great for a phone.
Interesting take. You know that the iPhone's OS was derived from OSX, right? Do you have any logical justification for your ridiculous feeling?
At a minimum, I need Android to give me the ability to easily quit an application when I leave it rather than just dumping it into the background where I'll need to launch a task manager application to finally get rid of it. At I/O, I used both my iPhone and the Droid they sent us prior to the conference.
How could you possibly have gone to a Google I/O event yet remain so ignorant? I seriously find this deeply suspicious. Yes, the whole tech world knows the free phones they gave out, so it hardly gives your words authority.
There is a widespread lack of knowledge of how Android's multitasking works among the technically ignorant, however in practical essence the app *does* quit the second you leave the activity (it isn't doing anything in the background, and any resources it consume(d) are primed for the plucking). This is the case for the overwhelming majority of applications that aren't spinning off services (which themselves live in virtual isolation). The upcoming addition of multitasking in the iPhone OS is a virtual clone of Android's long-existing multitasking.
I appreciate that end users don't know this, because they don't need to. They never need to manage what is running. But someone who claims to have gone to Google I/O? I don't buy it.
But when I was at WWDC 2009, during Phil Schiller's keynote, boy do I remember when they said "iPhones stink and are for poo poo heads". Couldn't believe it. I was there, right?
By that you mean that you'll take Android's version, right? Apple's version is an almost direct copy of Android's multitasking. The only real difference is that in Android you can spin off a service without getting the blessing of the king, but for general functionality and design they're 1:1.
These poor bastards are going to be burned at the afraid-of-sexuality stake
This sounds very erudite and post-contemporary, but it's also nonsensical cruft.
Playing Tetris is slacking off. Browsing porn at work is a sign of really, really questionable, almost "flaming out" judgment.
Re:Buying ARM for a leg?
on
Apple To Buy ARM?
·
· Score: 2, Insightful
Yuk yuk yuk.
It's "factual" in a purely paper sense. Apple had nothing to do with the creation of ARM the engineering team, the original processors, and so on. As pydev noted, Apple's participation was limited to being an investor when Acorn spun off ARM. It's essentially like saying the US Government invented the Chevy brand because now they are a part owner.
Boo hoo, Adobe. Apple doesn't have to support Adobe products on their platform.
They don't have to support anything. Adobe is the one that did the heavy lifting here and Apple has no onus at all. Apple did EXTRA work (and exposed themselves to ridicule and serious antagonism) to try to block Adobe, and it has bit them in the ass.
As to the other arguments, if "lowest common denominator" apps ran poorly, didn't use the hardware to its most, etc, then "native" superior options would win in that market. The fact that people seem to jump to conclusions on this, backed by nothing, is absurd.
Apple is perfectly within their rights to not allow programs that will then run arbitrary code on their devices.
Arbitrary code? WTF does that mean?
No, seriously, there is a stunning amount of technical ignorance about this whole subject, so I would really like to know what you think you are saying.
Apple provides a runtime for binary code. The runtime happens with a sandbox. Apple's toolchain produces binaries for that sandbox, but provides zero checks, constraints, etc: You can go hog wild writing it in inline ARM assembly if you want.
Adobe produced something that also targets that binary sandbox. Sorry, but there is zero technical legitimacy to Apple, or your, arguments in this case. You seem to be assuming that Apple's toolchain somehow guarantees efficient, secure code, which is nonsense.
Teradata and the other big relational db products (vertical, greenplum, etc) are all _analytical_ databases
Many of the very expensive database products focus on analytics because that is where the big, big money is.
Greenplum, however, is essentially a clusterable version of Postgresql. Column-oriented tables are an option for your table, but otherwise it's just a really, really scalable version of the open source product.
where adding new data to the system takes minutes if not hours
Not even Microsoft have been bastardly enough to so blatantly limit the user's freedom like Apple and Google are trying so hard to do.
I didn't watch the video, however I'm assuming that your post is of the "take the words about the iPad and turn it against Chrome OS" ilk.
If so, you are misdirected.
Here's the thing - from all that I've heard about the ChromeOS platform, it is intended to be usable only as essentially a web appliance. The apps that it uses are no different than the apps that anyone can use on any other modern web device, including, humorously, the iPad.
Your bank won't gloat about their new "ChromeOS Banking Application", nor will your doctor, etc. The public won't be bifurcated into various walled gardens.
Now personally I have zero interest in Chrome OS: Give me a real notebook any day. But I can see its purpose, and I don't see the same (gigantic) pitfalls that I see with the iPad.
It's a "carrier-neutral" chip, so you can activate the device on whatever carrier you like - GSM or CDMA.
Unlocking a phone makes it carrier neutral. What you're talking about is being communication standard neutral simply by supporting multiple standards, but that is increasingly a non-standard (CDMA was largely a North America only thing, but is increasingly a US-only thing -- basically Verizon -- after Bell and Telus in Canada left CDMA for GSM).
All of this having little to do with financing your phone.
Unlocked phones sold as devices by themselves would be *wonderful*. I'm hoping that this actually gets carried out.
But I doubt it. Firstly a lot of people are making hay about the fact that this phone is "designed by Google", but so was the G1 essentially (also branded as Google, given to Google employees, called a Google phone, etc). So it seems like a refresh of the position the G1 held and people are extrapolating a little too much.
Seriously, supposedly the reason apple has assiduously avoided background processes not to mention creating a centralized push notifier was to avoid battery life drain and process management headaches.
Do you believe their excuse? Somehow the iPhone has a battery life equal to or worse than a Droid, which is of course a phone with no such constraints.
Apple wants to control the experience and the ecosystem. It's as simple as that. Everything else is just noise.
It reminds me when Linux had to be custom compiled for every host, the necessities for the platform (e.g. network, video, etc) needing to be compiled into the kernel. At the time this was held up as some sort of great feature. Of course really it was just a technical deficiency and when they moved beyond it suddenly its imagined advantages were forgotten. The same for when the Apple PC world used PowerPC chips, those superstar processors that put the x86 world to shame. Then Apple went x86 and it was forgotten.
Only they attempt to be the only solution for everything. By your own logic why didn't the LSE go with an in-house.NET solution?
The fact that Microsoft held up the LSE as an example doesn't somehow excuse gross intellectual dishonesty in swinging entirely the other way. This submission specifically says that the LSE "rejects.NET for open source", when really they switched from a consulting co to running it in house through an acquisition. Oh but it doesn't stop there in parading ridiculous hyperbole, but goes on to say "pretty savage indictment of the costs of a complex.Net system" when it is nothing of the sort.
This is the sort of garbage that Slashdot suffered from years ago, so to see it still held up is honestly depressing. Didn't these people grow up and detach themselves a bit?
All of which contributes to overall system performance. Many ingredients a meal does have.
Yes, Linux is infinitely configurable. That's why I have it on my phone. That's why I have it on my NAS. That's why I have it on my PVR. However not only do I doubt that the company in question has customized it to much degree, a principal, extremely central part of their system is Oracle, which of course is the antithesis of open source.
This submission and many of the responses again herald back to years ago. Guess what, kids: Linux has already arrived. Linux runs lots of critical systems. This isn't some big revelation.
For which they spent $30 million on the last solution, and an up-front $10 million on this one, to who knows what ends, with expectations of a 3-unknown length deployment, running in parallel.
I agree that it superficially should be incredibly trivial. But it isn't.
Really? Where do you read that? Because, you know, no where does it say that. And maybe you're new to the industry, but metrics aren't always backed by the scientific method...
The article states quite clearly that the accenture/.net system could only achieve 2.7ms/transaction. It also states the system they bought achieves.4ms/transaction. So sure the LSE CTO could just be lying to make himself look good for spending $30 million to buy the company that built this platform...
The LSE has a bizarre technical leadership that tends to be very flamboyant about everything they do (remember the numbers they rolled around when switching to the.NET solution). Yes, I absolutely believe that it would go something like "estimate what the performance difference would be?" "Oh...I dunno...we'll have to set up a load testing enviro..." "No, just estimate...but make it good"......
You call yourself a realist... yet realistic perspective is dependent upon knowledge of the subject. It's well known that most trading platforms are faster than the piece of crap they had on the LSE... often more than 25ms faster, which means that it was faster to trade on Euronext.
Right. Yet even the LSE, in pimping this purchase, pegs their current trading times at less than 3ms. But...you know....Red Flayer says that the others are 25ms faster...so...uh...They pretrade. Must be designed by GS.
MilleniumIT has an established product in deployment in several exchanges. Their transaction times have been demonstrably faster than the POS that MS/Accenture developed on.NET framework
You keep making these claims based upon apparently nothing, vaguely waving your hands at the all-knowing Google. How about you provide some solid links?
"The sky is purple. Geeze....just google it! See, I've proven myself."
Failing that, are you telling us that you personally benchmarked this system running on the LSE's stack, with their trading load, and you've demonstrated this? Because, of course, you haven't.
Although, to be fair, it's also quite possible that what we have here are leftover MS astroturfers from the earlier anti-MS article. That would explain a lot, but maybe I'm just being paranoid again.
Paranoid, or just unable to see reality if it punched you in the face given how so utterly focused on your fanaticism you are.
Horseshit. This is switching from "Accenture writing a slow unstable trading platform with.NET via cheap labor in India" to "buying the company that produces a fast, stable platform on Linux via cheap labor in Sri Lanka".
And you're the one telling me to use Google?
When the LSE switched to the.NET solution, they heralded the fact that it dropped them from 130ms+ to 10ms trading times. Now that they're switching from.NET to this new solution, they're saying it brings them from 2.7ms to 0.4ms.
As far as reliability, purportedly the LSE had a single day of problems caused by never qualified reasons. That was enough for many to go "AHA!.NET!", but of course that's because they're ignorant morons who immediately demonstrate how little their opinion is worth. Failures at banks, airports, national systems, and so on, have happened on the gamut of platforms and systems, so it's marvelously telling when people demonstrate their ignorance.
Slashdot Mirror
Yet strangely many people are successfully doing this.
Completely wrong. Where are you getting this from?
Over 100,000 Android devices activated per day.
No you don't. You have no idea what you're talking about. And you're at Score:4 right now, which is shameful.
Ugh. You shouldn't have posted because almost everything you have said is just completely wrong.
The Android developer platform is extraordinarily universal. There's a density independent pixel format (which is how an app looks almost the same on a 320x480 screen as it looks on a 480x800 screen), support for varying screen ratios, a single way to inter-operate with the camera and send emails and read the GPS signal and get orientation signals, or even do advance OpenGL graphics.
One app to rule them all.
There are of course differences and occasionally "quirks". If you make a rich graphics game it's going to run terribly on a G1. Flash is only available on some devices. And of course if you have to target a newer API, presumably because it has a feature that you can't live without, you limit your app to that version and above (just as if I use Transactional Filesystem calls my Windows app would be Vista or newer).
No they aren't. Android 1.6+ has a very robust way of dealing with virtually any screen size.
http://developer.android.com/guide/practices/screens_support.html
Google is a terrible retailer. I love my Nexus One, but buying it from Google demonstrated that they aren't good at selling things, or being a conduit for selling things.
The marketplace is a good example. Us Canadians only recently got the ability to actually buy pay apps, and of course we (like most of the world) still don't have the ability to sell apps, at least with Google handling the transaction.
Which is why many apps have gone to either ad support, or some sort of activation key that you buy from a more world-capable transaction enabler like PayPal.
Well the "real browser" was a part of the original implementation. A part of the reason Google started the Chrome project was because they wanted a browser everywhere. It was actually a critical part of their strategy since day one.
It would have, absolutely, been much different. The original prototypes were very blackberry-esque. But here's the funny thing -- in the coming months we'll see Android strongly move back to that, having over-exerted on the "keyboardless" model. Personally I would rather have a good physical keyboard, coupled with appropriate on-screen touch elements.
On screen keyboards are not fun to type on, and they consume much of your limited screen real-estate unnecessarily.
But if your point is that Android ripped off what iPhone pioneered, to a degree that is absolutely true. It's also true that the favor is being returned. It's truer that Apple had a long litany of prior devices that they borrowed from.
As Steve Jobs himself said, tech is about stealing the ideas of others. Lately people have gone a little Kim Jong-il about him, however, and seem to have conveniently forgotten that.
Slashdot used to sneer at Apple and Apple devices. It is a relatively recent phenomena that Slashdot became so Apple-centric.
It's actually a bit bizarre because Slashdot's very pro-Apple slant seemed to take off just as disappeared from other sites like Hacker News. It's like all the Apple faithful migrated here.
Android -- an existing mobile company -- was bought by Google in 2005.
So how, exactly, do you rationalize seriously saying that it came about because of the iPhone?
No doubt Google has stolen some elements from the iPhone, just as the iPhone stole from many other devices. Yet it is infuriating seeing history rewritten, North Korea-style, until Steve Jobs invented the Internet, the stars and heavens, and so on.
BTW - You should send Google some thanks, as a wide range of features queued up for the iPhone 4 OS were cribbed directly from Android.
Interesting take. You know that the iPhone's OS was derived from OSX, right? Do you have any logical justification for your ridiculous feeling?
How could you possibly have gone to a Google I/O event yet remain so ignorant? I seriously find this deeply suspicious. Yes, the whole tech world knows the free phones they gave out, so it hardly gives your words authority.
There is a widespread lack of knowledge of how Android's multitasking works among the technically ignorant, however in practical essence the app *does* quit the second you leave the activity (it isn't doing anything in the background, and any resources it consume(d) are primed for the plucking). This is the case for the overwhelming majority of applications that aren't spinning off services (which themselves live in virtual isolation). The upcoming addition of multitasking in the iPhone OS is a virtual clone of Android's long-existing multitasking.
I appreciate that end users don't know this, because they don't need to. They never need to manage what is running. But someone who claims to have gone to Google I/O? I don't buy it.
But when I was at WWDC 2009, during Phil Schiller's keynote, boy do I remember when they said "iPhones stink and are for poo poo heads". Couldn't believe it. I was there, right?
By that you mean that you'll take Android's version, right? Apple's version is an almost direct copy of Android's multitasking. The only real difference is that in Android you can spin off a service without getting the blessing of the king, but for general functionality and design they're 1:1.
This sounds very erudite and post-contemporary, but it's also nonsensical cruft.
Playing Tetris is slacking off. Browsing porn at work is a sign of really, really questionable, almost "flaming out" judgment.
Yuk yuk yuk.
It's "factual" in a purely paper sense. Apple had nothing to do with the creation of ARM the engineering team, the original processors, and so on. As pydev noted, Apple's participation was limited to being an investor when Acorn spun off ARM. It's essentially like saying the US Government invented the Chevy brand because now they are a part owner.
They don't have to support anything. Adobe is the one that did the heavy lifting here and Apple has no onus at all. Apple did EXTRA work (and exposed themselves to ridicule and serious antagonism) to try to block Adobe, and it has bit them in the ass.
As to the other arguments, if "lowest common denominator" apps ran poorly, didn't use the hardware to its most, etc, then "native" superior options would win in that market. The fact that people seem to jump to conclusions on this, backed by nothing, is absurd.
Arbitrary code? WTF does that mean?
No, seriously, there is a stunning amount of technical ignorance about this whole subject, so I would really like to know what you think you are saying.
Apple provides a runtime for binary code. The runtime happens with a sandbox. Apple's toolchain produces binaries for that sandbox, but provides zero checks, constraints, etc: You can go hog wild writing it in inline ARM assembly if you want.
Adobe produced something that also targets that binary sandbox. Sorry, but there is zero technical legitimacy to Apple, or your, arguments in this case. You seem to be assuming that Apple's toolchain somehow guarantees efficient, secure code, which is nonsense.
The Nexus One has 512MB, and of course a ARM-based snapdragon. Many other current ARM Android phones have 512MB+.
Many of the very expensive database products focus on analytics because that is where the big, big money is.
Greenplum, however, is essentially a clusterable version of Postgresql. Column-oriented tables are an option for your table, but otherwise it's just a really, really scalable version of the open source product.
They load many TB per hour.
I didn't watch the video, however I'm assuming that your post is of the "take the words about the iPad and turn it against Chrome OS" ilk.
If so, you are misdirected.
Here's the thing - from all that I've heard about the ChromeOS platform, it is intended to be usable only as essentially a web appliance. The apps that it uses are no different than the apps that anyone can use on any other modern web device, including, humorously, the iPad.
Your bank won't gloat about their new "ChromeOS Banking Application", nor will your doctor, etc. The public won't be bifurcated into various walled gardens.
Now personally I have zero interest in Chrome OS: Give me a real notebook any day. But I can see its purpose, and I don't see the same (gigantic) pitfalls that I see with the iPad.
Unlocking a phone makes it carrier neutral. What you're talking about is being communication standard neutral simply by supporting multiple standards, but that is increasingly a non-standard (CDMA was largely a North America only thing, but is increasingly a US-only thing -- basically Verizon -- after Bell and Telus in Canada left CDMA for GSM).
All of this having little to do with financing your phone.
Unlocked phones sold as devices by themselves would be *wonderful*. I'm hoping that this actually gets carried out.
But I doubt it. Firstly a lot of people are making hay about the fact that this phone is "designed by Google", but so was the G1 essentially (also branded as Google, given to Google employees, called a Google phone, etc). So it seems like a refresh of the position the G1 held and people are extrapolating a little too much.
Do you believe their excuse? Somehow the iPhone has a battery life equal to or worse than a Droid, which is of course a phone with no such constraints.
Apple wants to control the experience and the ecosystem. It's as simple as that. Everything else is just noise.
It reminds me when Linux had to be custom compiled for every host, the necessities for the platform (e.g. network, video, etc) needing to be compiled into the kernel. At the time this was held up as some sort of great feature. Of course really it was just a technical deficiency and when they moved beyond it suddenly its imagined advantages were forgotten. The same for when the Apple PC world used PowerPC chips, those superstar processors that put the x86 world to shame. Then Apple went x86 and it was forgotten.
The fact that Microsoft held up the LSE as an example doesn't somehow excuse gross intellectual dishonesty in swinging entirely the other way. This submission specifically says that the LSE "rejects .NET for open source", when really they switched from a consulting co to running it in house through an acquisition. Oh but it doesn't stop there in parading ridiculous hyperbole, but goes on to say "pretty savage indictment of the costs of a complex .Net system" when it is nothing of the sort.
This is the sort of garbage that Slashdot suffered from years ago, so to see it still held up is honestly depressing. Didn't these people grow up and detach themselves a bit?
Yes, Linux is infinitely configurable. That's why I have it on my phone. That's why I have it on my NAS. That's why I have it on my PVR. However not only do I doubt that the company in question has customized it to much degree, a principal, extremely central part of their system is Oracle, which of course is the antithesis of open source.
This submission and many of the responses again herald back to years ago. Guess what, kids: Linux has already arrived. Linux runs lots of critical systems. This isn't some big revelation.
What information did they present?
None. Shill account back-ups or not, simply pretending that some valuable info was conveyed doesn't make it so.
For which they spent $30 million on the last solution, and an up-front $10 million on this one, to who knows what ends, with expectations of a 3-unknown length deployment, running in parallel.
I agree that it superficially should be incredibly trivial. But it isn't.
Do people still seriously use "M$" on Slashdot? I thought that died out when people grew up.
And to call me a Microsoft apologist is humorous.
Really? Where do you read that? Because, you know, no where does it say that. And maybe you're new to the industry, but metrics aren't always backed by the scientific method...
The LSE has a bizarre technical leadership that tends to be very flamboyant about everything they do (remember the numbers they rolled around when switching to the .NET solution). Yes, I absolutely believe that it would go something like "estimate what the performance difference would be?" "Oh...I dunno...we'll have to set up a load testing enviro..." "No, just estimate...but make it good".. ....
Right. Yet even the LSE, in pimping this purchase, pegs their current trading times at less than 3ms. But...you know....Red Flayer says that the others are 25ms faster...so...uh...They pretrade. Must be designed by GS.
You keep making these claims based upon apparently nothing, vaguely waving your hands at the all-knowing Google. How about you provide some solid links?
"The sky is purple. Geeze....just google it! See, I've proven myself."
Failing that, are you telling us that you personally benchmarked this system running on the LSE's stack, with their trading load, and you've demonstrated this? Because, of course, you haven't.
Paranoid, or just unable to see reality if it punched you in the face given how so utterly focused on your fanaticism you are.
And you're the one telling me to use Google?
When the LSE switched to the .NET solution, they heralded the fact that it dropped them from 130ms+ to 10ms trading times. Now that they're switching from .NET to this new solution, they're saying it brings them from 2.7ms to 0.4ms.
As far as reliability, purportedly the LSE had a single day of problems caused by never qualified reasons. That was enough for many to go "AHA! .NET!", but of course that's because they're ignorant morons who immediately demonstrate how little their opinion is worth. Failures at banks, airports, national systems, and so on, have happened on the gamut of platforms and systems, so it's marvelously telling when people demonstrate their ignorance.