I do motorpool for a company of about 800-1000 people. All of our executives and corporate staff wanna drive their goddammed Mustangs, Mercedes, Smart Cars and other personal wotsits or doo-dads to their meetings. Enough is a-freakin-nuff! We're a corporation and we need to maintain stability and compatibility over fancy and chic. You get a Lincoln Town Car. With FM Radio. And a driver... if you're lucky. Oh, and don't get me wrong... it's not like I'm being elitist or something. I love these new cars for home use. I drive all kinds of cars. But they belong AT HOME!
(Are you certain you're not trying to justify your own importance here? Your company gives people money and somehow, by some mystical force, they figure out how to purchase their own gas and oil, and arrive where they need to be without an in-house support organization.)
The problem is not mediocre apps that get no sells, it's the extremely good apps that cost a lot of money to make and get no visibility anywhere (its really really hard to make it to a top list), making it a big gamble.
If it cost a lot of money to make, you'd better protect your investment with marketing, ad buys, banners and good placement. Simply making the best thing isn't sufficient when you're selling to consumers -- they don't have the time to evaluate their options, conduct searches and exhaustively research, apps are impulse buys and move on emotional gratification. Take a page from the entertainment industry: half your budget has to be dedicated to telling people why they need your app.
(The first was the LG Prada. Apparently it had software limitations that made swipe gestures impractical due to the hurried development.)
Software development seems to be one of the few places where we're happy to give people brownie points for creating the first in a line of things, even when the first attempt didn't remotely work, wasn't practical, useable, or saleable. Thus, people still go around claiming that the LG Prada was the first of anything, when it was really just a brick with a GSM baseband and Flash interpreter (or, in other words, a brick).
But apparently, the fact that an LG Prada exists is supposed to be evidence that Apple can't patent swipe-to-unlock, not because the Prada had it, but because the Prada might have had it, or the existence of the Prada suddenly made anything you put on a touchscreen phone obvious (even though touchscreen phones existed for the previous decade), or something.
I think the real problem with software patents is that not enough people own them. They're all owned by a few very large corporations and easily sold, if license fees and royalties were raining down directly upon the hundreds of worker bees that actually invented this thing and that (more like copyrights), software developers would probably be much more sanguine.
It definitely depends on the ambient light where you're working if there's a tradeoff or not. Wether a matte finish oversaturates depends on if the screen's been calibrated properly, but they're definitely dimmer for the same backlight wattage -- glossy screens definitely save battery life.
If you want Android, you have a multitude of devices to choose from and it comes down to niche differences which one you will choose.
That's not really true, if you want Android, and you don't want one or either of (1) 4G, (2) a replaceable battery, (3) a 4" AMOLED screen, or (4) SD slots, you really don't have many choices, and the choices you have are really bad, second run phones with no software updates, often no Cyanogenmod support, and lower-tier CPUs and graphics.
It's remarkable that Android gives phone makers the ability to make basically whatever phone they'd want, but they don't make good candybar hardware keyboard phones anymore, they don't make good integrated battery phones, and the ones they do make are always hamstrung with last year's CPU, and their ideas about screen size and screen tech are remarkable in their monotony.
It's interesting that Android manufacturer's refuse to actually compete with the iPhone for its features; if they made a phone with an integrated battery, 3.5" screen, no moving parts and the best CPU available it would no doubt fail miserably, it could never compete with an iPhone. This is probably also reflected in the fact that the mobile manufacturers has basically ceded the entire market for wifi-only PDA/PMP handsets to the iPod Touch, no one makes anything remotely competitive.
(I admit I'm sort of weird, after my experience with the Treo 650 I made an oath to never buy a phone with a replaceable battery or SD slot again. I wish I could find something like my old Treo 290 that would run the Android. Despite Android's openness and potential for variety, Sammy, HTC and Motorola seem content to just manufacture the same phone with a slightly tweaked case design.)
So a unit shipped is a unit sold, because no business ships a unit without someone bloody paying for it.
The problem with this thinking is that if you're an application developer, or a cellular carrier, you can't count a phone in a warehouse as a customer, and you have to handicap them being potential customers, depending on if your customers actually want the phone. To a carrier or an electronics store, you've got phones and Samsung has the cash, but if the phones are undesirable phones and you have to liquidate half of the lot below wholesale, you're probably not going to reorder those phones.
Samsung reports shipped phones because that's their bottom line, but Samsung making money doesn't do anyone but Samsung any good. A "shipped phone" is a necessary but not sufficient good to a mobile developer, a mobile user/customer or a cellular carrier, and you shouldn't come to conclusions about what OS is better for you to develop for, or what phone you should buy, based on who is "shipping" more phones.
A "shipped" metric is pointless, unless you're a Sammy shareholder. If you're a developer, you should look for activation numbers, real sale numbers, the install base, the demographics of the install base, usage statistics (like web browser stats)...
If someone is only looking for a low-end smartphone, they aren't in the market for an iPhone and, if they didn't buy an iPhone, that wouldn't really be a lost sale for Apple, since they were never in the running.
This theory doesn't really hold, though, since you can now buy iPhones 3GS's and 4's for $0-$99 on contract, and a year-old iPhone would generally be comparable (in certain measures) to a middle-end Samsung.
OLEDs have terrible color, its over-saturated and makes people think the image looks better in a subjective way. ((S)AM)OLED is to cellphones as glossy screens are to laptops. Do a Pepsi challenge with a iPhone or even an original Droid, versus a Galaxy Nexus in a dark room, versus watching the image on a proper screen.
Pentile doesn't help either with the blacks or contrasty images.
And while people in Slashdot screams that iPhones are cheap (which are not), there's still people in many places that cannot afford or will not pay for them.
There are iPhones that are free on contract at this point. They have a ladder in hundred dollar increments now from $0 up to $399.
Re:Just seems like a well thought out list
on
The RMS Tour Rider
·
· Score: 2
Still doesn't explain why they needed two tubes of KY jelly in the dressing room.:)
This is biased logic, too. I can bias it the other way: if you're an application developer and you want a platform feature that is not yet released, and you need to wait for 6 months instead of one year, then you will get half your market six months sooner.
You shouldn't have to wait at all. 100% of your available market should have the new OS, or at least no encumbrance to acquiring it, and the platform vendor should aggressively push for adoption. There should optimally only be one target platform, and you shouldn't have to resort to functionality hacks that might have to get completely rewritten when the new version comes out. Apps are impulse buys, and if someone has to take more than even the most rudimentary steps to use your app, you'll lose a sale.
If you're talking about some sort of relative gain between platforms, none of your users is going to switch which phone he uses because of your app. If you have an iPhone app and you're afraid that you're losing sales to your competitor on Android, I assure you this isn't the case. No phone user is going to switch platforms to get your competitor's app, apps just aren't that important on an individual basis, particularly if there's only a 6 month differential.
I can tell you the worst limitation is the hardware, not the software.
I can see how this would be a good argument in the abstract, but we're talking about two phone platforms, one which has three handset models and one tablet model currently selling, and another which has a dozen models currently selling and a dozen more sold in the last three years, and more than a few tablets and e-readers of widely varying equipment.
But here's the thing that Apple fans are leaving out -- Apple is PAINFUL when it comes to their Mac OS X and what hardware it will (easily) install to. I recall back when G4 and G5 were new and which versions of Mac OS X would work on what hardware.
You're comparing installing a modern desktop OS on, in what that case was, a three year old computer with installing a modern cellphone OS on an 11-month old handset.
If you're an application developer and you're using a platform feature in the latest OS, if half of your potential customers are stuck a version behind it doesn't matter how many releases happen a year, you just lost half your market, for at least several months. "Release early, release often" is good for applications but for platforms it causes fragmentation and disruptions and all sorts of bad disincentives for developers.
For platform users, yes it's a different proposition.
My wife has never upgraded her HTC Aria to the current OS, while I have. Why hasn't she??? THERE WAS NO NEED TO.
On the individual basis there might not be a need to upgrade for a lot of people, but it's terrible if you're looking at Android from the standpoint of the developer. Want Fragments UI? Want low-latency audio? Want to integrate NFC beaming? Want to integrate with the calendar or visual vociemail? Or anything else?
If your app wants to merely use any of these, you'll have to maintain separate versions, builds, and perhaps even codebases. If your app would require any of these to do its magic, you're going to be locked out from wide swathes of the market at any one time.
Jobs solved the innovation dilemma by having a lot of engineers circled around him.
The fact that no other corporate CEO surrounds himself with engineers, designers and implementors should be instructive. I mean you make it sound easy but why on earth are most of the other tech companies led by the nose by their sales and marketing people?
You know, as a devout capitalist, I've always believed that customers service comes first.
A capitalist of the Adam Smith variety would say that profit comes first, and that good customer service and mutual benefit is a consequence of pursuing profit.
The fact that this doesn't work under a lot of different contexts, particularly the ones that Harvard MBAs get themselves learned in, is the guts of the story.
The other problem is the presence of search data when clicking through to unencrypted sites, if they are google customers. That means google's SSL service is a lie and your unencrypted searches will be sent to certain customers regardless of using http or https.
It seems sorta common sense that if you click on a link to a site, that site will know you clicked on it and where you're going. Similarly, if you have a cookie on a site, that site will know when you've been there and will be able to correlate all kinds of things you typed into that site with links, etc.
Google possesses this information, they can sell it. That your request travelled over HTTPS means it's secret between you and google, what either side of the transmission does with the information it obtained is strictly the business of either party.
Anything you can do with a search result from https://google.com/, like for instance, sharing a search result with a friend, google can do with your click stream, like, for instance, sharing it with their friend.
The CDC data [3] shows some serious concerns, 20,096 reaction reports including 71 verified deaths over 40 million doses. Of the 20,096 reports, 1,607 (8%) are considered serious.
That's a 0.05% rate of reported reactions per dose, 0.004% of which are considered serious, compared to your 0.16% total mortality in the population. Such a risk, taken by itself, is comparable to the risk of injury or death taking the stairs to the second floor of your house, versus taking a safer elevator.
You also need to consider the morbidity numbers for cervical cancer, surgical and post-operative complications, and deaths from other cancers secondary to cervical cancer. You have also left out other cancers caused by HPV, including oral, throat, and anal cancers in men and women -- a man giving oral sex to one woman (just one, in his entire life) doubles his chances of contracting HPV, and a man with more than five partners is 8.6x more likely.
I'm too old, like most of the people here, including you, to get an HPV vaccine, so I'm really not trying to justify any decision on my behalf, but I can run numbers like anyone else.
So what, exactly, are Bill's credentials to talk about education?
He's am entrepreneur who's made more money than the median pizza chain CEO. In the United States, that's what passes for civic leader and elite opinionmaker.
16GB is an awful lot of RAM, I'm really curious as to what it is that they're doing that's going to require more RAM than most of these devices have in total storage space.
These are the hardware requirements to compile the complete AOSP Android system and platform, and not the requirements to merely develop an application on it.
Slashdot Mirror
(Are you certain you're not trying to justify your own importance here? Your company gives people money and somehow, by some mystical force, they figure out how to purchase their own gas and oil, and arrive where they need to be without an in-house support organization.)
The problem is not mediocre apps that get no sells, it's the extremely good apps that cost a lot of money to make and get no visibility anywhere (its really really hard to make it to a top list), making it a big gamble.
If it cost a lot of money to make, you'd better protect your investment with marketing, ad buys, banners and good placement. Simply making the best thing isn't sufficient when you're selling to consumers -- they don't have the time to evaluate their options, conduct searches and exhaustively research, apps are impulse buys and move on emotional gratification. Take a page from the entertainment industry: half your budget has to be dedicated to telling people why they need your app.
It is a very old meme, the famous Kottke I-don't-want-a-holy-war-here post, and it was a 17 meg file, over 20 minutes. On an 8600 in 1998.
(The first was the LG Prada. Apparently it had software limitations that made swipe gestures impractical due to the hurried development.)
Software development seems to be one of the few places where we're happy to give people brownie points for creating the first in a line of things, even when the first attempt didn't remotely work, wasn't practical, useable, or saleable. Thus, people still go around claiming that the LG Prada was the first of anything, when it was really just a brick with a GSM baseband and Flash interpreter (or, in other words, a brick).
But apparently, the fact that an LG Prada exists is supposed to be evidence that Apple can't patent swipe-to-unlock, not because the Prada had it, but because the Prada might have had it, or the existence of the Prada suddenly made anything you put on a touchscreen phone obvious (even though touchscreen phones existed for the previous decade), or something.
I think the real problem with software patents is that not enough people own them. They're all owned by a few very large corporations and easily sold, if license fees and royalties were raining down directly upon the hundreds of worker bees that actually invented this thing and that (more like copyrights), software developers would probably be much more sanguine.
It definitely depends on the ambient light where you're working if there's a tradeoff or not. Wether a matte finish oversaturates depends on if the screen's been calibrated properly, but they're definitely dimmer for the same backlight wattage -- glossy screens definitely save battery life.
I think that having the image look better in a subjective way is exactly what I want. I don't need hyper-accurate colors on my smartphone.
Behold, readers, the triumph of marketing and showroom appeal, over, like, good engineering.
As I said, glossy screen of the cellphone world.
If you want Android, you have a multitude of devices to choose from and it comes down to niche differences which one you will choose.
That's not really true, if you want Android, and you don't want one or either of (1) 4G, (2) a replaceable battery, (3) a 4" AMOLED screen, or (4) SD slots, you really don't have many choices, and the choices you have are really bad, second run phones with no software updates, often no Cyanogenmod support, and lower-tier CPUs and graphics.
It's remarkable that Android gives phone makers the ability to make basically whatever phone they'd want, but they don't make good candybar hardware keyboard phones anymore, they don't make good integrated battery phones, and the ones they do make are always hamstrung with last year's CPU, and their ideas about screen size and screen tech are remarkable in their monotony.
It's interesting that Android manufacturer's refuse to actually compete with the iPhone for its features; if they made a phone with an integrated battery, 3.5" screen, no moving parts and the best CPU available it would no doubt fail miserably, it could never compete with an iPhone. This is probably also reflected in the fact that the mobile manufacturers has basically ceded the entire market for wifi-only PDA/PMP handsets to the iPod Touch, no one makes anything remotely competitive.
(I admit I'm sort of weird, after my experience with the Treo 650 I made an oath to never buy a phone with a replaceable battery or SD slot again. I wish I could find something like my old Treo 290 that would run the Android. Despite Android's openness and potential for variety, Sammy, HTC and Motorola seem content to just manufacture the same phone with a slightly tweaked case design.)
So a unit shipped is a unit sold, because no business ships a unit without someone bloody paying for it.
The problem with this thinking is that if you're an application developer, or a cellular carrier, you can't count a phone in a warehouse as a customer, and you have to handicap them being potential customers, depending on if your customers actually want the phone. To a carrier or an electronics store, you've got phones and Samsung has the cash, but if the phones are undesirable phones and you have to liquidate half of the lot below wholesale, you're probably not going to reorder those phones.
Samsung reports shipped phones because that's their bottom line, but Samsung making money doesn't do anyone but Samsung any good. A "shipped phone" is a necessary but not sufficient good to a mobile developer, a mobile user/customer or a cellular carrier, and you shouldn't come to conclusions about what OS is better for you to develop for, or what phone you should buy, based on who is "shipping" more phones.
A "shipped" metric is pointless, unless you're a Sammy shareholder. If you're a developer, you should look for activation numbers, real sale numbers, the install base, the demographics of the install base, usage statistics (like web browser stats)...
If someone is only looking for a low-end smartphone, they aren't in the market for an iPhone and, if they didn't buy an iPhone, that wouldn't really be a lost sale for Apple, since they were never in the running.
This theory doesn't really hold, though, since you can now buy iPhones 3GS's and 4's for $0-$99 on contract, and a year-old iPhone would generally be comparable (in certain measures) to a middle-end Samsung.
OLEDs have terrible color, its over-saturated and makes people think the image looks better in a subjective way. ((S)AM)OLED is to cellphones as glossy screens are to laptops. Do a Pepsi challenge with a iPhone or even an original Droid, versus a Galaxy Nexus in a dark room, versus watching the image on a proper screen.
Pentile doesn't help either with the blacks or contrasty images.
And while people in Slashdot screams that iPhones are cheap (which are not), there's still people in many places that cannot afford or will not pay for them.
There are iPhones that are free on contract at this point. They have a ladder in hundred dollar increments now from $0 up to $399.
Still doesn't explain why they needed two tubes of KY jelly in the dressing room. :)
This is biased logic, too. I can bias it the other way: if you're an application developer and you want a platform feature that is not yet released, and you need to wait for 6 months instead of one year, then you will get half your market six months sooner.
You shouldn't have to wait at all. 100% of your available market should have the new OS, or at least no encumbrance to acquiring it, and the platform vendor should aggressively push for adoption. There should optimally only be one target platform, and you shouldn't have to resort to functionality hacks that might have to get completely rewritten when the new version comes out. Apps are impulse buys, and if someone has to take more than even the most rudimentary steps to use your app, you'll lose a sale.
If you're talking about some sort of relative gain between platforms, none of your users is going to switch which phone he uses because of your app. If you have an iPhone app and you're afraid that you're losing sales to your competitor on Android, I assure you this isn't the case. No phone user is going to switch platforms to get your competitor's app, apps just aren't that important on an individual basis, particularly if there's only a 6 month differential.
I can tell you the worst limitation is the hardware, not the software.
I can see how this would be a good argument in the abstract, but we're talking about two phone platforms, one which has three handset models and one tablet model currently selling, and another which has a dozen models currently selling and a dozen more sold in the last three years, and more than a few tablets and e-readers of widely varying equipment.
But here's the thing that Apple fans are leaving out -- Apple is PAINFUL when it comes to their Mac OS X and what hardware it will (easily) install to. I recall back when G4 and G5 were new and which versions of Mac OS X would work on what hardware.
You're comparing installing a modern desktop OS on, in what that case was, a three year old computer with installing a modern cellphone OS on an 11-month old handset.
Android should be licensed in such a way that disallows proprietary drivers.
You realize Android is published by hardware manufacturers and cell carriers, right?
If you're an application developer and you're using a platform feature in the latest OS, if half of your potential customers are stuck a version behind it doesn't matter how many releases happen a year, you just lost half your market, for at least several months. "Release early, release often" is good for applications but for platforms it causes fragmentation and disruptions and all sorts of bad disincentives for developers.
For platform users, yes it's a different proposition.
My wife has never upgraded her HTC Aria to the current OS, while I have. Why hasn't she??? THERE WAS NO NEED TO.
On the individual basis there might not be a need to upgrade for a lot of people, but it's terrible if you're looking at Android from the standpoint of the developer. Want Fragments UI? Want low-latency audio? Want to integrate NFC beaming? Want to integrate with the calendar or visual vociemail? Or anything else?
If your app wants to merely use any of these, you'll have to maintain separate versions, builds, and perhaps even codebases. If your app would require any of these to do its magic, you're going to be locked out from wide swathes of the market at any one time.
The Dread Fragmentation rears its head.
Nor can your surgeon.
Jobs solved the innovation dilemma by having a lot of engineers circled around him.
The fact that no other corporate CEO surrounds himself with engineers, designers and implementors should be instructive. I mean you make it sound easy but why on earth are most of the other tech companies led by the nose by their sales and marketing people?
You know, as a devout capitalist, I've always believed that customers service comes first.
A capitalist of the Adam Smith variety would say that profit comes first, and that good customer service and mutual benefit is a consequence of pursuing profit.
The fact that this doesn't work under a lot of different contexts, particularly the ones that Harvard MBAs get themselves learned in, is the guts of the story.
The other problem is the presence of search data when clicking through to unencrypted sites, if they are google customers. That means google's SSL service is a lie and your unencrypted searches will be sent to certain customers regardless of using http or https.
It seems sorta common sense that if you click on a link to a site, that site will know you clicked on it and where you're going. Similarly, if you have a cookie on a site, that site will know when you've been there and will be able to correlate all kinds of things you typed into that site with links, etc.
Google possesses this information, they can sell it. That your request travelled over HTTPS means it's secret between you and google, what either side of the transmission does with the information it obtained is strictly the business of either party.
Anything you can do with a search result from https://google.com/, like for instance, sharing a search result with a friend, google can do with your click stream, like, for instance, sharing it with their friend.
The CDC data [3] shows some serious concerns, 20,096 reaction reports including 71 verified deaths over 40 million doses. Of the 20,096 reports, 1,607 (8%) are considered serious.
That's a 0.05% rate of reported reactions per dose, 0.004% of which are considered serious, compared to your 0.16% total mortality in the population. Such a risk, taken by itself, is comparable to the risk of injury or death taking the stairs to the second floor of your house, versus taking a safer elevator.
You also need to consider the morbidity numbers for cervical cancer, surgical and post-operative complications, and deaths from other cancers secondary to cervical cancer. You have also left out other cancers caused by HPV, including oral, throat, and anal cancers in men and women -- a man giving oral sex to one woman (just one, in his entire life) doubles his chances of contracting HPV, and a man with more than five partners is 8.6x more likely.
I'm too old, like most of the people here, including you, to get an HPV vaccine, so I'm really not trying to justify any decision on my behalf, but I can run numbers like anyone else.
So what, exactly, are Bill's credentials to talk about education?
He's am entrepreneur who's made more money than the median pizza chain CEO. In the United States, that's what passes for civic leader and elite opinionmaker.
That's great, but how do we do it?
Don't ask him, he's more of an idea rat.
16GB is an awful lot of RAM, I'm really curious as to what it is that they're doing that's going to require more RAM than most of these devices have in total storage space.
These are the hardware requirements to compile the complete AOSP Android system and platform, and not the requirements to merely develop an application on it.