You (and others) do realize that by saying that ranking neighborhoods by _crime_ levels is _racist_, you're implying "It's racist, because crime is for nig^W^W^Wrates are correlated to racial composition"? There's enough examples of racism here, idea in the original article isn't one.
As a side note, "Born in slums, achieved everything by his willpower and intellect" is not a bad image for a politician.
Gah, scratch the "- ergo, every Android user downloads an app, while not every Apple user does." part, same logic applies as in "doesn't necessarily mean "2.2% of Android devices never visit the Web"".
And again, you're trying to prove wrong point. I'll remind you: we were discussing "Are Android phones used just as plain old phones or as smartphones?". You, for some reason, try to prove "Apple is used much smartphoner than Android".
But let's look what *those statistics* show. Note that statistics are still not the absolute truth. First, apps. a) Android overtook iOS in mobile app downloads and now has 44 percent share worldwide vs. 31 percent for iOS b) iOS beats Android in terms of downloads per user by 2-to-1 c) Ergo, downloading users ratio is (44%/31% total app downloads)/(1/2 app downloads per user) ~= 3 Android user downloads per 1 Apple user d) Android’s install base now exceeds iOS by a factor of 2.4-to-1 worldwide - ergo, every Android user downloads an app, while not every Apple user does. Heya, you just kinda countered your own point.
Second, web usage.
OS Market Share by Audience Installed Base 3 Month Average Ending August 2011 Total Mobile Audience, U.S., Age 13+ Source: comScore MobiLens Google Android 34.1% OS Market Share by Digital Traffic (Browser-Based Page Views) August 2011 Total U.S. - Home and Work Locations Source: comScore Device Essentials Google Android 31.9%
Heya, check how those two numbers align. Note that it doesn't necessarily mean "2.2% of Android devices never visit the Web", as it would only be true if there were X devices and all the internet traffic was same X page views. But yeah, iPhone users are 133% more smartphoner by internet usage, as they drive 58% of traffic with 43% of devices in that statistic.
There's region filter in extended search, and choice seems to be prettypoor.
Re: permissions, data connections are in "Network" permission group, "multimedia" and "connectivity" represent access to phone's camera/mic and serial/IR/bluetooth. Sucks to be an "untrusted developer" on T-Mobile. Also missing in both cases is Read/Write User Data, which includes not only PIM data, but also user file access.
Duh, because we were discussing "Are Android phones are used just as plain old phones or as smartphones?", remember? Here, I'll quote you:
All the evidence shows that Android users statistically don't use their phones as "smart phones" as often as iOS users.
And then you proceed to talk about app profits, not app usage (also note, that Flurry is an analytics toolkit to be built in your apps. Second article talks about iOS devs preferring Flurry analytics, not something more general)
At least all Nokia phones are comparatively low-pain for Java ME development (which is generally fucked up experience). Consistent API, installation by simply puttin.jar somewhere in the storage and opening it on the phone.
(1) counts all iOS devices, not just phones and (2) speaks about in-app purchases stats.
I wonder, is there any data on what are estimated absolute numbers, not percentages, for all iOS devices vs all Android devices? Most stats give either "Market share of smartphones" or "Browser share", but no useful way to find out how many are there iPhones/iPods/iPads and Android phones/tablets/e-readers/players
But how "Global _mobile browser_ market-share" is useful outside of e-penis size competitions?
What would someone need it for? Ad targeting? Only matters to ad networks somewhat. Web masters? Nope. People who wish to advertise won't care for it too, they're more likely to care about phone sales or app sales.
For webmasters to optimize for specific browser quirks? Oh, wait, modern mobile browsers render quite consistently.
The only meaningful browser share statistics are those for your own site, as visitor demographics are wildly varying between sites - you might want to support IE6 or you might want to drop IE before 8 altogether, or you might want to give special attention to Chrome/Opera/Saphari/FF. Even then it doesn't mean much unless you're writing a rich web app and want to use some specific extensions.
Yes, desktop Linux is secure enough for only semi-loud malware story about it to be "someone uploaded trojan shell script masked as a Gnome addon to a third-party Gnome addons site, some people actually downloaded it and some even ran it". Can't remember did it try to get user to sudo it or just did what it could with user's permissions.
Server Linux, on the other hand, is very attractive target as it hosts a big part of the web and targeted software is not Linux per se, but usually buggy CMS's and unpatched Apache installations.
Windows, on the other hand, has a few nice MS-introduced OS level vulnerabilities discovered this year - not to forget about the beautiful times brought by LoveSan and alikes.
You know that "You wouldn't download a car" adage? Well, Jesus would and could.
He distributed illegal copies of bread and fish (see, no theft, just copying) depriving fishermen and bakers of their profits and circumvented DRM to upgrade water to wine bypassing the winery and proper grapes fermentation process.
Yes, because "Bah, say when" isn't coordination. Just make them tell you when's re-foobarization scheduled and then say "Great, 7-9AM is free, keep it so, and we'll roll in updates." and write it all down for both sides' future reference.
There are purely technical jobs in IT, but majority is still about working with people. Communication skills is a must. Consider yourself extended man-machine interface, because that's what we are, essentially.
Yes, having someone who can properly rephrase "We didn't finish re-foobarization process because IT does updates" to "This process takes N hours and costs X dollars. If it's interrupted, it has to be restarted, taking N hours and X dollars again. In this month we lost X*42 dollars due to untimely forced reboots. We need IT department to work with us on coordinating this issue" is the value of good management.
People higher up are usually quite good at understanding numbers, as in "losses of X dollars" and "N hours of lost productivity".
Who do you mean by "Call IT"? Calling low-level IT drones won't help you much, they just implement policies - if they're worth their pay, though, they should point towards people who write policies.
If it keeps happening, then you're calling wrong people. Usually, just showing the losses to someone up the chain would be enough for IT management to get instructed (hard and without lube) not to do it again and "Don't touch this system without go-ahead from Anonymous Coward" in big red letters to appear in the policies.
There's a reason most BASIC books taught bubble sort.
And the reason is that they're targeted at absolute novices in programming and algorithms, as bubble sort, insertion sort and selection sort are simple iterative algorithms and easy to comperhend?
You can always hack together some abstraction, for example storing file index and files in WebSQL database - supported across all mobile and Webkit plus Opera on desktop. This'll let you use somewhere in tens of megabytes on iOS - it asks for permission after 5Mb.
It's possible, just not on mobile. There's older FileReader API which lets you read files after user selects them through standard <input type="file"> field or drag-n-drops them onto browser window, and there's FileSystem API which lets you request a sandboxed piece of FS where you can do whatever you want.
Neither is available on iOS and only FileReader on Android.
Hate to break it to you, but it's not innovation, it's a new iteration of old idea at best. Graphical drag-n-drop IDEs count is in dozens, and they all do and will suck. Just because you didn't see any before doesn't mean it's something revolutionary.
All of them, except for those targeted at kids in elementary school, have the only redeeming quality - good old "code block" element, which - GASP - requires keyboard and typing code. Without it, they're even more of a head-ache than old 8-bit BASICs for anything outside of "Hello, world!"-ish range.
Being able to "script" things together by dragging items/objects/commands together into sequences and combining these sequences in various ways
Your "revolution" was done so many times and in so many ways, and it always stumbles into the "make it simple enough for even fool to use and only fool would use it" problem.
Graphical programming scales incredibly bad - doing simple things is OK, but once you step away from trivial 5-10 line scripts it all devolves in colorful spaghetti.
You _might_ get it to in a very specific cases, like when "c = a + b" is a part of basic block where data-flow analysis tells you that a and b are absolutely guaranteed to be plain ints on input.
Even then you still get something like "unpack a and b from tagged value representation, check for overflow, just do asm("add a,b") if no overflow, add_and_promote_to_big_int(a,b) if overflowed, add type tag, store c".
In most cases though you get extra weight like 'look-up JIT cache for this combination of parameter types, jump there if i'm lucky, update counters and interpret it or JIT compile it if not" and so on, and so on, and so on.
You might get decent output in some cases, but in general optimizing dynamically typed languages is hard or sometimes just impossible.
Slashdot Mirror
How is this is just like this except for "it's a keyboard and it's thin"?
Gah, the irony, it burns.
You (and others) do realize that by saying that ranking neighborhoods by _crime_ levels is _racist_, you're implying "It's racist, because crime is for nig^W^W^Wrates are correlated to racial composition"? There's enough examples of racism here, idea in the original article isn't one.
As a side note, "Born in slums, achieved everything by his willpower and intellect" is not a bad image for a politician.
Gah, scratch the "- ergo, every Android user downloads an app, while not every Apple user does." part, same logic applies as in "doesn't necessarily mean "2.2% of Android devices never visit the Web"".
Me stupid.
And again, you're trying to prove wrong point. I'll remind you: we were discussing "Are Android phones used just as plain old phones or as smartphones?". You, for some reason, try to prove "Apple is used much smartphoner than Android".
But let's look what *those statistics* show. Note that statistics are still not the absolute truth.
First, apps.
a) Android overtook iOS in mobile app downloads and now has 44 percent share worldwide vs. 31 percent for iOS
b) iOS beats Android in terms of downloads per user by 2-to-1
c) Ergo, downloading users ratio is (44%/31% total app downloads)/(1/2 app downloads per user) ~= 3 Android user downloads per 1 Apple user
d) Android’s install base now exceeds iOS by a factor of 2.4-to-1 worldwide - ergo, every Android user downloads an app, while not every Apple user does.
Heya, you just kinda countered your own point.
Second, web usage.
OS Market Share by Audience Installed Base 3 Month Average Ending August 2011 Total Mobile Audience, U.S., Age 13+ Source: comScore MobiLens Google Android 34.1%
OS Market Share by Digital Traffic (Browser-Based Page Views) August 2011 Total U.S. - Home and Work Locations Source: comScore Device Essentials Google Android 31.9%
Heya, check how those two numbers align. Note that it doesn't necessarily mean "2.2% of Android devices never visit the Web", as it would only be true if there were X devices and all the internet traffic was same X page views. But yeah, iPhone users are 133% more smartphoner by internet usage, as they drive 58% of traffic with 43% of devices in that statistic.
To clarify, "Network" is HTTP/HTTPS, plain TCP/UDP sockets are in "Low-level Network".
There's region filter in extended search, and choice seems to be pretty poor.
Re: permissions, data connections are in "Network" permission group, "multimedia" and "connectivity" represent access to phone's camera/mic and serial/IR/bluetooth. Sucks to be an "untrusted developer" on T-Mobile. Also missing in both cases is Read/Write User Data, which includes not only PIM data, but also user file access.
Duh, because we were discussing "Are Android phones are used just as plain old phones or as smartphones?", remember? Here, I'll quote you:
All the evidence shows that Android users statistically don't use their phones as "smart phones" as often as iOS users.
And then you proceed to talk about app profits, not app usage (also note, that Flurry is an analytics toolkit to be built in your apps. Second article talks about iOS devs preferring Flurry analytics, not something more general)
It's hard to tell what's "representative", when Nokia feature phones range from something like this to something like this. You might try browsing through some specs.
At least all Nokia phones are comparatively low-pain for Java ME development (which is generally fucked up experience). Consistent API, installation by simply puttin .jar somewhere in the storage and opening it on the phone.
Not as conclusive as you think.
(1) counts all iOS devices, not just phones and (2) speaks about in-app purchases stats.
I wonder, is there any data on what are estimated absolute numbers, not percentages, for all iOS devices vs all Android devices? Most stats give either "Market share of smartphones" or "Browser share", but no useful way to find out how many are there iPhones/iPods/iPads and Android phones/tablets/e-readers/players
s/Saphari/Safari/
How did this happen, I don't even.
But how "Global _mobile browser_ market-share" is useful outside of e-penis size competitions?
What would someone need it for? Ad targeting? Only matters to ad networks somewhat. Web masters? Nope. People who wish to advertise won't care for it too, they're more likely to care about phone sales or app sales.
For webmasters to optimize for specific browser quirks? Oh, wait, modern mobile browsers render quite consistently.
The only meaningful browser share statistics are those for your own site, as visitor demographics are wildly varying between sites - you might want to support IE6 or you might want to drop IE before 8 altogether, or you might want to give special attention to Chrome/Opera/Saphari/FF. Even then it doesn't mean much unless you're writing a rich web app and want to use some specific extensions.
"Sample reports", as in "reports used for illustration in the FAQ".
There might be reasons to believe their stats are incorrect, but "fabricated" is another thing.
Yes, desktop Linux is secure enough for only semi-loud malware story about it to be "someone uploaded trojan shell script masked as a Gnome addon to a third-party Gnome addons site, some people actually downloaded it and some even ran it". Can't remember did it try to get user to sudo it or just did what it could with user's permissions.
Server Linux, on the other hand, is very attractive target as it hosts a big part of the web and targeted software is not Linux per se, but usually buggy CMS's and unpatched Apache installations.
Windows, on the other hand, has a few nice MS-introduced OS level vulnerabilities discovered this year - not to forget about the beautiful times brought by LoveSan and alikes.
You know that "You wouldn't download a car" adage? Well, Jesus would and could.
He distributed illegal copies of bread and fish (see, no theft, just copying) depriving fishermen and bakers of their profits and circumvented DRM to upgrade water to wine bypassing the winery and proper grapes fermentation process.
Yes, because "Bah, say when" isn't coordination. Just make them tell you when's re-foobarization scheduled and then say "Great, 7-9AM is free, keep it so, and we'll roll in updates." and write it all down for both sides' future reference.
There are purely technical jobs in IT, but majority is still about working with people. Communication skills is a must. Consider yourself extended man-machine interface, because that's what we are, essentially.
Yes, having someone who can properly rephrase "We didn't finish re-foobarization process because IT does updates" to "This process takes N hours and costs X dollars. If it's interrupted, it has to be restarted, taking N hours and X dollars again. In this month we lost X*42 dollars due to untimely forced reboots. We need IT department to work with us on coordinating this issue" is the value of good management.
People higher up are usually quite good at understanding numbers, as in "losses of X dollars" and "N hours of lost productivity".
Who do you mean by "Call IT"? Calling low-level IT drones won't help you much, they just implement policies - if they're worth their pay, though, they should point towards people who write policies.
If it keeps happening, then you're calling wrong people. Usually, just showing the losses to someone up the chain would be enough for IT management to get instructed (hard and without lube) not to do it again and "Don't touch this system without go-ahead from Anonymous Coward" in big red letters to appear in the policies.
There's a reason most BASIC books taught bubble sort.
And the reason is that they're targeted at absolute novices in programming and algorithms, as bubble sort, insertion sort and selection sort are simple iterative algorithms and easy to comperhend?
You can always hack together some abstraction, for example storing file index and files in WebSQL database - supported across all mobile and Webkit plus Opera on desktop. This'll let you use somewhere in tens of megabytes on iOS - it asks for permission after 5Mb.
How many modules will you now have to load? How many APIs will you now have to learn?
Usually, you're one import('...') away from LINE, PLOT and CIRCLE likes.
Many are just packaged as "game engines". For example, Lua has LÃve - I think it's not much harder to save something like
to main.lua and run it with "love ."
It's possible, just not on mobile. There's older FileReader API which lets you read files after user selects them through standard <input type="file"> field or drag-n-drops them onto browser window, and there's FileSystem API which lets you request a sandboxed piece of FS where you can do whatever you want.
Neither is available on iOS and only FileReader on Android.
Nope. Same fallacy as with COBOL, "It's just like English, must be easy, right?"
Writing actual code is 10% of programming, constructing algorithms and structures is 90.
Hate to break it to you, but it's not innovation, it's a new iteration of old idea at best. Graphical drag-n-drop IDEs count is in dozens, and they all do and will suck. Just because you didn't see any before doesn't mean it's something revolutionary.
All of them, except for those targeted at kids in elementary school, have the only redeeming quality - good old "code block" element, which - GASP - requires keyboard and typing code. Without it, they're even more of a head-ache than old 8-bit BASICs for anything outside of "Hello, world!"-ish range.
Being able to "script" things together by dragging items/objects/commands together into sequences and combining these sequences in various ways
Your "revolution" was done so many times and in so many ways, and it always stumbles into the "make it simple enough for even fool to use and only fool would use it" problem.
Graphical programming scales incredibly bad - doing simple things is OK, but once you step away from trivial 5-10 line scripts it all devolves in colorful spaghetti.
You _might_ get it to in a very specific cases, like when "c = a + b" is a part of basic block where data-flow analysis tells you that a and b are absolutely guaranteed to be plain ints on input.
Even then you still get something like "unpack a and b from tagged value representation, check for overflow, just do asm("add a,b") if no overflow, add_and_promote_to_big_int(a,b) if overflowed, add type tag, store c".
In most cases though you get extra weight like 'look-up JIT cache for this combination of parameter types, jump there if i'm lucky, update counters and interpret it or JIT compile it if not" and so on, and so on, and so on.
You might get decent output in some cases, but in general optimizing dynamically typed languages is hard or sometimes just impossible.