Name one component in the PS3 where there are manufacturers not willing to release OSS drivers (i.e. where Sony would be restricted to picking parts that have open drivers).
WLAN, Bluetooth, USB, audio? Being OSS-friendly, especially in the embedded field, is an exception rather than the rule.
Sony never provided access to the GPU other than through Hypervisor calls and a dumb frame-buffer so that doesn't count.
GPU doesn't count for the OSS problem, but it still counts for the cost problem, because you still have to write the drivers for the privileged side of the hypervisor.
The real kick-starter for the hackers was the desire to run OtherOS on the slim PS3 (which didnt support it even though there is no technical reason it couldn't, just Sony wanting to stop people buying PS3s for Linux only and not buying games)
Playing the devil's advocate - although there are no technical reasons for the slim PS3 not to run Linux, there's plenty of commercial reasons: writing and maintaining GPL drivers costs time and money. And the slim ps3 was all about cutting costs. Also, the need to support OSS drivers means that further cost-cutting in future releases of the console by replacing hardware components could be hindered by the fact that Sony had to restrict their choices to those components whose manufacturers allowed OSS drivers to be written - Microsoft and Nintendo hadn't this problem.
The other manufacturers already sold cheaper consoles and were about to introduce even cheaper new variants. Sony has probably considered that the very small slice of their market made up of people who would buy the console for the OtherOS support wasn't worth the much bigger share of the market made up of people who wouldn't buy the console because it was too expensive.
Does IE really allow untrusted bytes coming from the internet to be interpreted by an open-ended infrastructure such as the DirectShow filter system? Isn't that a security nightmare?
I am also surprised that employees of Nokia have the power to remove apps from Apple's app store, I wonder why they don't remove them all. It would be a good thing for their employer.
Google "iPhone" and "jailbreak" sometime. Interesting stuff.
Jailbreaking is not "a way given you by Apple to install software on the hardware you own without abiding to their terms".
The key words are "by Apple".
Stuff that hackers have been used to do for ages, before the entitlement culture made people lazy bitches that demanded the manufacturer do the "hacking" for them.
Lucky for those bitches, other manufacturers still provide them with hardware that requires no such "hacking".
What "profits"? It's a free app! Submitted by some developer! Who is not Apple! Get a grip!
The profits coming from people buying the iPhone because it allows them to run VLC. Or do you think that Apple developed the iOS SDK for philanthropy?
Sure they are, find where in the licence it says you cannot do this
Ok, here you are. I suggest reading the last paragraph, in particular.
And there's no warranty that they'll be able to continue doing so on future hardware/software releases.
More apple hater FUD since you simply restore the phone to factory OS before servicing - and if the phone is dead they have no idea what was on it anyway.
I didn't mean this. I meant that you can't know in advance if the iPhone 5 will be hackable, or if the next iOS update will be hackable, or even if it will brick already-hacked iPhones. You can't know it while you're in a shop, deciding whether to buy an iPhone. That's because jailbreaking is not supported by Apple.
Since you introduced another argument, I can remind you that proper phones include an encrypted storage area, not accessible by the user, which can hold security information (e.g. the IMEI number, which for obvious reasons shouldn't be changeable by phone thieves) and can log (authorized or not) firmware updates. The PDS in Motorola phones used to do that. I don't know if the iPhone has one.
To me it seems like a good reason to avoid purchasing the hardware altogether, as the OP was saying.
If you have an ideological slant that is fine; but at least be honest to admit there is no valid technical backing to your position.
I haven't used words of ideology. I didn't talk about "fud" or "haters". My technical position is the following:
The iPhone does not support the installation of user-supplied applications without resorting to hacker-devised modifications which are officially unsupported, violate the iOS license, void the warranty, and aren't guaranteed to work in future firmware releases.
In comparison, Android (Symbian,...) phones support running user-provided applications simply by accepting their installation using the phone's standard user interface.
You might not care about the difference between these two situations and it's fine. You're in good company. But it has nothing to do with my honesty, or the OP decision.
Dist-upgrading Ubuntu (unstable) just arbitrarily changed my plain Italian keyboard layout to a dual setup alternating between US and Afghan.
I was hoping in the apple realm that because you pay more maybe you get what you pay for
And you do. I wrote that post just because I have the impression that while Apple's advantages are well publicized, the dark sides aren't.
It just seems to me that the amount of f'ing around with desktops to do basic things should have gone down over the past 5 years on linux, and it seems to have gone up instead.
I share your impression. I think this happens as the average Linux userspace is changing from a set of simple, legacy, documented, user-developed programs, to a set of complex, experimental, undocumented, distro-oriented self-sentient subsystems that often seem designed to serve every use case except the ones you need. And that get deprecated before getting completed.
Leave VLC aside and look at the question from a more abstract point of view.
Allows installation of user-provided applications:
iPhone - NO
Android - YES
That said, given the source of the information, I'm really confident that VLC for Android will be relased soon. Then we'll be able to see with hindsight who got more and who got less.
IMHO it also gets +1 because, should VLC violate the rules of the Android market, people would still be free to download the.apk from the VLC home page, as they would do with any other OS, and install it on their phones, by their own choice and at their own risk.
I think he's referring to the fact that he has to cope with the store's terms because his hardware won't allow software from other sources to be installed on it.
I'm a dedicated linux user for the past 5 years and I'm thinking of dumping it all and going to macs.
Instead I'm a dedicated mac user and I'm thinking of dumping it all and going to linux.;)
I spend way too much time fiddling and screwing around trying to get normal things to work. The other day for example my wife wanted a sound to come on when she got an email in thunderbird. Should be easy, no? Doesn't work on ubuntu without some serious googling/screwing around. Google earth which for some reason vanished from the medibuntu repository... same deal. For some reason the installer set the symlink to point somewhere else. And still the fonts are screwed up, don't know why.
I spend way too much time fiddling and screwing around trying to get normal things to work. And Apple keeps deleting my posts when I whine on their support forums.
The other day for example I wanted to connect my Android phone to my macbook. Should be easy, no? Doesn't work on a mac at all because "RNDIS is a Microsoft protocol".
Or one of my favorites? Kdenlive, a great video editor, can't export to h.264 out of the box on ubuntu because it uses lame so you have to put your own custom export in.
After I switched to linux, kdenlive exported to h.264 out of the box simply by choosing "H.264" from the format list (which included HDV, DV, MPEG2, MPEG4, Xvid, Flash, RealVideo, Theora and Webm).
Of course, that only worked after I enabled the restricted codecs, which aren't "restricted" by any technical reason, but only by the illiberal laws of some countries which sacrifice civil liberties to create monopolies for the profit of big enterprises. Thankfully I don't live in one of them.
Or a recent clean install of Kubuntu 10.10 that left the master mixing channel muted (not through kmix but through alsamixer).
Effectively I have some problems with that %$&%# PulseAudio which keeps eating all of my CPU just to play an MP3:D...
Or the fact that the newest ubuntu amarok packages kill it's ability to talk to my wife's ipod.
...but I'm happy anyway because I switched to a non-Apple mp3 player which doesn't require me to either use a closed, buggy, heavy, alien, limited application or to rely on amateur reverse-engineered libraries just to transfer music on it. It also costed much less and does have a removable battery.
Look, I loves me linux, but I have 3 kids, a wife, a job, and a life.
Look, I love Apple devices, but I have an underpaid IT job, so I can't afford to spend 3x the money to buy underpowered hardware.
Good for you. And good thing that the terms of license of your operating system allow you to install another media player without breaking the contract between you and its vendor.
Do you realize the insanity of what you're saying -- in THIS context? You're whining that Apple has terms of service for its store, when the software in question was pulled by VLC's developers because publishing it didn't comply with THEIR terms (the GPL).
No, he's legitimately protesting because Apple give you no way to install software on the hardware you own, without abiding to their terms.
So the people who caused this action -- that you're supporting -- are enforcing THEIR rules. Get it?
In fact, they want to stop Apple from gaining profits stealing their work by copying it without respecting its license terms. What's wrong with that?
The GPL lovers are hypocrites. They don't want to give people freedom. They simply want everyone else to be forced to make the same choices that THEY do.
Bullshit. The GPL lovers have no control over what people DO with software. They do control how people can copy the software THEY have developed. Can I, in comparison, download iLife for free? Can I reverse-engineer iOS? Can I run OS X in a virtual machine?
Did Apple pay a cent to get the base for the HTML rendering engine which made the first iPhone so successful? No, because it was LGPL. Did they share the improvements they did to it with the community, and in particular with other companies, which led to an explosion of web-capable devices all running the same engine? Yes, and *only* because it was LGPL.
I think there wasn't a big performance gap for 16-bit code between a 386 and a 286 at the same frequency, so our experiences must have been comparable.
Now that you talk about it, I do remember some slowness; I remember that by the slowdown, you could get a clue of how many enemy units you had to expect behind the fog-of-war.
Also, perhaps because I only had 1 MB of RAM, I couldn't hear all the voice samples; for instance, I heard "Harkonnen unit" instead of "Harkonnen unit, approaching, from the east".
But overall I can't say that the 286 didn't let me enjoy the game.
This starts to remind me of the 286's way of addressing memory (where you fought with conventional memory, upper memory blocks, high memory area, expanded memory, extended memory, real mode, protected mode, A20 line and segmentation on top of that), and how happy I was when I got a 32-bit PC capable of addressing all of its memory (and more) with a wide, beautiful, consistent, scalar, plain 32 bit pointer.
Although segmentation offered additional security benefits over flat memory, just like NAT does in the IP world, just nobody has ever wanted to have anything to do with it after that, with the 386, it stopped being a necessity.
Not necessarily. PDF does not preserve text flow. It breaks up paragraphs into lines (or less if kerning has been altered), and places them accurately on the page.
This is not true. PDF is capable of preserving text flow if the document contains such information. See this as an example: if you open it in acrobat reader and move the text cursor using the down arrow, you'll see it travel correctly among columns and paragraphs.
No page description format will help if the page has been generated in a broken way: for instance, try extracting text from the tables of an html page generated by javascript.
If you have a multi-column layout, then a pdf-to-text algorithm (first step in screen reading) is likely to put column-2-line-1 between column-1-lines-{1 and 2}. Best of luck sorting that out.
In this case it is the pdf-to-text algorithm to be broken, and should be fixed.
I've done some benchmarks comparing Java with equivalently coded native code. For the kinds of code I need, Java code runs almost as fast as native code: slower in number crunching, faster in high level code (e.g. string handling, memory allocation).
Java UIs appear slow not so much because of their speed, but because of their responsiveness, which may be hindered by the pauses induced by garbage collection. Work is being done to address that. They also require more memory than native UIs.
Swing's API is not bad. It allows you to create a working mirrored, 35-rotated radio button with three lines of code. It's true that built-in layout managers suck, but BoxLayout is almost usable. And if you don't like them, then don't use them at all, and fall back to absolute positioning, which is what most other toolkits do anyway. Moreover, the original poster didn't require a widget toolkit, but rather a 2D graphics API.
For example, why the hell does java.awt.Graphics.drawRectangle() not accept a java.awt.Rectangle as a parameter?
Graphics.drawRectangle() does not exist. Are you talking about Graphics.drawRect()? It's a legacy API from Java 1.0 which was intended to run on machines where the overhead of converting coordinates into a heap-allocated Rectangle structure could be significant.
Instead you should use Graphics2D.draw(), which consistently supports any Shape, including a Rectangle. It also supports floating point coodinates, transforms, etc. Please update your trolling to 2005, thank you.
I do still agree that Java is one of the better choices for cross-platform GUI building, by virtue of not being overcomplicated and error-prone like C++ or unbearably slow like Python; but if this is the best we can do, I weep for the future of the human race.
Instead, in my opinion Java2D is the most elegant, intuitive and powerful 2D "canvas" API I've ever used. For example, it allows you, with a few lines of code, to load an OpenType font, format a string using it, and convert it to a Shape which can be rendered using the current stroke and fill properties. Any step of the process can be customized using the same API: you can break the string in lines using a layout, you can iterate the vertices of the resulting Shape, you can apply transforms and clipping, the usual Porter-Duff composition, and you can control the rasterization from Shapes into pixels. All of this implemented mostly in pure Java, with fully commented source code available, and it's available out of the box in all Java implementations.
No, Android is not a Linux distribution. It's an embedded platform supporting Google's business model; it happens to run on the Linux kernel just because it was the best strategic choice. Android might drop the Linux kernel tomorrow, and Android applications would hardly notice it.
What's stopping Google from using code in the MeeGo base?
I'm 99% sure that it's nothing, and that if the code's good enough, we'll probably see a lot of cross contamination between Android and MeeGo kernel code.
Android already uses Nokia code. For instance, Bluez, and Nokia-contributed kernel code.
Palm, too, used some results of a collaboration between Nokia and TI on their products (at least this is what was said at the first MeeGo conference).
The Linux kernel already makes use of Google-contributed code.
If Nokia and Google were smart, Nokia would steal Google's UI code and Dalvik and Google would steal their best threading, i/o, and whatever other code is probably superior in MeeGo that isn't probably going to wind up in the base kernel trunk and call the whole thing a "cross platform open source collaboration" or some other nonsense string of buzzwords that make marketers beyond happy.
I think that, from a collaboration point of view, MeeGo are doing their share, by developing their software on the open, using public repositories and mailing lists, reusing existing open source projects instead of inventing new ones, and especially with the "upstream first" policy. It is yet to be seen if such approach will be successful. Or even feasible: for example, Maemo's changes for GTK were rejected by upstream IIRC.
Android chose a more business-oriented approach, by forking the software they needed, to keep it under their control, developing it behind closed doors and "revealing" it only after handsets have already been manufactured. This approach has been very successful so far, but I doubt the Linux Foundation would approve it.
But even if the two worlds were compatible, I wouldn't be happy if they merged. Competition is good even in the open source territory: in the end, after all choices have been made, a single project, open source or not, can only evolve in a single direction, and that direction need not be the best one. Or, it might make some users happy and others angry. It's good to have more projects, with differing philosophies and targets, see the different results they obtain, and be able to choose among them.
Slashdot Mirror
Name one component in the PS3 where there are manufacturers not willing to release OSS drivers (i.e. where Sony would be restricted to picking parts that have open drivers).
WLAN, Bluetooth, USB, audio? Being OSS-friendly, especially in the embedded field, is an exception rather than the rule.
Sony never provided access to the GPU other than through Hypervisor calls and a dumb frame-buffer so that doesn't count.
GPU doesn't count for the OSS problem, but it still counts for the cost problem, because you still have to write the drivers for the privileged side of the hypervisor.
The real kick-starter for the hackers was the desire to run OtherOS on the slim PS3 (which didnt support it even though there is no technical reason it couldn't, just Sony wanting to stop people buying PS3s for Linux only and not buying games)
Playing the devil's advocate - although there are no technical reasons for the slim PS3 not to run Linux, there's plenty of commercial reasons: writing and maintaining GPL drivers costs time and money. And the slim ps3 was all about cutting costs. Also, the need to support OSS drivers means that further cost-cutting in future releases of the console by replacing hardware components could be hindered by the fact that Sony had to restrict their choices to those components whose manufacturers allowed OSS drivers to be written - Microsoft and Nintendo hadn't this problem.
The other manufacturers already sold cheaper consoles and were about to introduce even cheaper new variants. Sony has probably considered that the very small slice of their market made up of people who would buy the console for the OtherOS support wasn't worth the much bigger share of the market made up of people who wouldn't buy the console because it was too expensive.
Does IE really allow untrusted bytes coming from the internet to be interpreted by an open-ended infrastructure such as the DirectShow filter system? Isn't that a security nightmare?
Yes, ISO and IETF have a different definition of what is an "open standard".
I am also surprised that employees of Nokia have the power to remove apps from Apple's app store, I wonder why they don't remove them all. It would be a good thing for their employer.
Do I, as a user, care about things that developers can do but I can't do? - NO
Can I trust the VLC development team when they say that VLC for Android is nearly complete and will be released in early 2011 - I'D SAY YES
Jailbreaking is not "a way given you by Apple to install software on the hardware you own without abiding to their terms".
The key words are "by Apple".
Lucky for those bitches, other manufacturers still provide them with hardware that requires no such "hacking".
The profits coming from people buying the iPhone because it allows them to run VLC. Or do you think that Apple developed the iOS SDK for philanthropy?
Ok, here you are. I suggest reading the last paragraph, in particular.
I didn't mean this. I meant that you can't know in advance if the iPhone 5 will be hackable, or if the next iOS update will be hackable, or even if it will brick already-hacked iPhones. You can't know it while you're in a shop, deciding whether to buy an iPhone. That's because jailbreaking is not supported by Apple.
Since you introduced another argument, I can remind you that proper phones include an encrypted storage area, not accessible by the user, which can hold security information (e.g. the IMEI number, which for obvious reasons shouldn't be changeable by phone thieves) and can log (authorized or not) firmware updates. The PDS in Motorola phones used to do that. I don't know if the iPhone has one.
I haven't used words of ideology. I didn't talk about "fud" or "haters". My technical position is the following:
The iPhone does not support the installation of user-supplied applications without resorting to hacker-devised modifications which are officially unsupported, violate the iOS license, void the warranty, and aren't guaranteed to work in future firmware releases.
In comparison, Android (Symbian, ...) phones support running user-provided applications simply by accepting their installation using the phone's standard user interface.
You might not care about the difference between these two situations and it's fine. You're in good company. But it has nothing to do with my honesty, or the OP decision.
Dist-upgrading Ubuntu (unstable) just arbitrarily changed my plain Italian keyboard layout to a dual setup alternating between US and Afghan.
And you do. I wrote that post just because I have the impression that while Apple's advantages are well publicized, the dark sides aren't.
I share your impression. I think this happens as the average Linux userspace is changing from a set of simple, legacy, documented, user-developed programs, to a set of complex, experimental, undocumented, distro-oriented self-sentient subsystems that often seem designed to serve every use case except the ones you need. And that get deprecated before getting completed.
Allows installation of user-provided applications:
iPhone - NO
Android - YES
That said, given the source of the information, I'm really confident that VLC for Android will be relased soon. Then we'll be able to see with hindsight who got more and who got less.
And there's no warranty that they'll be able to continue doing so on future hardware/software releases.
To me it seems like a good reason to avoid purchasing the hardware altogether, as the OP was saying.
IMHO it also gets +1 because, should VLC violate the rules of the Android market, people would still be free to download the .apk from the VLC home page, as they would do with any other OS, and install it on their phones, by their own choice and at their own risk.
I think he's referring to the fact that he has to cope with the store's terms because his hardware won't allow software from other sources to be installed on it.
I'm a dedicated linux user for the past 5 years and I'm thinking of dumping it all and going to macs.
Instead I'm a dedicated mac user and I'm thinking of dumping it all and going to linux. ;)
I spend way too much time fiddling and screwing around trying to get normal things to work. The other day for example my wife wanted a sound to come on when she got an email in thunderbird. Should be easy, no? Doesn't work on ubuntu without some serious googling/screwing around. Google earth which for some reason vanished from the medibuntu repository... same deal. For some reason the installer set the symlink to point somewhere else. And still the fonts are screwed up, don't know why.
I spend way too much time fiddling and screwing around trying to get normal things to work. And Apple keeps deleting my posts when I whine on their support forums. The other day for example I wanted to connect my Android phone to my macbook. Should be easy, no? Doesn't work on a mac at all because "RNDIS is a Microsoft protocol".
Or one of my favorites? Kdenlive, a great video editor, can't export to h.264 out of the box on ubuntu because it uses lame so you have to put your own custom export in.
After I switched to linux, kdenlive exported to h.264 out of the box simply by choosing "H.264" from the format list (which included HDV, DV, MPEG2, MPEG4, Xvid, Flash, RealVideo, Theora and Webm). Of course, that only worked after I enabled the restricted codecs, which aren't "restricted" by any technical reason, but only by the illiberal laws of some countries which sacrifice civil liberties to create monopolies for the profit of big enterprises. Thankfully I don't live in one of them.
Or a recent clean install of Kubuntu 10.10 that left the master mixing channel muted (not through kmix but through alsamixer).
Effectively I have some problems with that %$&%# PulseAudio which keeps eating all of my CPU just to play an MP3 :D ...
Or the fact that the newest ubuntu amarok packages kill it's ability to talk to my wife's ipod.
...but I'm happy anyway because I switched to a non-Apple mp3 player which doesn't require me to either use a closed, buggy, heavy, alien, limited application or to rely on amateur reverse-engineered libraries just to transfer music on it. It also costed much less and does have a removable battery.
Look, I loves me linux, but I have 3 kids, a wife, a job, and a life.
Look, I love Apple devices, but I have an underpaid IT job, so I can't afford to spend 3x the money to buy underpowered hardware.
Good for you. And good thing that the terms of license of your operating system allow you to install another media player without breaking the contract between you and its vendor.
Do you realize the insanity of what you're saying -- in THIS context? You're whining that Apple has terms of service for its store, when the software in question was pulled by VLC's developers because publishing it didn't comply with THEIR terms (the GPL).
No, he's legitimately protesting because Apple give you no way to install software on the hardware you own, without abiding to their terms.
So the people who caused this action -- that you're supporting -- are enforcing THEIR rules. Get it?
In fact, they want to stop Apple from gaining profits stealing their work by copying it without respecting its license terms. What's wrong with that?
The GPL lovers are hypocrites. They don't want to give people freedom. They simply want everyone else to be forced to make the same choices that THEY do.
Bullshit. The GPL lovers have no control over what people DO with software. They do control how people can copy the software THEY have developed. Can I, in comparison, download iLife for free? Can I reverse-engineer iOS? Can I run OS X in a virtual machine?
Did Apple pay a cent to get the base for the HTML rendering engine which made the first iPhone so successful? No, because it was LGPL. Did they share the improvements they did to it with the community, and in particular with other companies, which led to an explosion of web-capable devices all running the same engine? Yes, and *only* because it was LGPL.
Now that you talk about it, I do remember some slowness; I remember that by the slowdown, you could get a clue of how many enemy units you had to expect behind the fog-of-war. Also, perhaps because I only had 1 MB of RAM, I couldn't hear all the voice samples; for instance, I heard "Harkonnen unit" instead of "Harkonnen unit, approaching, from the east".
But overall I can't say that the 286 didn't let me enjoy the game.
I remember I had to fix my uncle's computer because he had deleted COMMAND.COM thinking it was part of Commander Keen.
Hey, you could also run Dune II on a 286. With it you could get most of the fun of Command & Conquer (which required a 386 instead).
Yes, if you want you can preserve those and much more. See for yourself, there are standard tags to describe most of the document structure.
Although segmentation offered additional security benefits over flat memory, just like NAT does in the IP world, just nobody has ever wanted to have anything to do with it after that, with the 386, it stopped being a necessity.
Not necessarily. PDF does not preserve text flow. It breaks up paragraphs into lines (or less if kerning has been altered), and places them accurately on the page.
This is not true. PDF is capable of preserving text flow if the document contains such information. See this as an example: if you open it in acrobat reader and move the text cursor using the down arrow, you'll see it travel correctly among columns and paragraphs.
No page description format will help if the page has been generated in a broken way: for instance, try extracting text from the tables of an html page generated by javascript.
If you have a multi-column layout, then a pdf-to-text algorithm (first step in screen reading) is likely to put column-2-line-1 between column-1-lines-{1 and 2}. Best of luck sorting that out.
In this case it is the pdf-to-text algorithm to be broken, and should be fixed.
Java UIs appear slow not so much because of their speed, but because of their responsiveness, which may be hindered by the pauses induced by garbage collection. Work is being done to address that. They also require more memory than native UIs.
Swing's API is not bad. It allows you to create a working mirrored, 35-rotated radio button with three lines of code. It's true that built-in layout managers suck, but BoxLayout is almost usable. And if you don't like them, then don't use them at all, and fall back to absolute positioning, which is what most other toolkits do anyway. Moreover, the original poster didn't require a widget toolkit, but rather a 2D graphics API.
Graphics.drawRectangle() does not exist. Are you talking about Graphics.drawRect()? It's a legacy API from Java 1.0 which was intended to run on machines where the overhead of converting coordinates into a heap-allocated Rectangle structure could be significant.
Instead you should use Graphics2D.draw(), which consistently supports any Shape, including a Rectangle. It also supports floating point coodinates, transforms, etc. Please update your trolling to 2005, thank you.
Instead, in my opinion Java2D is the most elegant, intuitive and powerful 2D "canvas" API I've ever used. For example, it allows you, with a few lines of code, to load an OpenType font, format a string using it, and convert it to a Shape which can be rendered using the current stroke and fill properties. Any step of the process can be customized using the same API: you can break the string in lines using a layout, you can iterate the vertices of the resulting Shape, you can apply transforms and clipping, the usual Porter-Duff composition, and you can control the rasterization from Shapes into pixels. All of this implemented mostly in pure Java, with fully commented source code available, and it's available out of the box in all Java implementations.
Tutorial.
Isn't that what Android is for?
No, Android is not a Linux distribution. It's an embedded platform supporting Google's business model; it happens to run on the Linux kernel just because it was the best strategic choice. Android might drop the Linux kernel tomorrow, and Android applications would hardly notice it.
What's stopping Google from using code in the MeeGo base?
I'm 99% sure that it's nothing, and that if the code's good enough, we'll probably see a lot of cross contamination between Android and MeeGo kernel code.
Android already uses Nokia code. For instance, Bluez, and Nokia-contributed kernel code.
Palm, too, used some results of a collaboration between Nokia and TI on their products (at least this is what was said at the first MeeGo conference).
The Linux kernel already makes use of Google-contributed code.
If Nokia and Google were smart, Nokia would steal Google's UI code and Dalvik and Google would steal their best threading, i/o, and whatever other code is probably superior in MeeGo that isn't probably going to wind up in the base kernel trunk and call the whole thing a "cross platform open source collaboration" or some other nonsense string of buzzwords that make marketers beyond happy.
I think that, from a collaboration point of view, MeeGo are doing their share, by developing their software on the open, using public repositories and mailing lists, reusing existing open source projects instead of inventing new ones, and especially with the "upstream first" policy. It is yet to be seen if such approach will be successful. Or even feasible: for example, Maemo's changes for GTK were rejected by upstream IIRC.
Android chose a more business-oriented approach, by forking the software they needed, to keep it under their control, developing it behind closed doors and "revealing" it only after handsets have already been manufactured. This approach has been very successful so far, but I doubt the Linux Foundation would approve it.
But even if the two worlds were compatible, I wouldn't be happy if they merged. Competition is good even in the open source territory: in the end, after all choices have been made, a single project, open source or not, can only evolve in a single direction, and that direction need not be the best one. Or, it might make some users happy and others angry. It's good to have more projects, with differing philosophies and targets, see the different results they obtain, and be able to choose among them.