Granted: the Metro UI was genuinely innovative on a phone-sized device, and I can see that there are actually people for whom it's a better UI than iOS or Android. Microsoft did something actually innovative and good for the handheld with this UI.
But it does not work on a standard keyboard-and-mouse desktop environment. It also doesn't really work in the set-top environment -- I say this both because of my experiments with Kinect-driven Metro on the latest XBox, and because of my experiments with Windows 8 and 720p displays. (Windows 8 will not show you your address book on a 720p display, because it says it's not high resolution enough. At 1280x720. To show your address book.)
Now, it's possible that Microsoft will be able to force this through. But in my workplace, where people already have a choice between Windows, Linux, and MacOS, I'm expecting a noticeable uptick in Linux and MacOS desktop users over the 18 months after the release of Windows 8.
(I'm planning to familiarize myself more with Ubuntu/Unity in the near future, so I can be helpful during this time. And yes, I'm also familiarizing myself with Windows 8, running the consumer preview both natively and under VMs, and conducting experiments and compatibility tests. And my own day-to-day desktop environment is already MacOS.)
That "or (nowadays) Android" coupled with the kernel 3.3 news gives me some optimism for the future.
I'm really looking forward to, say, Debian or Ubuntu just bundling Dalvik and a large portion of the Android stack. I'm looking forward to the day when one of my alternatives to Office on the Linux desktop is actually the Android version of "Documents to Go", running full screen "Lion"-style or "Windows 8 Metro"-style.
(Likewise, I'm looking forward to CyanogeMod bundling large portions of Debian or Ubuntu, where it makes sense to do so, so I can maybe run Eclipse or at least Emacs on my Nook Color, with the full surrounding ecosystem that those environments expect.)
(Full disclosure: from about 1994 to 2000 I used Linux heavily everywhere, but today I only use it on my servers, using MacOS on the desktop/laptop. Also, I have an Android tablet -- a hacked Nook Color with CM7.1 on it -- but most of my mobile work today is on iOS. The reason is usability, coupled with enthusiasm for Objective-C left over from my NeXTstep days. But a merged Android and Linux might bring me back "into the fold", so to speak... though I still consider myself "in the fold", because Linux is my OS of choice for all my servers as it has been since the mid 1990s.)
It's true. In the mid-1990s, I was using a Newton connected to a phone line to receive faxes, annotate them, and send them back. It was crude, but extremely valuable to me at the time.
The devices running the General Magic software, "MagicCap", were a bit better at it, since it was almost all they could do!
(They were pen-based devices like the Newton, but unlike the Newton, they did not attempt handwriting recognition. So when you wrote a mail message to someone, you sent them a drawing, not anything a computer would recognize as text. The OS's infrastructure for handling that stuff was largely centered around "annotating" the documents with "stamps" or "stickers". Like, the way you'd mark a mail message as "important" was to annotate its envelope with what visually looked like an "important" rubber stamp. The OS bound meaning to the annotation. Pretty slick, for its day.)
(I still own four working Newtons and three working MagicCap devices. And yes, I also own several PalmOS and WinCE devices, and even an ancient Poqet, a palmtop with MS-DOS in ROM!)
I'd have no problem with 300 most of the time, but I'm not a fan of multiplayer games that require twitch reflexes, nor do I do anything significant on OnLive.
Another place you might have trouble is streamed content -- this is one of the reasons I do not like streamed content. I'd much rather use the model of "download the content, use it, and then discard it", which is much less sensitive to latency, lets you get higher quality regardless of bandwidth (as long as you allow the download time to be longer than the viewing time), and lets you schedule your traffic for times when the network isn't congested.
(Alas, it seems to me that content-owners think of streaming-instead-of-downloading as a form of DRM, and so attempt to make the other options look less attractive. Makes it more like TV/cable, I guess.)
...they are already going to have to include over 4,000 people in the game's credits. Anyone who donates $100 or more gets "special thanks in the game's credits".
Just FYI: you can use off-the-shelf USB barcode scanners with the iPad. I have done it. If the barcode scanner presents itself as a USB keyboard, you just plug it into the iPad's camera connection kit, and the iPad then thinks it's got an external hardware keyboard connected. Scans work.
This isn't theory -- I tested it with my own USB barcode scanner. It does work.
You're explaining why the version of the Linux kernel commonly used for desktops today cannot provide all the features Android apps use today.
What's to stop someone from basing a desktop (or tablet) distribution on the Android kernel? Sure, it's not what the "Linux mainstream" is doing, but can't it be done?
(Maybe it can't, in which case I look forward to being educated. But I've seen fairly complete userspace tool collections available to add on to Android distributions, to the point where people talk about "Ubuntu for Android"...)
1) We have DirecTV (in the United States), and some of our content is 1080p (It's mostly "on-demand" movies that download to the DVR's hard drive before playing, and take way longer for us to download than actually play back, but so what? I can let it download all day while I'm at work and then watch it in full quality when I get home in the evening. Fuck streaming, it's not the only option.).
2) We have multiple devices that send a signal to the TV that's locally generated, not a movie or TV show or something, and they're visibly better when connected at 1080p.
One example is our XBox 360. The user interface elements are sharper, for example.
Another example is an actual computer. They're coming with HDMI ports these days. No need for VGA or DVI or DisplayPort or whatever, just plug in the HDMI cable and get video and audio and everything. For this, 1080p is definitely better. Just try both and see.
(Heck, we're even thinking of getting an HDMI/USB KVM, since we've got multiple devices that use HDMI for both video and audio and use USB for peripherals. Then I'd be able to share a keyboard, mouse, and wired XBox game controllers across multiple systems.)
On the XBox 360, I can play four (well, five and a half, but four that count) different kinds of game: on-disc games, XBLA download games, "on-demand" download games, and "indie" games.
All four can be patched. Is the claim that even on the cheapest of these, a patch costs $40,000? That's a bit hard for me to swallow, because I've seen "indy" games get patched, and those are often from very small shops.
(There's also "original XBox" games, now obsolete, but also capable of being patched. There's also interactive content on video DVDs, but almost nobody distributes games that way, and these cannot be patched.)
I'm already using it this way, using it as an arbitrary file store, from my desktop via the browser and on my iOS devices via multiple applications like "GoodReader".
The article didn't give me an idea of what they're adding. Transparent background sync with local filesystem? Standards-compliant WebDAV access?
Loss-leaders are certainly of benefit to those that decide to engage in them. I think that's obvious. You seemed to think I didn't see that. What about my commend gave you that conclusion?
That does not mean that they're of net benefit to consumers, or to society as a whole.
The arguments you make are similar in character to the arguments against regulating pollution. Pollution regulations would not be put in place for the benefit of polluters, but for everyone else.
My preference is for the market to be made up of fair competition among a huge number of small players. I tend to have a negative view of practices that give incumbents a leg up. Is that enough on its own to explain my stance to you, or do you need for more of the dots to be connected?
Why do you think there's a high risk of hurting their market share? Do you think most of the masses buying Android phones are buying them because they run Android? Do you think they care about the Android brand, or about Google services (like marketplace) specifically?
If so: you may be correct, but it's certainly not self-evidently obvious. Amazon's selling Kindle Fire systems like mad. Why couldn't Samsung do the same with a fork for smartphones?
Most hate towards PHP comes from elitist snobs who don't know how to use the language.
Maybe. But not all of it.
I have been using PHP, for both business and personal stuff, since it was called "PHP/FI" ("personal home page / forms interface"). My old startup company would never have been successful without stuff we did in PHP. I've written and contributed C-based PHP extension code which was distributed with the core for years (until the underlying libraries fell so far out of use that there was no point including it anymore).
I still use PHP sometimes. But I prefer not to use it for important stuff that needs to stick around for a long time if it has any complexity.
Here's why: in my own personal experience (YMMV, of course), PHP has not been terrific about separating security fixes and bug fixes from feature evolution. It was far too easy to write code that would break with a version upgrade (eg. because an API or behavior changed), and to be forced into that version upgrade because security hole fixes were only available for the newest versions.
It's certainly the case that if you were very careful, you could avoid the problems. But, you had to be careful, and I saw a lot of novice programmers who weren't. (Our startup company provided PHP APIs for our product -- as well as Perl, Python, Java, and even TCL, this being the mid-to-late 1990s -- so I got to interact with a lot of novice PHP programmers.)
It's also certainly the case that for home/hobbyist use, you can mitigate some of the trouble by getting your PHP as part of a Debian distribution instead of getting it directly from upstream sources. The Debian folks are kinda insane about backporting security fixes to regarded-as-obsolete versions of software, and I love 'em for it. This is what I do on my household servers. (Yes, I still run my servers myself, not via a hosting provider -- have since about 1988, when they were Sun and VAX systems instead of the Linux I use today.)
It's certainly possible that PHP has gotten significantly better about this than it used to be. I haven't had reason to go check.
But anyhow, that's where my own... hatred is too strong a word, let's say maybe "concern"?... for PHP came from.
I don't know what you mean by "more catholic than the Pope" here.
Google has never made it clear that they do not mind unauthorized Market access. If they authorized it, I'd install it. But, were you around when the UNISYS patent came out of nowhere and nailed all the GIF users? I'm not going to start using something that I know has intellectual property landmines in it, trusting that the IP-owner is going to be reasonable. Been burned that way before.
They can make it clear that the use is authorized, or I can ignore it and use something else. Not going to risk it.
Tablets are faster, cheaper, have far more features, and are no more difficult to use.
Can you send me a pointer to a full tablet that's faster, cheaper, has more features, and is no more difficult to use than a Nook Color, please? I've never seen or heard of one, myself.
A few minutes on slatedroid, half an hour of effort and your favorite Chinese tablet is running the full market.
Alas, not (unarguably) legally.
My only Android device is a Nook Color. Yes, I've got CM7 installed on a microSD card. But have you noticed that CM doesn't come with marketplace? Know why?
Yes, if you install CM on a device that originally came with marketplace, it's pretty well agreed upon that it's legal to back up that copy of marketplace, install CM, and then restore the marketplace. But my Nook Color never came with marketplace.
Some folks argue that if you have another Android device that has the Google apps (including marketplace), it's kosher to back them up from that device and restore them to a device that didn't have it, as long as they're both your devices. I actually do buy that argument. But, I have no other Android devices, so that argument does not apply to me.
There is no legal way for me to install the Google marketplace app on my own Nook Color, rooted or not, CM7 or not. If I'm unwilling to entertain the possibility of piracy, what are my options?
Fortunately, Amazon's Android marketplace is freely downloadable from Amazon's web servers. That lets me get a bunch of software without doing anything that anyone would interpret as piracy, so, when I'm running CyanogenMod, that's where all my apps come from. I simply have no other realistic option.
I do wish Google would say "look, it's okay to install this stuff, here's a spot where you can download it from, go nuts, as long as it's just for your own personal use and you're not repackaging it". I'd do so, and Google (and developers in the marketplace) would start getting some of my money. But as things stand, well, if Google doesn't want my money, okay, that's their call.
1) There are people who want a backlit color LCD display even for simple reading. While many people have eyes that are better off staring at e-ink, there are also many people for whom e-ink has problems, for whom looking at a backlit LCD is actually better. (Both I and Charlie Stross happen to be in that set. I recommend everyone try both -- forget propaganda/marketing, you need to figure out what's better for your eyes, under your reading conditions. Trust no one, test it yourself.)
2) There are people, and a lot of them, who want more than an e-reader, but less than a tablet. They want to consume not just B&W text, but color text, and in particular animations.
To elaborate on #2: a slate that can run e-reader software plus colorful animated "Dr. Seuss" books plus Netflix and Hulu is not (necessarily) a full tablet. A full-fledged tablet can do more than just consume content, and some people do just want to consume content (but more content than just books).
This does not impact the Nook Color in any significant way.
Both the Nook Color and Nook Tablet will try to boot off microSD first if they can. That's not part of the OS. However, the Nook Tablet requires a signed kernel to boot, and the Nook Color does not. So, this change results in a significant loss of hackability for the Nook Tablet, since you had to "jailbreak" it in some sense to do anything. It does not result in a significant loss of hackability for the older Nook Color, since you can still just write an unsigned kernel to a microSD card and you're off and running.
Disclaimer: this is my understanding from scouring the xda-dev forums for details and from hacking my own Nook Color. I've confirmed that 1.4.1 on the Nook Color does close the sideloading "hole", and that a 1.4.1 Nook Color will still boot stuff like CM7.1 from microSD card. The rest of it, I have not personally verified myself, but am summarizing my understanding from reading experts talking about it all.
..that it's in our best interests, as a society, to have legions of unemployed scientists out there with a grudge against those fools at the academy...
I find it impossible to answer without knowing more about both your customers and the market you're operating in.
My own startup open sourced as much of our code as we legally could (though the term hadn't quite come into wide usage yet). We wrote software that implemented communication protocols (financial transaction protocols, specifically) that were only available under NDA and only "legal" to use after the software had undergone rigorous certification by financial institutions. We put that into the smallest C library we could, and shipped that library in both static and dynamic linked forms (so folks could replace it without rebuilding everything), and gave away the source code to everything on top of that, including the shims that made our C library available in Perl, Python, PHP, TCL, and Java. (In fact, this is how my code and name got into the core PHP distribution for years.)
We had a few "apps", like a GUI cash register and a batch script for running transactions stored in a CSV file, and we gave away the full source for those and encouraged our users to modify them and share the changes.
However, this only worked for us because of the market we were in (internet commerce) and who our customers were (people building out internet commerce in the mid 1990s, who were hackers who wanted to get real business done).
If our customers had been people who just wanted the apps, and never needed to tweak them, this could have been counterproductive from a business standpoint, and I might not own a seven bedroom house today. But we knew who our customers were and thankfully we techies were able to put this in place before we brought business types in to help us run things. (I suspect that if the business types had been there at the start, this strategy would have been vetoed.)
Help me figure this out: is there something about the <video> tag that makes DRM not work? Because if there is, I cannot see it.
You put in the video tag and you list sources. Let's for example assume I've got a FnordPlay protected MPEG-4 video file that I'm entitled to view. Everything I can see leads me to believe that if I put that file on my own web server, and put an HTML5 page with a video tag there next to it, I'd have a web page that I could navigate to and see the content, but which others would navigate to and couldn't see the content, unless they authenticated with FnordPlay with my credentials first (and then they could).
I see absolutely nothing that would interfere with that in the spec.
Slashdot Mirror
I'd call this an understatement. Seen this?
http://www.youtube.com/watch?v=v4boTbv9_nU
Granted: the Metro UI was genuinely innovative on a phone-sized device, and I can see that there are actually people for whom it's a better UI than iOS or Android. Microsoft did something actually innovative and good for the handheld with this UI.
But it does not work on a standard keyboard-and-mouse desktop environment. It also doesn't really work in the set-top environment -- I say this both because of my experiments with Kinect-driven Metro on the latest XBox, and because of my experiments with Windows 8 and 720p displays. (Windows 8 will not show you your address book on a 720p display, because it says it's not high resolution enough. At 1280x720. To show your address book.)
Now, it's possible that Microsoft will be able to force this through. But in my workplace, where people already have a choice between Windows, Linux, and MacOS, I'm expecting a noticeable uptick in Linux and MacOS desktop users over the 18 months after the release of Windows 8.
(I'm planning to familiarize myself more with Ubuntu/Unity in the near future, so I can be helpful during this time. And yes, I'm also familiarizing myself with Windows 8, running the consumer preview both natively and under VMs, and conducting experiments and compatibility tests. And my own day-to-day desktop environment is already MacOS.)
That "or (nowadays) Android" coupled with the kernel 3.3 news gives me some optimism for the future.
I'm really looking forward to, say, Debian or Ubuntu just bundling Dalvik and a large portion of the Android stack. I'm looking forward to the day when one of my alternatives to Office on the Linux desktop is actually the Android version of "Documents to Go", running full screen "Lion"-style or "Windows 8 Metro"-style.
(Likewise, I'm looking forward to CyanogeMod bundling large portions of Debian or Ubuntu, where it makes sense to do so, so I can maybe run Eclipse or at least Emacs on my Nook Color, with the full surrounding ecosystem that those environments expect.)
(Full disclosure: from about 1994 to 2000 I used Linux heavily everywhere, but today I only use it on my servers, using MacOS on the desktop/laptop. Also, I have an Android tablet -- a hacked Nook Color with CM7.1 on it -- but most of my mobile work today is on iOS. The reason is usability, coupled with enthusiasm for Objective-C left over from my NeXTstep days. But a merged Android and Linux might bring me back "into the fold", so to speak... though I still consider myself "in the fold", because Linux is my OS of choice for all my servers as it has been since the mid 1990s.)
It's true. In the mid-1990s, I was using a Newton connected to a phone line to receive faxes, annotate them, and send them back. It was crude, but extremely valuable to me at the time.
The devices running the General Magic software, "MagicCap", were a bit better at it, since it was almost all they could do!
(They were pen-based devices like the Newton, but unlike the Newton, they did not attempt handwriting recognition. So when you wrote a mail message to someone, you sent them a drawing, not anything a computer would recognize as text. The OS's infrastructure for handling that stuff was largely centered around "annotating" the documents with "stamps" or "stickers". Like, the way you'd mark a mail message as "important" was to annotate its envelope with what visually looked like an "important" rubber stamp. The OS bound meaning to the annotation. Pretty slick, for its day.)
(I still own four working Newtons and three working MagicCap devices. And yes, I also own several PalmOS and WinCE devices, and even an ancient Poqet, a palmtop with MS-DOS in ROM!)
Good point -- VoIP never even occurs to me, so I had forgotten to include it. You're completely right. Ditto for video.
I'd have no problem with 300 most of the time, but I'm not a fan of multiplayer games that require twitch reflexes, nor do I do anything significant on OnLive.
Another place you might have trouble is streamed content -- this is one of the reasons I do not like streamed content. I'd much rather use the model of "download the content, use it, and then discard it", which is much less sensitive to latency, lets you get higher quality regardless of bandwidth (as long as you allow the download time to be longer than the viewing time), and lets you schedule your traffic for times when the network isn't congested.
(Alas, it seems to me that content-owners think of streaming-instead-of-downloading as a form of DRM, and so attempt to make the other options look less attractive. Makes it more like TV/cable, I guess.)
...they are already going to have to include over 4,000 people in the game's credits. Anyone who donates $100 or more gets "special thanks in the game's credits".
(Yes, my name will be one of those.)
Just FYI: you can use off-the-shelf USB barcode scanners with the iPad. I have done it. If the barcode scanner presents itself as a USB keyboard, you just plug it into the iPad's camera connection kit, and the iPad then thinks it's got an external hardware keyboard connected. Scans work.
This isn't theory -- I tested it with my own USB barcode scanner. It does work.
You're explaining why the version of the Linux kernel commonly used for desktops today cannot provide all the features Android apps use today.
What's to stop someone from basing a desktop (or tablet) distribution on the Android kernel? Sure, it's not what the "Linux mainstream" is doing, but can't it be done?
(Maybe it can't, in which case I look forward to being educated. But I've seen fairly complete userspace tool collections available to add on to Android distributions, to the point where people talk about "Ubuntu for Android"...)
This is false:
1) We have DirecTV (in the United States), and some of our content is 1080p (It's mostly "on-demand" movies that download to the DVR's hard drive before playing, and take way longer for us to download than actually play back, but so what? I can let it download all day while I'm at work and then watch it in full quality when I get home in the evening. Fuck streaming, it's not the only option.).
2) We have multiple devices that send a signal to the TV that's locally generated, not a movie or TV show or something, and they're visibly better when connected at 1080p.
One example is our XBox 360. The user interface elements are sharper, for example.
Another example is an actual computer. They're coming with HDMI ports these days. No need for VGA or DVI or DisplayPort or whatever, just plug in the HDMI cable and get video and audio and everything. For this, 1080p is definitely better. Just try both and see.
(Heck, we're even thinking of getting an HDMI/USB KVM, since we've got multiple devices that use HDMI for both video and audio and use USB for peripherals. Then I'd be able to share a keyboard, mouse, and wired XBox game controllers across multiple systems.)
On the XBox 360, I can play four (well, five and a half, but four that count) different kinds of game: on-disc games, XBLA download games, "on-demand" download games, and "indie" games.
All four can be patched. Is the claim that even on the cheapest of these, a patch costs $40,000? That's a bit hard for me to swallow, because I've seen "indy" games get patched, and those are often from very small shops.
(There's also "original XBox" games, now obsolete, but also capable of being patched. There's also interactive content on video DVDs, but almost nobody distributes games that way, and these cannot be patched.)
I'm confused ... isn't Google Docs already this?
I'm already using it this way, using it as an arbitrary file store, from my desktop via the browser and on my iOS devices via multiple applications like "GoodReader".
The article didn't give me an idea of what they're adding. Transparent background sync with local filesystem? Standards-compliant WebDAV access?
Eh?
Loss-leaders are certainly of benefit to those that decide to engage in them. I think that's obvious. You seemed to think I didn't see that. What about my commend gave you that conclusion?
That does not mean that they're of net benefit to consumers, or to society as a whole.
The arguments you make are similar in character to the arguments against regulating pollution. Pollution regulations would not be put in place for the benefit of polluters, but for everyone else.
My preference is for the market to be made up of fair competition among a huge number of small players. I tend to have a negative view of practices that give incumbents a leg up. Is that enough on its own to explain my stance to you, or do you need for more of the dots to be connected?
That certainly would be nice, wouldn't it?
Why do you think there's a high risk of hurting their market share? Do you think most of the masses buying Android phones are buying them because they run Android? Do you think they care about the Android brand, or about Google services (like marketplace) specifically?
If so: you may be correct, but it's certainly not self-evidently obvious. Amazon's selling Kindle Fire systems like mad. Why couldn't Samsung do the same with a fork for smartphones?
Maybe. But not all of it.
I have been using PHP, for both business and personal stuff, since it was called "PHP/FI" ("personal home page / forms interface"). My old startup company would never have been successful without stuff we did in PHP. I've written and contributed C-based PHP extension code which was distributed with the core for years (until the underlying libraries fell so far out of use that there was no point including it anymore).
I still use PHP sometimes. But I prefer not to use it for important stuff that needs to stick around for a long time if it has any complexity.
Here's why: in my own personal experience (YMMV, of course), PHP has not been terrific about separating security fixes and bug fixes from feature evolution. It was far too easy to write code that would break with a version upgrade (eg. because an API or behavior changed), and to be forced into that version upgrade because security hole fixes were only available for the newest versions.
It's certainly the case that if you were very careful, you could avoid the problems. But, you had to be careful, and I saw a lot of novice programmers who weren't. (Our startup company provided PHP APIs for our product -- as well as Perl, Python, Java, and even TCL, this being the mid-to-late 1990s -- so I got to interact with a lot of novice PHP programmers.)
It's also certainly the case that for home/hobbyist use, you can mitigate some of the trouble by getting your PHP as part of a Debian distribution instead of getting it directly from upstream sources. The Debian folks are kinda insane about backporting security fixes to regarded-as-obsolete versions of software, and I love 'em for it. This is what I do on my household servers. (Yes, I still run my servers myself, not via a hosting provider -- have since about 1988, when they were Sun and VAX systems instead of the Linux I use today.)
It's certainly possible that PHP has gotten significantly better about this than it used to be. I haven't had reason to go check.
But anyhow, that's where my own ... hatred is too strong a word, let's say maybe "concern"? ... for PHP came from.
I don't know what you mean by "more catholic than the Pope" here.
Google has never made it clear that they do not mind unauthorized Market access. If they authorized it, I'd install it. But, were you around when the UNISYS patent came out of nowhere and nailed all the GIF users? I'm not going to start using something that I know has intellectual property landmines in it, trusting that the IP-owner is going to be reasonable. Been burned that way before.
They can make it clear that the use is authorized, or I can ignore it and use something else. Not going to risk it.
Can you send me a pointer to a full tablet that's faster, cheaper, has more features, and is no more difficult to use than a Nook Color, please? I've never seen or heard of one, myself.
Alas, not (unarguably) legally.
My only Android device is a Nook Color. Yes, I've got CM7 installed on a microSD card. But have you noticed that CM doesn't come with marketplace? Know why?
Yes, if you install CM on a device that originally came with marketplace, it's pretty well agreed upon that it's legal to back up that copy of marketplace, install CM, and then restore the marketplace. But my Nook Color never came with marketplace.
Some folks argue that if you have another Android device that has the Google apps (including marketplace), it's kosher to back them up from that device and restore them to a device that didn't have it, as long as they're both your devices. I actually do buy that argument. But, I have no other Android devices, so that argument does not apply to me.
There is no legal way for me to install the Google marketplace app on my own Nook Color, rooted or not, CM7 or not. If I'm unwilling to entertain the possibility of piracy, what are my options?
Fortunately, Amazon's Android marketplace is freely downloadable from Amazon's web servers. That lets me get a bunch of software without doing anything that anyone would interpret as piracy, so, when I'm running CyanogenMod, that's where all my apps come from. I simply have no other realistic option.
I do wish Google would say "look, it's okay to install this stuff, here's a spot where you can download it from, go nuts, as long as it's just for your own personal use and you're not repackaging it". I'd do so, and Google (and developers in the marketplace) would start getting some of my money. But as things stand, well, if Google doesn't want my money, okay, that's their call.
I think you're missing two things:
1) There are people who want a backlit color LCD display even for simple reading. While many people have eyes that are better off staring at e-ink, there are also many people for whom e-ink has problems, for whom looking at a backlit LCD is actually better. (Both I and Charlie Stross happen to be in that set. I recommend everyone try both -- forget propaganda/marketing, you need to figure out what's better for your eyes, under your reading conditions. Trust no one, test it yourself.)
2) There are people, and a lot of them, who want more than an e-reader, but less than a tablet. They want to consume not just B&W text, but color text, and in particular animations.
To elaborate on #2: a slate that can run e-reader software plus colorful animated "Dr. Seuss" books plus Netflix and Hulu is not (necessarily) a full tablet. A full-fledged tablet can do more than just consume content, and some people do just want to consume content (but more content than just books).
This does not impact the Nook Color in any significant way.
Both the Nook Color and Nook Tablet will try to boot off microSD first if they can. That's not part of the OS. However, the Nook Tablet requires a signed kernel to boot, and the Nook Color does not. So, this change results in a significant loss of hackability for the Nook Tablet, since you had to "jailbreak" it in some sense to do anything. It does not result in a significant loss of hackability for the older Nook Color, since you can still just write an unsigned kernel to a microSD card and you're off and running.
Disclaimer: this is my understanding from scouring the xda-dev forums for details and from hacking my own Nook Color. I've confirmed that 1.4.1 on the Nook Color does close the sideloading "hole", and that a 1.4.1 Nook Color will still boot stuff like CM7.1 from microSD card. The rest of it, I have not personally verified myself, but am summarizing my understanding from reading experts talking about it all.
..that it's in our best interests, as a society, to have legions of unemployed scientists out there with a grudge against those fools at the academy...
I agree -- it clearly needs to be "Xbox 361 for Workgroups".
I find it impossible to answer without knowing more about both your customers and the market you're operating in.
My own startup open sourced as much of our code as we legally could (though the term hadn't quite come into wide usage yet). We wrote software that implemented communication protocols (financial transaction protocols, specifically) that were only available under NDA and only "legal" to use after the software had undergone rigorous certification by financial institutions. We put that into the smallest C library we could, and shipped that library in both static and dynamic linked forms (so folks could replace it without rebuilding everything), and gave away the source code to everything on top of that, including the shims that made our C library available in Perl, Python, PHP, TCL, and Java. (In fact, this is how my code and name got into the core PHP distribution for years.)
We had a few "apps", like a GUI cash register and a batch script for running transactions stored in a CSV file, and we gave away the full source for those and encouraged our users to modify them and share the changes.
However, this only worked for us because of the market we were in (internet commerce) and who our customers were (people building out internet commerce in the mid 1990s, who were hackers who wanted to get real business done).
If our customers had been people who just wanted the apps, and never needed to tweak them, this could have been counterproductive from a business standpoint, and I might not own a seven bedroom house today. But we knew who our customers were and thankfully we techies were able to put this in place before we brought business types in to help us run things. (I suspect that if the business types had been there at the start, this strategy would have been vetoed.)
It's not that simple. If this were always true, I wouldn't own a seven bedroom house today.
(That said, I've no idea what the Zynga contracts are like.)
Help me figure this out: is there something about the <video> tag that makes DRM not work? Because if there is, I cannot see it.
You put in the video tag and you list sources. Let's for example assume I've got a FnordPlay protected MPEG-4 video file that I'm entitled to view. Everything I can see leads me to believe that if I put that file on my own web server, and put an HTML5 page with a video tag there next to it, I'd have a web page that I could navigate to and see the content, but which others would navigate to and couldn't see the content, unless they authenticated with FnordPlay with my credentials first (and then they could).
I see absolutely nothing that would interfere with that in the spec.