You are (almost) in luck. Firefox's integrated auto-updater will now, as part of the dialog telling you there's a new version and asking if you want to upgrade, list your extensions and highlight which are and aren't compatible (and lets you do a bulk "check for updates" at the same time). It's quite slick, I was impressed.
But that doesn't help you if you're upgrading from 2.0.x or if you're not receiving the new version through the built-in updater.
Microsoft sees itself competing with Sony for the living room: the PS3's loss is the XBox 360's gain. I'm not sure what Microsoft's long-term expectations were, but for the past two years it's been apparent that the longer that HD DVD remained viable (or, rather, the longer Blu-Ray failed to "win") the better the XBox 360 could compete the PS3 in the long term. And in the off-chance that HD DVD did win, that would be a huge blow to Sony and the PS3 -- and a gain for Microsoft, by default. Blu-Ray, having won, now makes the PS3 look a fair bit more attractive than it did a couple months ago.
I've heard that in general the licensing money Microsoft gets from HD DVD is not all that different from Blu-Ray (i.e., once you ignore the fact that all of the early HD DVD titles and none of the early Blu-Ray titles used VC-1).
I think it's fair to say that Microsoft's motivations were neither to see HD DVD specifically succeed, nor to see both formats to fail. They hedged their long-term bets on all sides (including the eventual streaming/downloading possibilities) and then saw that their best short-term strategy was to casually undermine and hold off Blu-Ray's domination. So: promote and contribute to HD DVD, but only in ways that don't cost a lot of money.
You realize that you can remap both of those shortcuts in System Preferences, right?
I'll grant that the default screenshot shortcut seems more than a bit arbitrary, but seriously: the vast majority of users don't take screenshots so often that it should have a dedicated F key out-of-the-box.
And with your Finder issue, you can also open a folder in list view by hitting Cmd-Down, which is (on most Mac keyboards) a whole lot easier to hit than most other individual keys on the keyboard.
Your video streams aren't going over the Internet; they're coming from some SaskTel server. To them it's all internal (i.e. free) bandwidth -- the streams' bandwidth does not affect SaskTel's upstream requirements, and costs them nothing.
5) it's gotta honk'in large cache so the 4800 rpm disk is not going to be that big a drag (afterall the macbooks and mac mini are only 5400 rpm and have smaller caches) What on earth are you talking about? If it's L2 cache... that's not at all relevant, and every single MacBook processor has 4MB on-die.
The Air's disk is 4200 rpm, not 4800, by the way, and if you were talking about the disk's buffer (which shouldn't be confused with cache), I haven't seen the size listed anywhere, at all, and there's no reason to believe it's anything special.
Well, the Mac Pro uses bloody expensive Fully Buffered RAM, and bleeding-edge Intel Xeon chips, neither of which are even desired by cost-conscious home users, but your point mostly stands. An Apple-branded Intel desktop machine that's (on the inside) more or less like every other normal Intel desktop machine. Price it at a premium, it'll still sell.
Of course, there's perhaps some irony here: Apple does not sell (and has never sold) any machine that uses Intel's Desktop processors. The Minis and iMacs both use notebook parts, and the XServes and Mac Pros use server/pro-workstation parts. There might be some interesting stipulations about this in whatever deal exists between Intel and Apple.
And all that obviously goes double for the new MacBook Air.
It seems to me -- and not just because I'm describing my own current position -- that many, if not most, of the people who've been buying MacBooks over the past two years are first-time Mac users. To extend the RV metaphor, eventually these people will get tired of traveling and want to go home. They might probably want to continue the Mac experience there! But they already have perfectly good monitors and keyboards and mice, and they know how nice it is to be able to toss in more hard drives and RAM and swap video cards a couple years down the line.
Apple is missing the opportunity to lock these people in as dedicated Mac users.
I do that too, sometimes, but find the required angles and desk space to be quite awkward. The MBP's screen only really works nicely when you're sitting at it's keyboard; trying to use the MBP's screen as a secondary screen while it's hooked into a normal keyboard/mouse/monitor desk setup is... hard to arrange, lopsided, and doesn't come close to the usefulness of a proper symmetric dual-head setup.
It's not a deal breaker, for me, but I do sympathize with the grandparent poster.
That's only half-true. Or, rather, the grandparent post is only half-untrue.
Neither Parallels nor Fusion support 3D acceleration to the hardware's full level of capability (i.e., DirectX 9 with shaders, which is what most recent games require). They don't give the virtualized system real access to the GPU -- instead, afaik, they're providing their own still-limited drivers within the virtualized OS that feed the calls back through OS X.
The games you list only work because they're old or designed to be compatible with older computers.
Parallels only supports up to DirectX 8, and the latest beta of Fusion has experimental DirectX 9 support but no support for shaders. Both programs are definitely progressing but there's still a significant tax to be paid in image quality, speed, and compatibility.
TransGaming's Cider implements common multimedia Windows APIs such as Direct3D, DirectInput, DirectSound and many others by mapping them to Mac equivalents.
It's virtualization at the API level, rather than the hardware level. I don't know if that was the grandparent poster's meaning, but I'd say it's a fair use of the term. The Mac's native OpenGL drivers are "virtualized" into DirectX drivers through the injection of an additional layer of abstraction that the original game doesn't realize is there. Just like virtualization software places an additional layer of abstraction between the OS and the real hardware that the OS doesn't realize is there, as a means of making it play nice with others.
You can argue over the definition, but the concept behind Cider is quite analogous to virtualization in the general sense.
Ahh, so you don't know about dithering, because your statement makes no sense.
Dithering is principally related to sample depth (eg 16 bit), not sample rate (44.1 kHz). Dithering is not a hack; it's quite ingenious and well researched. It has nothing in particular to do with 44.1 kHz CDs and everything to do with fundamental issues of digital sampling. Dithering (and it's fancier cousin, noise shaping) minimize quantization distortion. You want this. You need this. Done properly, the "noise" that it adds is invisible, while the distortion it eliminates is quite tangible.
No matter how fast or how deep you sample, you will still want dithering.
Actually it's not the normal backup power -- it's a secondary, emergency backup power:
The NRU reactor has eight independent main cooling pumps. In addition to the regular power supply, four of these pumps have a DC motor connected to backup power supplies - the normal backup supply. Two of those pumps, P-104 & P-105, are used to provide emergency core cooling and are to be connected to an additional back-up power supply, known as the Emergency Power Supply (EPS).
Judging from a more technical explanation, the issue with the pumps seems to have been widely misreported.
The NRU reactor has eight independent main cooling pumps. In addition to the regular power supply, four of these pumps have a DC motor connected to backup power supplies - the normal backup supply. Two of those pumps, P-104 & P-105, are used to provide emergency core cooling and are to be connected to an additional back-up power supply, known as the Emergency Power Supply (EPS). The issue is that those last two pumps, for unclear reasons, were not connected to the Emergency Power Supply (a backup to the backup) when they should have been.
The reactor is fifty years old, and has been subject to many many upgrades; the regulator reviews the state of things every couple years and asks for upgrades to keep pace with modern nuclear safety standards. The most recent review asked for this secondary backup, but it was discovered on a routine inspection that it had not been installed when they were under the impression it had been.
The talk of "pump installation" is actually about installing new motor starters for the pumps that are connected to the EPS; the first new motor starter had already been installed when the Government intervened to bring the reactor back online. The second one is being prepared and will now be installed in an upcoming scheduled maintenance. (Even under normal operation, the reactor is taken offline every two week for a few days.)
We have built an entire new plant. Two of them, each with double the capacity of the NRU. They've been built and operational for several years, but commissioning has been held up because of problems with the emergency shut-off systems that were discovered in testing.
Each of these plants is individually capable of (practically) meeting the world's demand for medical isotopes, once they passing commission testing.
It doesn't help, but it probably doesn't hurt given the low expectations for all timed release movie-to-game adaptations. They make the game because it will sell at predictable, profitable numbers to kids who like the movie and don't know to play better games. As far as the movie companies are concerned, it's like t-shirts or lunch boxes: they make them because people will buy them. Enough people to turn a profit. And, naturally, they make them as cheaply as they can without it hurting sales or causing complaints.
For what it's worth, after a year with a MacBook Pro, I actually do agree with you for the most part, but your personal example probably isn't the best one to illustrate this.
Apple does try to ensure things Just Work, but only if you do everything Steve's Way (or rather, the latest revision of Steve's Way). They want you to subscribe entirely to the Apple lifestyle and only support other standards and platforms when they're so stable and ubiquitous that ignoring them would have significant repercussions. They don't seem to particularly care about other usage models, and they never show any signs of caring about regressions in what they see as edge cases. Standard practice when people report these things seems to be to stonewall. I understand and agree that they need to prioritize on the things that affect 80%+ users, over hard little problems that frustrate 1% of users. It would be a lot nicer if they could drop the air of infallibility and actually just admit this from time to time.
I know that many people have had incredibly frustrating experiences of this. I've run up against little ones, here and there, but overall my experience has been positive. In the worst cases, there is no real workaround. In yours, there's a cheap hardware solution which is probably what should've been done in the first place. (In other words, I'm not sure I can really fault Apple all that much for ignoring your particular problem.)
You have a household with at least half a dozen active computers, including a web server, but you're too cheap to spend $20 to get a proper wifi access point? Sharing the connection through an ad hoc wireless network is not really a recommended long-term practice. It doesn't matter what OS you're using: the wifi card in your desktop is not a replacement for a proper access point.
Indeed, I know of no example where the wisdom of crowds is really effectively demonstrated. Ant colonies. The behaviour of an individual ant is, unsurprisingly, quite simple and predictable (allowing for a certain degree of randomness). But a colony of ants is actually quite smart and adaptable: with a few thousand ants all walking around, there is an emergent behaviour, and collectively they are successful. The ants aren't voting, and there's no central decision-making -- they're all just doing what they're programmed to do (ie, "feed myself, and help my siblings find food and shelter") using largely independent decisions and knowledge, and the overall movement of that crowd exhibits greater wisdom through the process.
Your representation of the "wisdom of crowds" argument is a bit off-the-mark. Firstly, not all crowds are wise; the book spends quite a while exploring what is needed to make a wise crowd, and the contexts in which to use them. Secondly, part of the idea is that a large crowd will likely contain better experts than a small group of generalists, and you're better off listening to those experts than trying to do it all yourself. That part seems obvious -- we invented it a few thousands years ago and called it civilization -- but the communication and information tools of the Internet now allow us to exploit this diversity in ways we're only just beginning to grasp.
The biggest exposure is in combination with insecure home WiFi, and if you've got that you're already skiing naked through a briar patch.
I'd expect that an equal-or-bigger problem would be with notebooks, in situations like university campus wifi connections, which are relatively open and typically have large numbers of unadministered machines talking link-locally. A Bonjour virus could spread quickly in an environment link that (before admins caught on and started filtering), and then the movement of the machines themselves would carry it to other networks.
In such a situation, there is a certain irony in how the attack vector already is an efficient means of finding vulnerable machines.
I'm not disagreeing with you, but there is one thing I think is noteworthy and relevant:
The claimed vulnerability here is with mDNSResponder -- i.e., Bonjour/Zeroconf
By default, Mac OS X runs this service, listening to an open UPD port.
If, in the System Preferences tool, you turn the firewall on, turn off all listed services and open ports, and check off "Block UDP Traffic" -- in short, do everything the default GUI offers to lock down the system -- mDNSResponder is still running, and its incoming ports are still open. The only ways to block or disable it involve the command line. The command to do this are trivial and don't have any far-reaching side effects if you don't depend on Bonjour features, but the point stands: a vulnerability in this is potentially a vulnerability in practically every consumer Mac.
(Mostly this is a criticism that Apple doesn't include a simple and transparent preference for disabling Bonjour.)
The one limiting factor is that the multicast DNS traffic that mDNSResponder listens to isn't normally routed across the open Internet, but I'm told that it it is routable.
It's not a question of "vulnerability". The community has not "lost" anything. Apple now owns the source, and can go off and do their own things with it, but since it's been released under a mix of GPL and LGPL, anyone could fork the code and make their own version with a new name -- the only restriction is that the fork must stay under GPL/LGPL.
That's the fundamental part of how OSI-approved open source licenses work: the copyright holder cannot "take away" already released code by later changing the license to something more restrictive.
I'm not the grandparent poster, but I just want to point out that in both those incidents, it looks like the confusion was actually a result of internal miscommunication rather than indecision (which is how I would interpret "flip-flopping"). The marketing people not reading last week's emails from the engineers, effectively.
Your point still stands, however: Off-hand comments and speculation of this nature is not remotely reliable at the moment.
There are different types of LED backlight technology -- generally, in comparison with CCFL (cold cathode fluorescent lamp) backlights it breaks down into:
1) the type of LED backlight that uses lots of power but is brighter, offers wider gamut, and is now starting to appear in high-end desktop LCDs.
2) the type of LED backlight that uses somewhat less power, offers roughly equivalent brightness gamut, and is now starting to appear in notebook LCDs.
The new MBPs, of course, use the latter. The technology for the first type can't (or hasn't yet) been made thin enough to suite a notebook LCD, and uses up too much power. Note that the primary reason cited by Apple for the switch was environmental concerns: CCFLs have lead and are hard to dispose of cleanly. Not improving image quality.
The complaints from the past six months about MacBook Pro screens are incredibly uninformed. The screens do have some issues -- but it's not what most people think. First, the ~45% NTSC color gamut and 6-bit pixel depth offered by MacBook Pro screens is the same as every other notebook screen you can possibly buy in volume. People asking for better fail to realize it doesn't exist.
Apple, like most other companies in their position, actually uses a different models of LCD panel from a few different manufacturers in the MacBook Pro, based on current supply and logistics. The problem is that two of the three(?) models actively used by Apple over the past six months are somewhat crappy. There's no way to know in advance which is in a given machine, and it's hard to look up even if it's in front of you, so whether or not you get a 'good' screen is luck of the draw based on where and when the machine was assembled.
The actual issues are that many of those LCD panels have:
a) Somewhat uneven backlighting. b) Extremely narrow range of 'good' viewing angles, which makes the backlighting seem much more uneven. c) An odd/excessive anti-glare coating that causes the "grainy" or "sparkly" effect people have been complaining about.
None of those are things Apple has made any claim about you could use for a lawsuit. Many people have mistakenly attributed that last point to the screen's (completely standard) dithering of 6-bits to 8-bits. Dithering involves rapidly changing pixels -- it simply does not produce the visible, fixed patterns the characterize the "grainy" appearance people complain about.
Moving to LED-backlit panels changes the nature of a). It might improve it, it might hamper it; I really don't know. Simply avoiding the cheaper manufacturers could alleviate b) -- and, afaik, those manufacturers don't have 15.4" LED-backlit models yet. The issue with the anti-glare coating is up to Apple. Hopefully they've had enough complaints to figure out the problem.
So, these new screens? Probably somewhat better than the previous ones, but for completely different reasons than most people are asking about.
Slashdot Mirror
You are (almost) in luck. Firefox's integrated auto-updater will now, as part of the dialog telling you there's a new version and asking if you want to upgrade, list your extensions and highlight which are and aren't compatible (and lets you do a bulk "check for updates" at the same time). It's quite slick, I was impressed.
But that doesn't help you if you're upgrading from 2.0.x or if you're not receiving the new version through the built-in updater.
Microsoft sees itself competing with Sony for the living room: the PS3's loss is the XBox 360's gain. I'm not sure what Microsoft's long-term expectations were, but for the past two years it's been apparent that the longer that HD DVD remained viable (or, rather, the longer Blu-Ray failed to "win") the better the XBox 360 could compete the PS3 in the long term. And in the off-chance that HD DVD did win, that would be a huge blow to Sony and the PS3 -- and a gain for Microsoft, by default. Blu-Ray, having won, now makes the PS3 look a fair bit more attractive than it did a couple months ago.
I've heard that in general the licensing money Microsoft gets from HD DVD is not all that different from Blu-Ray (i.e., once you ignore the fact that all of the early HD DVD titles and none of the early Blu-Ray titles used VC-1).
I think it's fair to say that Microsoft's motivations were neither to see HD DVD specifically succeed, nor to see both formats to fail. They hedged their long-term bets on all sides (including the eventual streaming/downloading possibilities) and then saw that their best short-term strategy was to casually undermine and hold off Blu-Ray's domination. So: promote and contribute to HD DVD, but only in ways that don't cost a lot of money.
You realize that you can remap both of those shortcuts in System Preferences, right?
I'll grant that the default screenshot shortcut seems more than a bit arbitrary, but seriously: the vast majority of users don't take screenshots so often that it should have a dedicated F key out-of-the-box.
And with your Finder issue, you can also open a folder in list view by hitting Cmd-Down, which is (on most Mac keyboards) a whole lot easier to hit than most other individual keys on the keyboard.
Your video streams aren't going over the Internet; they're coming from some SaskTel server. To them it's all internal (i.e. free) bandwidth -- the streams' bandwidth does not affect SaskTel's upstream requirements, and costs them nothing.
The Air's disk is 4200 rpm, not 4800, by the way, and if you were talking about the disk's buffer (which shouldn't be confused with cache), I haven't seen the size listed anywhere, at all, and there's no reason to believe it's anything special.
I'd add "Voice Recorder" as the fourth most-requested item, which the E200 also has.
Well, the Mac Pro uses bloody expensive Fully Buffered RAM, and bleeding-edge Intel Xeon chips, neither of which are even desired by cost-conscious home users, but your point mostly stands. An Apple-branded Intel desktop machine that's (on the inside) more or less like every other normal Intel desktop machine. Price it at a premium, it'll still sell.
Of course, there's perhaps some irony here: Apple does not sell (and has never sold) any machine that uses Intel's Desktop processors. The Minis and iMacs both use notebook parts, and the XServes and Mac Pros use server/pro-workstation parts. There might be some interesting stipulations about this in whatever deal exists between Intel and Apple.
And all that obviously goes double for the new MacBook Air.
It seems to me -- and not just because I'm describing my own current position -- that many, if not most, of the people who've been buying MacBooks over the past two years are first-time Mac users. To extend the RV metaphor, eventually these people will get tired of traveling and want to go home. They might probably want to continue the Mac experience there! But they already have perfectly good monitors and keyboards and mice, and they know how nice it is to be able to toss in more hard drives and RAM and swap video cards a couple years down the line.
Apple is missing the opportunity to lock these people in as dedicated Mac users.
I do that too, sometimes, but find the required angles and desk space to be quite awkward. The MBP's screen only really works nicely when you're sitting at it's keyboard; trying to use the MBP's screen as a secondary screen while it's hooked into a normal keyboard/mouse/monitor desk setup is ... hard to arrange, lopsided, and doesn't come close to the usefulness of a proper symmetric dual-head setup.
It's not a deal breaker, for me, but I do sympathize with the grandparent poster.
That's only half-true. Or, rather, the grandparent post is only half-untrue.
Neither Parallels nor Fusion support 3D acceleration to the hardware's full level of capability (i.e., DirectX 9 with shaders, which is what most recent games require). They don't give the virtualized system real access to the GPU -- instead, afaik, they're providing their own still-limited drivers within the virtualized OS that feed the calls back through OS X.
The games you list only work because they're old or designed to be compatible with older computers.
http://en.wikipedia.org/wiki/Comparison_of_VMware_Fusion_and_Parallels_Desktop
Parallels only supports up to DirectX 8, and the latest beta of Fusion has experimental DirectX 9 support but no support for shaders. Both programs are definitely progressing but there's still a significant tax to be paid in image quality, speed, and compatibility.
You skipped the next sentence:
TransGaming's Cider implements common multimedia Windows APIs such as Direct3D, DirectInput, DirectSound and many others by mapping them to Mac equivalents.
It's virtualization at the API level, rather than the hardware level. I don't know if that was the grandparent poster's meaning, but I'd say it's a fair use of the term. The Mac's native OpenGL drivers are "virtualized" into DirectX drivers through the injection of an additional layer of abstraction that the original game doesn't realize is there. Just like virtualization software places an additional layer of abstraction between the OS and the real hardware that the OS doesn't realize is there, as a means of making it play nice with others.
You can argue over the definition, but the concept behind Cider is quite analogous to virtualization in the general sense.
Ahh, so you don't know about dithering, because your statement makes no sense.
Dithering is principally related to sample depth (eg 16 bit), not sample rate (44.1 kHz). Dithering is not a hack; it's quite ingenious and well researched. It has nothing in particular to do with 44.1 kHz CDs and everything to do with fundamental issues of digital sampling. Dithering (and it's fancier cousin, noise shaping) minimize quantization distortion. You want this. You need this. Done properly, the "noise" that it adds is invisible, while the distortion it eliminates is quite tangible.
No matter how fast or how deep you sample, you will still want dithering.
The reactor is fifty years old, and has been subject to many many upgrades; the regulator reviews the state of things every couple years and asks for upgrades to keep pace with modern nuclear safety standards. The most recent review asked for this secondary backup, but it was discovered on a routine inspection that it had not been installed when they were under the impression it had been.
The talk of "pump installation" is actually about installing new motor starters for the pumps that are connected to the EPS; the first new motor starter had already been installed when the Government intervened to bring the reactor back online. The second one is being prepared and will now be installed in an upcoming scheduled maintenance. (Even under normal operation, the reactor is taken offline every two week for a few days.)
We have built an entire new plant. Two of them, each with double the capacity of the NRU. They've been built and operational for several years, but commissioning has been held up because of problems with the emergency shut-off systems that were discovered in testing.
Each of these plants is individually capable of (practically) meeting the world's demand for medical isotopes, once they passing commission testing.
It doesn't help, but it probably doesn't hurt given the low expectations for all timed release movie-to-game adaptations. They make the game because it will sell at predictable, profitable numbers to kids who like the movie and don't know to play better games. As far as the movie companies are concerned, it's like t-shirts or lunch boxes: they make them because people will buy them. Enough people to turn a profit. And, naturally, they make them as cheaply as they can without it hurting sales or causing complaints.
For what it's worth, after a year with a MacBook Pro, I actually do agree with you for the most part, but your personal example probably isn't the best one to illustrate this.
Apple does try to ensure things Just Work, but only if you do everything Steve's Way (or rather, the latest revision of Steve's Way). They want you to subscribe entirely to the Apple lifestyle and only support other standards and platforms when they're so stable and ubiquitous that ignoring them would have significant repercussions. They don't seem to particularly care about other usage models, and they never show any signs of caring about regressions in what they see as edge cases. Standard practice when people report these things seems to be to stonewall. I understand and agree that they need to prioritize on the things that affect 80%+ users, over hard little problems that frustrate 1% of users. It would be a lot nicer if they could drop the air of infallibility and actually just admit this from time to time.
I know that many people have had incredibly frustrating experiences of this. I've run up against little ones, here and there, but overall my experience has been positive. In the worst cases, there is no real workaround. In yours, there's a cheap hardware solution which is probably what should've been done in the first place. (In other words, I'm not sure I can really fault Apple all that much for ignoring your particular problem.)
You have a household with at least half a dozen active computers, including a web server, but you're too cheap to spend $20 to get a proper wifi access point? Sharing the connection through an ad hoc wireless network is not really a recommended long-term practice. It doesn't matter what OS you're using: the wifi card in your desktop is not a replacement for a proper access point.
Your representation of the "wisdom of crowds" argument is a bit off-the-mark. Firstly, not all crowds are wise; the book spends quite a while exploring what is needed to make a wise crowd, and the contexts in which to use them. Secondly, part of the idea is that a large crowd will likely contain better experts than a small group of generalists, and you're better off listening to those experts than trying to do it all yourself. That part seems obvious -- we invented it a few thousands years ago and called it civilization -- but the communication and information tools of the Internet now allow us to exploit this diversity in ways we're only just beginning to grasp.
The biggest exposure is in combination with insecure home WiFi, and if you've got that you're already skiing naked through a briar patch.
I'd expect that an equal-or-bigger problem would be with notebooks, in situations like university campus wifi connections, which are relatively open and typically have large numbers of unadministered machines talking link-locally. A Bonjour virus could spread quickly in an environment link that (before admins caught on and started filtering), and then the movement of the machines themselves would carry it to other networks.
In such a situation, there is a certain irony in how the attack vector already is an efficient means of finding vulnerable machines.
I'm not disagreeing with you, but there is one thing I think is noteworthy and relevant:
The claimed vulnerability here is with mDNSResponder -- i.e., Bonjour/Zeroconf
By default, Mac OS X runs this service, listening to an open UPD port.
If, in the System Preferences tool, you turn the firewall on, turn off all listed services and open ports, and check off "Block UDP Traffic" -- in short, do everything the default GUI offers to lock down the system -- mDNSResponder is still running, and its incoming ports are still open. The only ways to block or disable it involve the command line. The command to do this are trivial and don't have any far-reaching side effects if you don't depend on Bonjour features, but the point stands: a vulnerability in this is potentially a vulnerability in practically every consumer Mac.
(Mostly this is a criticism that Apple doesn't include a simple and transparent preference for disabling Bonjour.)
The one limiting factor is that the multicast DNS traffic that mDNSResponder listens to isn't normally routed across the open Internet, but I'm told that it it is routable.
It's not a question of "vulnerability". The community has not "lost" anything. Apple now owns the source, and can go off and do their own things with it, but since it's been released under a mix of GPL and LGPL, anyone could fork the code and make their own version with a new name -- the only restriction is that the fork must stay under GPL/LGPL.
That's the fundamental part of how OSI-approved open source licenses work: the copyright holder cannot "take away" already released code by later changing the license to something more restrictive.
I'm not the grandparent poster, but I just want to point out that in both those incidents, it looks like the confusion was actually a result of internal miscommunication rather than indecision (which is how I would interpret "flip-flopping"). The marketing people not reading last week's emails from the engineers, effectively.
Your point still stands, however: Off-hand comments and speculation of this nature is not remotely reliable at the moment.
There are different types of LED backlight technology -- generally, in comparison with CCFL (cold cathode fluorescent lamp) backlights it breaks down into:
1) the type of LED backlight that uses lots of power but is brighter, offers wider gamut, and is now starting to appear in high-end desktop LCDs.
2) the type of LED backlight that uses somewhat less power, offers roughly equivalent brightness gamut, and is now starting to appear in notebook LCDs.
The new MBPs, of course, use the latter. The technology for the first type can't (or hasn't yet) been made thin enough to suite a notebook LCD, and uses up too much power. Note that the primary reason cited by Apple for the switch was environmental concerns: CCFLs have lead and are hard to dispose of cleanly. Not improving image quality.
The complaints from the past six months about MacBook Pro screens are incredibly uninformed. The screens do have some issues -- but it's not what most people think. First, the ~45% NTSC color gamut and 6-bit pixel depth offered by MacBook Pro screens is the same as every other notebook screen you can possibly buy in volume . People asking for better fail to realize it doesn't exist.
Apple, like most other companies in their position, actually uses a different models of LCD panel from a few different manufacturers in the MacBook Pro, based on current supply and logistics. The problem is that two of the three(?) models actively used by Apple over the past six months are somewhat crappy. There's no way to know in advance which is in a given machine, and it's hard to look up even if it's in front of you, so whether or not you get a 'good' screen is luck of the draw based on where and when the machine was assembled.
The actual issues are that many of those LCD panels have:
a) Somewhat uneven backlighting.
b) Extremely narrow range of 'good' viewing angles, which makes the backlighting seem much more uneven.
c) An odd/excessive anti-glare coating that causes the "grainy" or "sparkly" effect people have been complaining about.
None of those are things Apple has made any claim about you could use for a lawsuit. Many people have mistakenly attributed that last point to the screen's (completely standard) dithering of 6-bits to 8-bits. Dithering involves rapidly changing pixels -- it simply does not produce the visible, fixed patterns the characterize the "grainy" appearance people complain about.
Moving to LED-backlit panels changes the nature of a). It might improve it, it might hamper it; I really don't know. Simply avoiding the cheaper manufacturers could alleviate b) -- and, afaik, those manufacturers don't have 15.4" LED-backlit models yet. The issue with the anti-glare coating is up to Apple. Hopefully they've had enough complaints to figure out the problem.
So, these new screens? Probably somewhat better than the previous ones, but for completely different reasons than most people are asking about.