Try ALL Disney DVDs, not just animated ones. Some of them aren't even movie previews. I've seen outright ads for toys and other such crap on there. I'm paying them for the privilege of watching their movie, and then they're forcing me to watch that stuff? Not gonna happen. Stick the disc in, switch to my TiVo, watch a 30-minute show, switch back, and watch the movie.
There actually was one shuttle abort, STS-51F---one of Challenger's last missions. Also, STS-93 ended up in a lower orbit due to a fuel leak in one of the main engines. They didn't classify it as an ATO because they just plain ran out of oxidizer; no human explicitly hit an abort button. That same flight also had a major electrical short causing multiple main engine controller failures on that flight. Had the backup controllers also failed, you would doubtless have seen an ATLS on that flight.
And of course, there have only been two in-flight catastrophes, neither of which was detected in time to prevent it using an abort. The final Challenger and Columbia missions did not result in aborts because nobody hit the abort button, not because an abort wouldn't have been possible, given an early enough determination of the seriousness of the underlying failure event.
Either way, probably the biggest reason we don't see lots of shuttle disasters is that there are so many redundant systems and electronic safety checks on the shuttle. And it's a good thing, too, given the rate of failures. Most failures, statistically speaking, happen and are detected before the bird gets into the air. The number of launchpad aborts for the shuttle is significant. By my count, pad aborts occurred in almost one in ten launches (14 aborts out of 145 launch attempts). And that's not counting scrubbed launches due to weather or other non-mechanical causes. That's just the actual equipment failures occurring prior to SRB ignition....
As an aside, if airplanes had even one tenth the engine failure rate that the shuttles have seen, we'd see an airplane flight cancelled once per minute in the U.S. alone. Food for thought next time you start pondering whether the government should be in the business of designing payload launch systems without adult supervision.
Possibly. It certainly isn't completely out of the question. Either way, we'll never know if they could have done anything because they weren't given the opportunity to try.
First, the shuttle can land safely with certain tiles missing. It does that with regularity. There's a sizable safety margin everywhere but the leading edge of the wing. They could have spacewalked somebody to pry a tile or two off of less sensitive areas (e.g. near the OMS engines where a dozen tiles fell off during Columbia's maiden flight) and affixed them to the leading edge of the wing. They wouldn't have been as good as the higher-temperature tiles that are supposed to be there, but they would have been better than a gaping hole. Secondarily, they could have affixed additional metal plating over the affected area, ripped from consoles inside the orbiter and bent to suit, using whatever screws and drills that they had handy. You know, an Apollo-style repair.
Second, had the orbiter been reprogrammed by ground engineers, it might have been possible to land at an angle that takes some of the heat off of the damaged wing, either by favoring the opposite side during the hottest parts of reentry or by adjusting the angle of entry to be steeper, thus A. putting more heat on the bottom tiles and less on the leading edge, and B. dropping them more quickly into an area with greater air density against the wing that might provide better thermal transfer. I'm not sure if that last part would make up for the extra heating due to falling like a rock, but that's for somebody to actually model, not for me to guess about. In any case, they could have experimented with several computer models to find the reentry vector that gave them the best chance at survival instead of just using the standard reentry vector.
Third, in the absence of a way to patch things up, the U.S. military has thousands of launch vehicles that could readily be hacked up to carry a food and power payload into space on short notice. We call them ICBMs. They're theoretically fueled up and ready to launch at a moment's notice, and they have enough thrust to match the shuttle's speed. Rip out the warheads, carefully balance your payload components, fasten them in properly, put the thing up there, match speed with the shuttle, blow the payload in the general direction of the cargo bay, then restart the missile's engines to bring it back down to Earth. Don't get me wrong, hitting the target would make the missile defense tests look like a cakewalk by comparison, but it would also involve precisely known trajectories and two vehicles that can, at least to some extent, adjust course to meet each other.... And of course, there are enough missiles sitting in silos that if you screwed up with the first one, you could launch another one 90 minutes later on the next trip around. It would have been the single most expensive piece of foam in the history of the world, but then again, it was anyway, so....
Finally, there was a Soyuz ISS mission scheduled just two months later that could have been diverted as an emergency return vehicle. Pulled back a few weeks and combined with an ICBM for supply delivery, there is every possibility that they could have gotten the crew home safely, at which time the shuttle could have been landed on autopilot (optionally out in the middle of the ocean or desert if they were sufficiently worried) without risk to the crew.
Either way, regardless of whether it would or would not have been possible to save the crew, if they had put thousands of people to work looking for a solution, there's at least some possibility that they might have found one. As it stood, they didn't, so they couldn't.
No, minor failures happen with regularity. It's just selective memory. Nobody remembers the minor ones. The last minor shuttle equipment failure was just five weeks ago.
Yes, and no. There are certain classes of failures that are not preventable, sure, but in the case of both shuttle disasters, the accidents were entirely preventable. The first one was caused by NASA ignoring the thermal specifications for the SRBs and launching anyway. The second one was caused by NASA ignoring all the engineers screaming for inspection of the heat tiles and landing anyway. In both cases, the primary fault rests squarely on human error, and the secondary fault lies in design error---a single design error, in fact.
Both shuttle disasters were caused by a single design error. The shuttle is on the side of the stack instead of on top. Had the Challenger been on top of the stack, it would not been so severely damaged by the explosion, and the crew could have readily performed an abort-to-launch-site or similar. Had the Columbia been on top of the stack, it would not have gotten hit by flying chunks of foam. Sure, there's still the risk of bird impacts, but the risk is much, much lower than the near-inevitable foam impacts
There were, of course, other design flaws that contributed---insufficient metal shielding on the leading edge of the wing that makes the thermal tiles more critical than necessary, dubious lack of enough layers of O-rings in the SRBs, lack of an emergency capsule ejection system that activates automatically in the event of catastrophic failure, poor weight-to-surface-area ratio for landing in order to reduce drag during launch, the mere existence of foam on the outside of a tank instead of sealed inside it, and loads of other design decisions or oversights that in retrospect probably weren't the best idea, but ultimately it was the placement of the shuttle on the side of the stack that proved to be its undoing on two separate occasions. Such a fundamental mistake isn't easy to correct without throwing out the design completely and starting over.
I'd usually agree for important variables that get used throughout a function, but there's nothing more pointless than having to put loop iterators that only get used in two lines of code up at the top of the function. It just makes things so much harder to read than:
Just out of curiosity, what part of "No peasants are starving, our economy is great, everyone else in the world envies/fears us, and by the way we just perfected nuclear fusion!" is a *mild* hallucination?
It's only a hallucination/delusion if he believes it. Otherwise, it's just Pseudologia fantastica.
You're right. I should have called it an internet hosting provider. Either way, the DMCA safe harbor very clearly defines "service provider" to include hosting providers:
(1) Service provider. -- (A) As used in subsection (a), the term "service provider" means an entity offering the transmission, routing, or providing of connections for digital online communications, between or among points specified by a user, of material of the user's choosing, without modification to the content of the material as sent or received.
(B) As used in this section, other than subsection (a), the term "service provider" means a provider of online services or network access, or the operator of facilities therefor, and includes an entity described in subparagraph (A).
Note that subsection (a) is the section "Transitory Digital Network Communications", which doesn't really apply to hosting providers in any meaningful way, hence "service providers" is defined differently for that section. For all other parts, hosting providers are service providers. In particular, the relevant safe harbor rule is 512(c), "Information Residing on Systems or Networks at Direction of Users," which is covered by definition (B).
Agreed. Hear that sound? That's the sound of the world's tiniest violin playing the saddest song in the world.
Sorry, Hollywood, but YouTube is precisely the kind of site that safe harbor exemptions are for. It is an ISP whose material is provided exclusively by its users. What, you thought that somehow a site shouldn't count because it is getting too big for you to bully? Waaaah. Cry me a freaking river.
It has occurred to many people. Indeed, that's precisely what extended attributes are for. However, it isn't feasible for a filesystem to provide multiple views into the data based on EA searches, as that would either require a very rigorously defined set of valid EA structures that can be searched or would require extensive application-specific knowledge.
Windows made an attempt at putting that sort of multi-view design into their filesystem with WinFS and half a decade later, it hasn't shipped. Apple tried the "leave it up to the importer" approach with Spotlight, and that actually worked reasonably well, though there are a few warts. Either way, the specific views that are most useful tend to be application-specific, so anything beyond keyword searches is problematic to implement in a general way (which is why nobody does it).
The thing is, apps like iPhoto already have abstractions that are better than filesystems. As long as the collection metaphor is hierarchical, they inherently scale at least as well as a filesystem does (because you can organize them in precisely the same way if you want). However, they also provide a much richer set of navigation than a filesystem can ever realistically provide by providing multiple views into the data.
For example, in iPhoto (personally, I use Lightroom, but the same principle applies), they have tagging, they have folders/collections, they have faces and places.... They have the ability to view things not just in a strict hierarchical view like a filesystem, but also in numerous logical views (e.g. show me photos taken in France with Pierre in them) that go way beyond what any filesystem view can feasibly provide. Thus, when it comes to larger data sets, filesystems generally get in the way far more than they help.
Note: I'm not saying iPhoto was always good about this. I do remember a time just a few versions back when scrolling through my iPhoto library took a small eternity. So I can certainly understand having trouble dealing with large libraries in iPhoto. However, that's not a flaw in the concept so much as a flaw in the implementation of that concept.
Not at all. One of the biggest flaws in computer UI design today is that there are lots of things that are not stored as files but are still basically indivisible units of data, whether they're mail messages or database records or... you name it. Because so many of these things are not, in fact, files, a purely file-based view is a fairly clumsy way to represent that content. For most users, they don't need to know or care whether data is in a file or a database record or an email message in an mbox file. Abstracting those details away from the user results in a better user experience with more ability to manage the actual content than a pure file-based interface can provide.
It's not like the filesystem in Ubuntu Light will cease to exist or will become inaccessible to power users. You'll just have to install tools to reach it. At least I assume that this is the case.
Other way around. AT&T works well in population centers. It is terrible in rural areas compared with Verizon. My parents regularly travel a 30 mile stretch of major 4-lane highway in TN with solid Verizon coverage and no AT&T coverage whatsoever. Here in California, when you get up into the hills, Verizon works, AT&T doesn't. And so on. AT&T's 2G coverage is anything but incredible. Their 3G coverage is simply an abomination, so people are so shellshocked that the 2G coverage seems great by comparison....
AT&T's 3G coverage works exclusively in major population centers; people have problems because AT&T's 3G coverage doesn't fail over to 2G very well. Disable 3G voice and you'll have many fewer dropped calls and a much better overall experience.
I would agree with you about the thin thing if you were talking about the MacBook Air, but an iPad really doesn't seem like it was intended to be your main computing device. It seems designed to be a device for carrying around content from your computer, working with it when you're on the go, then syncing it back to your computer when you return. When you're tethered to a desk somewhere, you'll probably use a traditional computer, not an iPad, so having connectors on the device is of marginal utility unless it's something you think you'll need with you when you're traveling. Having lots of connectors is antithetical to the entire concept of an iPad because hooking it up to lots of peripherals and tying yourself down to a desk is exactly the opposite of the way it was designed to be used.
Don't get me wrong, I'm not about to buy one. If my main machine were a desktop, I had no iPhone, and my portable needs were less esoteric, I could see it being a useful device, particularly for airplane flights. Unfortunately, the largest one could barely hold my photo library alone. And, of course, there's not an app for most of the things I use my laptop for anyway (writing software, composing large ensemble musical scores, designing websites, etc.). For my parents, on the other hand, were it not for AT&T's roughly 30-mile-long dead spot on their regular commute, I'd have already ordered one just so they could check email and browse the web while away from home. Different tools for different users.
Actually, it does. If you can't feasibly control who has access to information, then any open, distributed alternative to Facebook is going to have exactly zero success at preventing the same sort of data mining that they're complaining about Facebook allowing... unless, of course, you deliberately obfuscate the data so that it is hard to mine in any reasonable programmatic fashion, but doing so would also make it much less useful for your actual friends.
Don't get me wrong, Facebook shouldn't be playing fast and loose with privacy, and it's important to chew them out on principal every time they screw up, particularly as it applies to someone's public profile. However, I'm not convinced that any similar service that anybody else comes up with won't have at least as many similar problems. Coming up with good access policies is hard. Designing user-interfaces that expose those settings in a way your grandparents can understand is even harder.
First, those standards are pure algorithmic patents, and thus are probably invalid to begin with. It's just a question of somebody making the right arguments in a patent nullification suit.
Second, even if we assume the patents are valid, companies would not be "forced to give up their IP if they join a consortium." Companies would be free to declare IP that they care about to the consortium but not license it to them, at which point it could not be part of the standard, but they would retain full rights to it.
Look, I have exactly zero sympathy for any company that pushes their patented technology on the world as part of an open standard. Interchange standards should be just that---standards that can be used anywhere. Patents and cross-platform compatibility/ubiquity are fundamentally incompatible. We either do away with patents for file formats or those patents do away with compatibility. It's really that simple.
That said, adding patent laws to cover consortiums is still the wrong way to handle this. Consumers should have a fundamental right to any data of their creation, period, whether they are encoded using software from a consortium or from an individual corporation. What we need is a fair use law that applies to file format patents, e.g. something like this:
Use of a patent without a license is not infringement if all of the following conditions are met:
a.) The patent is or has ever been licensed for use in software or hardware designed for the storage, retrieval, encryption, decryption, compression, or decompression of user-generated data, including (but not limited to) photos, video, audio, or text.
b.) Either:
1.) It is generally believed to be impractical to design software to convert between the stored form and the original, uncompressed, unencrypted user data without infringing the patent or
2.) The embodiment of the invention is in the form of hardware and the sole infringement is the use of that hardware by an end user in a way that falls outside the scope of the patent license.
c.) The primary purpose for the unlicensed used of the patent is for manipulating user-generated data originally produced by a licensed implementation.
d.) Either:
1.) The patent is not strictly limited to covering the mechanical act of storing data on a physical medium or
2.) The patent is strictly limited to covering the mechanical act of storing data on a physical medium, but the licensed hardware does not make the data available to the end user.
Boom. No more problems with interchange formats and patents, all with minimal collateral damage. It's basically the patent equivalent of the DMCA's compatibility exemption. For example, this protects things for which alternatives exist (e.g. novel ways to implement MPEG in hardware), but does so in a way that protects consumers' right to their own data and their right to use hardware that they paid for.
It could compress the entire screen update into a single byte and it still wouldn't matter. The latency of cellular communications would still kill you. Imagine typing a keystroke and waiting a quarter second for it to show up on your screen. And that's the round-trip time using ssh. Any screen-based protocol is going to add additional overhead on top of that.
Multi-hundred-millisecond latency figures are not at all unusual when you're dealing with cellular data, and if you're out of 3G territory, you're talking about whole seconds.
Okay, so, your solution to using your under-powered over-price feature-less piece of crap is to go out and spend more exorbitant amounts of money?
If you're saying that iPhone should have those connectors exposed, I would point out that USB and VGA connectors would make it a lot thicker. If you're saying that an iPad should come with these kits, go whine on the Apple feedback page. I'm sure they'd love to hear your feedback.
Either way, the fact remains that an iPad can do nearly everything on the list, and it is incorrect to claim otherwise merely because you have to buy an adapter that fans out pins on the dock connector to appropriate connectors.
I'm pretty sure the iPad and netbook demographics are completely different.
That depends on whether those people are buying netbooks for the form factor or for the price. It overlaps with some netbook buyers. Obviously not you.
...there's no VALUE to using an iPad as a development server or a graphics workstation.
Yes, there is. There's immense value to being able to work with a single file format. For art files, that format is PSD, like it or not. That doesn't mean I plan to do all my graphics work on an iPad, but if you use a stylus with it, it could be a really nice graphics tablet for certain types of work. That said, unless you can move files transparently back and forth to and from the device, you can't realistically use it for that apart from basic hobby use.
As for a development server, although there's no good reason to use it for one, there's also no good reason not to use it for one. As recently as ten years ago, I ran production servers on hardware that's less than a tenth the speed of an iPad. Either way, my point wasn't that the iPad should be able to do that stuff, but rather that if it is supposed to replace a netbook, then it would need to be able to do those things.
It doesn't bother me that an iPad can't do everything that a netbook can because what the iPad does, it does very well.
It's not a good match for my needs, but it would have been a great choice for my parents during their regular commute were it not for the 30-mile-long dead spot that covers half their county and most of the next.... And the complete lack of 3G for hundreds of miles. (No 3G coverage even in cities of 60,000+.)
The iPad is a consumer appliance. It's optimized for consuming information, rather than creating it.
And it is precisely that attitude that holds back real innovation. You see iPad for what it is. I see iPad for what it has the potential to be. Its potential is enormous---far greater than a mere consumer of information. But so long as people keep thinking of it as an overpriced display, it cannot reach its full potential.
You do realize that almost everything on your list is compete and utter trolling, right? An iPad can basically do almost everything on your list. It's not at all hard to come up with things that an iPad can't do, yet still you proceeded to dump a huge list of things that it can! Strawman argument, perhaps?
I can use its camera - iPad cannot
That's easy. Buy the iPad Camera Connector Kit and use a real camera. Oh, you meant a built-in camera on the Netbook? Those are universally crap anyway....:-)
I can install anything I want - iPad cannot
I can write a program and use it instantly - iPad cannot
Sure it can. You just need a $99 iPhone developer program membership. Or jailbreak it. Either way.
I can use various memory sticks - iPad cannot
Again, easy. Buy the iPad Camera Connector Kit. It supports memory sticks as long as they are capable of working correctly on low-power USB. In other words, if you can plug them into your keyboard, they should work. If they only work when you plug them into a computer, they won't.
I can dump my camera to it - iPad cannot
Apparently you haven't heard of the iPad Camera Connector Kit? Because that's exactly what it's for....
It has a real keyboard - iPad does not
I'm pretty sure you can connect any standard USB keyboard using the iPad Camera Connector Kit. At least that's what a number of folks have discovered.
Also, there's always the iPad Keyboard Dock that provides a standard laptop-sized keyboard (no keypad)
I can plug a real keyboard and mouse in now - iPad cannot
Again, I'm not sure it's officially supported, but the iPad Camera Connector Kit does this as long as you're connecting a standard HID-class-compliant keyboard and mouse. I'm starting to sound like a broken record here.
I can plug in an external monitor with hires - iPad cannot
Actually, yes it can. It can output up to 720p if you buy the iPad VGA adapter. Sure, it's not DVI, but if you wanted that, netbooks cannot, either.
I can use dual monitors on it - iPad cannot
On what netbook!?! I've never seen a netbook with anything more than a single VGA output. If you're going to compare an iPad to a netbook, at least compare it to a netbook that doesn't live in fantasy-land. Oh, you meant outputting something different than what's on your screen? Yup. iPad can. The app has to be coded to do so, but plenty of apps do this.
It has multiple USB ports - iPad does not
USB hub. $10 at Fry's. Seriously, you're not trying very hard, are you?
I run multi-boot OS's on in (Win7 / Linux) - iPad cannot
There are, no doubt folks working on that, too.... It's pretty much inevitable.
Slashdot Mirror
Try ALL Disney DVDs, not just animated ones. Some of them aren't even movie previews. I've seen outright ads for toys and other such crap on there. I'm paying them for the privilege of watching their movie, and then they're forcing me to watch that stuff? Not gonna happen. Stick the disc in, switch to my TiVo, watch a 30-minute show, switch back, and watch the movie.
No, I just read a lot.
Only hearts can pump the fluid through themselves, so I'd imagine it is somewhat limited, but I'm just guessing here.
That said, the first time I read this headline, I read "...can keep a head viable 10 days outside the body" and imagined something from Futurama....
There actually was one shuttle abort, STS-51F---one of Challenger's last missions. Also, STS-93 ended up in a lower orbit due to a fuel leak in one of the main engines. They didn't classify it as an ATO because they just plain ran out of oxidizer; no human explicitly hit an abort button. That same flight also had a major electrical short causing multiple main engine controller failures on that flight. Had the backup controllers also failed, you would doubtless have seen an ATLS on that flight.
And of course, there have only been two in-flight catastrophes, neither of which was detected in time to prevent it using an abort. The final Challenger and Columbia missions did not result in aborts because nobody hit the abort button, not because an abort wouldn't have been possible, given an early enough determination of the seriousness of the underlying failure event.
Either way, probably the biggest reason we don't see lots of shuttle disasters is that there are so many redundant systems and electronic safety checks on the shuttle. And it's a good thing, too, given the rate of failures. Most failures, statistically speaking, happen and are detected before the bird gets into the air. The number of launchpad aborts for the shuttle is significant. By my count, pad aborts occurred in almost one in ten launches (14 aborts out of 145 launch attempts). And that's not counting scrubbed launches due to weather or other non-mechanical causes. That's just the actual equipment failures occurring prior to SRB ignition....
As an aside, if airplanes had even one tenth the engine failure rate that the shuttles have seen, we'd see an airplane flight cancelled once per minute in the U.S. alone. Food for thought next time you start pondering whether the government should be in the business of designing payload launch systems without adult supervision.
Possibly. It certainly isn't completely out of the question. Either way, we'll never know if they could have done anything because they weren't given the opportunity to try.
First, the shuttle can land safely with certain tiles missing. It does that with regularity. There's a sizable safety margin everywhere but the leading edge of the wing. They could have spacewalked somebody to pry a tile or two off of less sensitive areas (e.g. near the OMS engines where a dozen tiles fell off during Columbia's maiden flight) and affixed them to the leading edge of the wing. They wouldn't have been as good as the higher-temperature tiles that are supposed to be there, but they would have been better than a gaping hole. Secondarily, they could have affixed additional metal plating over the affected area, ripped from consoles inside the orbiter and bent to suit, using whatever screws and drills that they had handy. You know, an Apollo-style repair.
Second, had the orbiter been reprogrammed by ground engineers, it might have been possible to land at an angle that takes some of the heat off of the damaged wing, either by favoring the opposite side during the hottest parts of reentry or by adjusting the angle of entry to be steeper, thus A. putting more heat on the bottom tiles and less on the leading edge, and B. dropping them more quickly into an area with greater air density against the wing that might provide better thermal transfer. I'm not sure if that last part would make up for the extra heating due to falling like a rock, but that's for somebody to actually model, not for me to guess about. In any case, they could have experimented with several computer models to find the reentry vector that gave them the best chance at survival instead of just using the standard reentry vector.
Third, in the absence of a way to patch things up, the U.S. military has thousands of launch vehicles that could readily be hacked up to carry a food and power payload into space on short notice. We call them ICBMs. They're theoretically fueled up and ready to launch at a moment's notice, and they have enough thrust to match the shuttle's speed. Rip out the warheads, carefully balance your payload components, fasten them in properly, put the thing up there, match speed with the shuttle, blow the payload in the general direction of the cargo bay, then restart the missile's engines to bring it back down to Earth. Don't get me wrong, hitting the target would make the missile defense tests look like a cakewalk by comparison, but it would also involve precisely known trajectories and two vehicles that can, at least to some extent, adjust course to meet each other.... And of course, there are enough missiles sitting in silos that if you screwed up with the first one, you could launch another one 90 minutes later on the next trip around. It would have been the single most expensive piece of foam in the history of the world, but then again, it was anyway, so....
Finally, there was a Soyuz ISS mission scheduled just two months later that could have been diverted as an emergency return vehicle. Pulled back a few weeks and combined with an ICBM for supply delivery, there is every possibility that they could have gotten the crew home safely, at which time the shuttle could have been landed on autopilot (optionally out in the middle of the ocean or desert if they were sufficiently worried) without risk to the crew.
Either way, regardless of whether it would or would not have been possible to save the crew, if they had put thousands of people to work looking for a solution, there's at least some possibility that they might have found one. As it stood, they didn't, so they couldn't.
No, minor failures happen with regularity. It's just selective memory. Nobody remembers the minor ones. The last minor shuttle equipment failure was just five weeks ago.
Yes, and no. There are certain classes of failures that are not preventable, sure, but in the case of both shuttle disasters, the accidents were entirely preventable. The first one was caused by NASA ignoring the thermal specifications for the SRBs and launching anyway. The second one was caused by NASA ignoring all the engineers screaming for inspection of the heat tiles and landing anyway. In both cases, the primary fault rests squarely on human error, and the secondary fault lies in design error---a single design error, in fact.
Both shuttle disasters were caused by a single design error. The shuttle is on the side of the stack instead of on top. Had the Challenger been on top of the stack, it would not been so severely damaged by the explosion, and the crew could have readily performed an abort-to-launch-site or similar. Had the Columbia been on top of the stack, it would not have gotten hit by flying chunks of foam. Sure, there's still the risk of bird impacts, but the risk is much, much lower than the near-inevitable foam impacts
There were, of course, other design flaws that contributed---insufficient metal shielding on the leading edge of the wing that makes the thermal tiles more critical than necessary, dubious lack of enough layers of O-rings in the SRBs, lack of an emergency capsule ejection system that activates automatically in the event of catastrophic failure, poor weight-to-surface-area ratio for landing in order to reduce drag during launch, the mere existence of foam on the outside of a tank instead of sealed inside it, and loads of other design decisions or oversights that in retrospect probably weren't the best idea, but ultimately it was the placement of the shuttle on the side of the stack that proved to be its undoing on two separate occasions. Such a fundamental mistake isn't easy to correct without throwing out the design completely and starting over.
I'd usually agree for important variables that get used throughout a function, but there's nothing more pointless than having to put loop iterators that only get used in two lines of code up at the top of the function. It just makes things so much harder to read than:
for (int i=3; i<maxval; i++) {...}
It's only a hallucination/delusion if he believes it. Otherwise, it's just Pseudologia fantastica.
You're right. I should have called it an internet hosting provider. Either way, the DMCA safe harbor very clearly defines "service provider" to include hosting providers:
Note that subsection (a) is the section "Transitory Digital Network Communications", which doesn't really apply to hosting providers in any meaningful way, hence "service providers" is defined differently for that section. For all other parts, hosting providers are service providers. In particular, the relevant safe harbor rule is 512(c), "Information Residing on Systems or Networks at Direction of Users," which is covered by definition (B).
Well, yes and no. According to Wikipedia, OFS wasn't the same project, though WinFS was based on the same concept.
Agreed. Hear that sound? That's the sound of the world's tiniest violin playing the saddest song in the world.
Sorry, Hollywood, but YouTube is precisely the kind of site that safe harbor exemptions are for. It is an ISP whose material is provided exclusively by its users. What, you thought that somehow a site shouldn't count because it is getting too big for you to bully? Waaaah. Cry me a freaking river.
It has occurred to many people. Indeed, that's precisely what extended attributes are for. However, it isn't feasible for a filesystem to provide multiple views into the data based on EA searches, as that would either require a very rigorously defined set of valid EA structures that can be searched or would require extensive application-specific knowledge.
Windows made an attempt at putting that sort of multi-view design into their filesystem with WinFS and half a decade later, it hasn't shipped. Apple tried the "leave it up to the importer" approach with Spotlight, and that actually worked reasonably well, though there are a few warts. Either way, the specific views that are most useful tend to be application-specific, so anything beyond keyword searches is problematic to implement in a general way (which is why nobody does it).
The thing is, apps like iPhoto already have abstractions that are better than filesystems. As long as the collection metaphor is hierarchical, they inherently scale at least as well as a filesystem does (because you can organize them in precisely the same way if you want). However, they also provide a much richer set of navigation than a filesystem can ever realistically provide by providing multiple views into the data.
For example, in iPhoto (personally, I use Lightroom, but the same principle applies), they have tagging, they have folders/collections, they have faces and places.... They have the ability to view things not just in a strict hierarchical view like a filesystem, but also in numerous logical views (e.g. show me photos taken in France with Pierre in them) that go way beyond what any filesystem view can feasibly provide. Thus, when it comes to larger data sets, filesystems generally get in the way far more than they help.
Note: I'm not saying iPhoto was always good about this. I do remember a time just a few versions back when scrolling through my iPhoto library took a small eternity. So I can certainly understand having trouble dealing with large libraries in iPhoto. However, that's not a flaw in the concept so much as a flaw in the implementation of that concept.
Not at all. One of the biggest flaws in computer UI design today is that there are lots of things that are not stored as files but are still basically indivisible units of data, whether they're mail messages or database records or... you name it. Because so many of these things are not, in fact, files, a purely file-based view is a fairly clumsy way to represent that content. For most users, they don't need to know or care whether data is in a file or a database record or an email message in an mbox file. Abstracting those details away from the user results in a better user experience with more ability to manage the actual content than a pure file-based interface can provide.
It's not like the filesystem in Ubuntu Light will cease to exist or will become inaccessible to power users. You'll just have to install tools to reach it. At least I assume that this is the case.
Other way around. AT&T works well in population centers. It is terrible in rural areas compared with Verizon. My parents regularly travel a 30 mile stretch of major 4-lane highway in TN with solid Verizon coverage and no AT&T coverage whatsoever. Here in California, when you get up into the hills, Verizon works, AT&T doesn't. And so on. AT&T's 2G coverage is anything but incredible. Their 3G coverage is simply an abomination, so people are so shellshocked that the 2G coverage seems great by comparison....
AT&T's 3G coverage works exclusively in major population centers; people have problems because AT&T's 3G coverage doesn't fail over to 2G very well. Disable 3G voice and you'll have many fewer dropped calls and a much better overall experience.
I would agree with you about the thin thing if you were talking about the MacBook Air, but an iPad really doesn't seem like it was intended to be your main computing device. It seems designed to be a device for carrying around content from your computer, working with it when you're on the go, then syncing it back to your computer when you return. When you're tethered to a desk somewhere, you'll probably use a traditional computer, not an iPad, so having connectors on the device is of marginal utility unless it's something you think you'll need with you when you're traveling. Having lots of connectors is antithetical to the entire concept of an iPad because hooking it up to lots of peripherals and tying yourself down to a desk is exactly the opposite of the way it was designed to be used.
Don't get me wrong, I'm not about to buy one. If my main machine were a desktop, I had no iPhone, and my portable needs were less esoteric, I could see it being a useful device, particularly for airplane flights. Unfortunately, the largest one could barely hold my photo library alone. And, of course, there's not an app for most of the things I use my laptop for anyway (writing software, composing large ensemble musical scores, designing websites, etc.). For my parents, on the other hand, were it not for AT&T's roughly 30-mile-long dead spot on their regular commute, I'd have already ordered one just so they could check email and browse the web while away from home. Different tools for different users.
Actually, it does. If you can't feasibly control who has access to information, then any open, distributed alternative to Facebook is going to have exactly zero success at preventing the same sort of data mining that they're complaining about Facebook allowing... unless, of course, you deliberately obfuscate the data so that it is hard to mine in any reasonable programmatic fashion, but doing so would also make it much less useful for your actual friends.
Don't get me wrong, Facebook shouldn't be playing fast and loose with privacy, and it's important to chew them out on principal every time they screw up, particularly as it applies to someone's public profile. However, I'm not convinced that any similar service that anybody else comes up with won't have at least as many similar problems. Coming up with good access policies is hard. Designing user-interfaces that expose those settings in a way your grandparents can understand is even harder.
First, those standards are pure algorithmic patents, and thus are probably invalid to begin with. It's just a question of somebody making the right arguments in a patent nullification suit.
Second, even if we assume the patents are valid, companies would not be "forced to give up their IP if they join a consortium." Companies would be free to declare IP that they care about to the consortium but not license it to them, at which point it could not be part of the standard, but they would retain full rights to it.
Look, I have exactly zero sympathy for any company that pushes their patented technology on the world as part of an open standard. Interchange standards should be just that---standards that can be used anywhere. Patents and cross-platform compatibility/ubiquity are fundamentally incompatible. We either do away with patents for file formats or those patents do away with compatibility. It's really that simple.
That said, adding patent laws to cover consortiums is still the wrong way to handle this. Consumers should have a fundamental right to any data of their creation, period, whether they are encoded using software from a consortium or from an individual corporation. What we need is a fair use law that applies to file format patents, e.g. something like this:
Boom. No more problems with interchange formats and patents, all with minimal collateral damage. It's basically the patent equivalent of the DMCA's compatibility exemption. For example, this protects things for which alternatives exist (e.g. novel ways to implement MPEG in hardware), but does so in a way that protects consumers' right to their own data and their right to use hardware that they paid for.
Which makes this whole story rather pointless, don't you think?
It could compress the entire screen update into a single byte and it still wouldn't matter. The latency of cellular communications would still kill you. Imagine typing a keystroke and waiting a quarter second for it to show up on your screen. And that's the round-trip time using ssh. Any screen-based protocol is going to add additional overhead on top of that.
Multi-hundred-millisecond latency figures are not at all unusual when you're dealing with cellular data, and if you're out of 3G territory, you're talking about whole seconds.
If you're saying that iPhone should have those connectors exposed, I would point out that USB and VGA connectors would make it a lot thicker. If you're saying that an iPad should come with these kits, go whine on the Apple feedback page. I'm sure they'd love to hear your feedback.
Either way, the fact remains that an iPad can do nearly everything on the list, and it is incorrect to claim otherwise merely because you have to buy an adapter that fans out pins on the dock connector to appropriate connectors.
That depends on whether those people are buying netbooks for the form factor or for the price. It overlaps with some netbook buyers. Obviously not you.
Yes, there is. There's immense value to being able to work with a single file format. For art files, that format is PSD, like it or not. That doesn't mean I plan to do all my graphics work on an iPad, but if you use a stylus with it, it could be a really nice graphics tablet for certain types of work. That said, unless you can move files transparently back and forth to and from the device, you can't realistically use it for that apart from basic hobby use.
As for a development server, although there's no good reason to use it for one, there's also no good reason not to use it for one. As recently as ten years ago, I ran production servers on hardware that's less than a tenth the speed of an iPad. Either way, my point wasn't that the iPad should be able to do that stuff, but rather that if it is supposed to replace a netbook, then it would need to be able to do those things.
It's not a good match for my needs, but it would have been a great choice for my parents during their regular commute were it not for the 30-mile-long dead spot that covers half their county and most of the next.... And the complete lack of 3G for hundreds of miles. (No 3G coverage even in cities of 60,000+.)
And it is precisely that attitude that holds back real innovation. You see iPad for what it is. I see iPad for what it has the potential to be. Its potential is enormous---far greater than a mere consumer of information. But so long as people keep thinking of it as an overpriced display, it cannot reach its full potential.
Wait. Correction. An iPad can deal with a HID-class keyboard. I don't think it would do anything with a mouse. :-)
You do realize that almost everything on your list is compete and utter trolling, right? An iPad can basically do almost everything on your list. It's not at all hard to come up with things that an iPad can't do, yet still you proceeded to dump a huge list of things that it can! Strawman argument, perhaps?
That's easy. Buy the iPad Camera Connector Kit and use a real camera. Oh, you meant a built-in camera on the Netbook? Those are universally crap anyway.... :-)
This is, of course, coming soon.
Sure it can. You just need a $99 iPhone developer program membership. Or jailbreak it. Either way.
Again, easy. Buy the iPad Camera Connector Kit. It supports memory sticks as long as they are capable of working correctly on low-power USB. In other words, if you can plug them into your keyboard, they should work. If they only work when you plug them into a computer, they won't.
Apparently you haven't heard of the iPad Camera Connector Kit? Because that's exactly what it's for....
I'm pretty sure you can connect any standard USB keyboard using the iPad Camera Connector Kit. At least that's what a number of folks have discovered.
Also, there's always the iPad Keyboard Dock that provides a standard laptop-sized keyboard (no keypad)
Again, I'm not sure it's officially supported, but the iPad Camera Connector Kit does this as long as you're connecting a standard HID-class-compliant keyboard and mouse. I'm starting to sound like a broken record here.
Actually, yes it can. It can output up to 720p if you buy the iPad VGA adapter. Sure, it's not DVI, but if you wanted that, netbooks cannot, either.
On what netbook!?! I've never seen a netbook with anything more than a single VGA output. If you're going to compare an iPad to a netbook, at least compare it to a netbook that doesn't live in fantasy-land. Oh, you meant outputting something different than what's on your screen? Yup. iPad can. The app has to be coded to do so, but plenty of apps do this.
USB hub. $10 at Fry's. Seriously, you're not trying very hard, are you?
There are, no doubt folks working on that, too.... It's pretty much inevitable.