By the same token, Newton's law of gravitation has clearly lost ALL predictive power, since it breaks down in the relativistic realm. So feel free not to get out of the way next time there's an anvil falling toward your head.
And yet if you get out of the way when there's a moon falling towards your head, they call you loony. Ba-dump bump.
Its no different to an undercover cop pretending to be a drug dealer and engaging in a drug deal so the hidden cops can spring forth and bust the bad guy.
Uh, it isn't even similar. In a sting operation, the operation ends when the person buys the drugs. The purpose is to get them to buy the drugs, but at that point, they arrest the person. In this case, they download the fake porn that tattles on them, but they also continue to download real porn from systems run by the government, upload new porn to systems run by the government, and trade porn in a marketplace run by the government over an extended period of time.
This is more closely equivalent to a rogue CIA agent infiltrating a drug cartel and then continuing to run it for a decade, growing and selling drugs, killing members of competing cartels, and advertising the availability of drugs in an effort to entice (entrap) people who otherwise might not have bought them, all under the pretense that while his or her organization deals drugs, he or she will also periodically slip the police a list of some of the people who are buying so that they can eventually do a sting operation and bust them....
This is so far outside the bounds of what should legitimately be legal in a free society that it is downright terrifying. There's a bright line between pretending to be part of a criminal organization and participating in the operation of a criminal organization, and from what I've read about this operation, I would say that they crossed that line by a large enough distance that they couldn't even see it in their rearview mirrors....
They've already done that. The remaining problem is that the user's passcode is entangled with a hardware key that can only be obtained by either using a custom OS (which Apple would have to agree to produce) or by destroying evidence—specifically, uncapping the CPU and then using an electron microscope on its guts, if memory serves.
If the backups were disabled by the user at that point, then that would be a show-stopper. If it simply stopped backing up, Apple could always bump up the quota on that account, and the device would automatically detect that it had space, and then it would back itself up.
Unfortunately, somebody changed the password on the iCloud account, making that approach impossible.
Nobody sane is going to encode unicode data as UTF-32. I mean sure, you can do it, but it is horribly inefficient. In fact, with the current UTF-8 standard, most Chinese, Japanese, and Korean characters are encoded using 3 bytes, with the exceptions being characters that are relatively uncommonly used. So unless your text consists entirely of special mathematical symbols, emoji, and certain less-common Chinese, Japanese, and Korean characters, you're wasting a lot of space by storing it in that way. And there is no situation in which UTF-32 is more efficient than UTF-8, now that the five- and six-byte characters are considered illegal byte sequences.
... unless you just enjoy using about 33% more space for your Chinese characters and 4x as much space for your ASCII characters, in which case, be my guest. But in general, the situations in which it makes sense to inspect the character at a particular index without looking through the preceding characters are few and far between.
Neither. You make it a varchar so that the space it uses varies based on the actual size of the data, and then you don't worry about the length of a particular character.
Actually, I think I understand why there's a difference. The OTA updates look like they run inside iOS, similar to the way minor OS X updates work, whereas the non-OTA updates seem to involve booting a from a separate installer root like major OS X upgrades work. So if that install DMG's OS contained a bug, it would affect the non-OTA updates during the upgrade process itself, but would have no impact on OTA updates.
That also means that you ought to have been able to get around the problem (albeit without being able to complete the upgrade process) by doing whatever the iOS equivalent of an NVRAM reset is. Does entering and leaving DFU mode have that effect, by any chance?
Also, to get the new build of 9.2.1, you apparently need to download it through iTunes, not over your iDevice's Wi-Fi connection [macrumors.com].
Supposedly (though I can't imagine why this would be the case) updating OTA to the earlier 9.2.1 build didn't cause the error to appear. So there may be no need to rev the OTA update.
With that said, I seem to recall that over-the-air updates require additional carrier approval because they're big and they can be DLed over the cellular network (depending on the carrier and the size of the update). So for an emergency update revision like this one, the OTA update could lag behind the normal release. If that recollection is correct, then the OTA update might get revved to match the normal iTunes downloadable update in a few days.
No more so than they could without changing the scanner. This change doesn't enable fake fingerprint scanners. It just lets you continue to use the device with a passcode as though the fingerprint scanner weren't there.
Apple does this quite frequently when they make a minor mistake in an update, silently releasing a new build with the same version number. What this signifies is that for 99.9999% of users, there's no functional difference between the two builds, so they didn't feel the need to turn a new build number and force everyone to update over something that affects probably a single-digit number of users.
By turning the build, they're ensuring that no new users encounter the problem going forwards, and providing a mechanism for the few affected users to get their devices up and running again (by manually reinstalling the current update). It wouldn't be an automated update anyway, because the devices won't let you use them, so for affected users, bumping the version number gains them nothing. And bumping the version number for everyone else would have resulted in everybody downloading a patch that they really don't need, and worse, would have caused anybody upgrading from 9.2 to (for example) 9.2.1a to get hit with a much larger combo update because they skipped the quirky 9.2.1 build.
When the next OS release happens, everybody will be back in sync, and until then, the differences are minor enough that they really don't matter for the most part.
With that said, if you want to know which version you are running, go to Settings > General > About, and look at the Version field. If it says 9.2.1 (13D15), you're running the older build. If it says 9.2.1 (13D20), you're running the newer build.
I mostly agree with what you're saying, and for a primary drive, you're absolutely right. But your original post didn't limit it to primary drives. That was my objection.
Don't get me wrong. I love my SSD's performance. It makes the laptop much, much faster than my pre-retina MBP. I don't want a spinning drive in lieu of an SSD. I want an internal spinning drive in addition to the SSD. Basically, the frequently used data (OS, apps, current projects) would live on the SSD, and bulk data (photo library, audio and video recordings, etc.) would live on the spinning drive, which would be spun down most of the time, but still available inside the computer for when the data is needed, without the clumsiness of an external drive.
External drives on portable machines are a royal pain in the backside. That's what I'm doing now as a workaround for the inadequate storage on my laptop. It means that every few months, I offload half a terabyte of photos from my laptop to that hard drive, and whenever I want to do anything with the older photos, I end up having to grab that hard drive. It isn't unusable by any means, but it isn't nearly as convenient as it would be if I could actually buy several terabytes of flash storage, or even a modern Mac laptop with a 3+ TB fusion drive setup. If I had that setup, my photos would go onto the spinning storage from day one. As for battery life, Lightroom uses so much battery even with an SSD that the extra battery hit probably wouldn't be a big deal. So when I'm running that app, my battery life would suck, but that's the price of working with RAW photos.:-)
What really bugs me is that my current laptop (a retina MacBook Pro) has the same disk capacity as I had in my laptop back in 2010 (and a third less storage than I had at various points in between). While my storage has been stuck at 1 TB, desktop hard drives went from 3 TB to 8 TB, and laptop hard drives went from 1 TB to 2 TB (or 3 TB if you have room for a 15mm drive).
I love the performance of SSD, but hate with a fiery passion the fact that capacity growth, at least in the laptop world, has ground to a halt for the past six years, while my actual storage needs are increasing at an ever faster pace. I just want my laptop to be able to hold my entire personal photo collection without having to connect external hard drives, and to not have to constantly delete things that I would prefer to keep around because I'm running out of disk space.....
With that said, for non-laptop purposes, I wouldn't expect power to be a significant piece of the equation, at least if you compare SSDs against modern enterprise spinning storage. An 8 TB HGST helium hard drive draws about 9.1W typically, which is only 80 kWh per year, or ~$28 per year in power costs at PG&E's top tier (which is some of the most expensive power in the country). Cooling is typically about a third of the total power consumption, so that's still under $40 per year difference for an 8 TB drive even if we assume that the flash drive uses zero power.
I mean yes, power factors in, but until your cost per drive of comparable capacity is within tens of dollars, it shouldn't be enough to erase the difference in hardware costs. Other factors, like capacity per unit of physical space are likely to be far more important, because that determines when you need to add an additional machine. So it will be interesting to see what happens with these 10 TB SSDs in that regard.
I think that Apple is just doing all they can to make buying a $700 phone enticing.
Except they aren't doing nearly all that they can. One big change they could make that would both improve sales and reduce prices is to add a micro-SD card slot to the iPhone, and then stop selling phones with larger capacity built-in.
Apple tacks on an extra $100 to move from 16 GB to 64 GB, and another $100 to move to 128 GB. I can buy a 64 GB micro-SD card for $30 from Amazon, or a 128 GB card for $45. So Apple is charging about 4x the normal market price for iPhone flash upgrades. And nearly everybody has to pay that first $100 simply because 16 GB is too small for most people, and because they're afraid that they will run out of space, and then will suffer until they replace their phone in three years.
This requirement that users know how much space they're going to use before they buy the phone is a real burden, both on users and on the flash manufacturing industry. It causes people to over-buy (most of those folks would be fine with 32 GB), and worse, causes people to re-buy the flash components every time they replace their devices. They could easily move their photos and other large content to their new phone by moving the flash card were it not soldered to the phone's main board.
If every iPhone came with only 16 GB and stored additional content on the micro-SD card, then there would be only one model of iPhone per form factor, which means less customer confusion, fewer supply chain issues, and fewer problems for the companies selling the products (customers not being able to find the particular model that they want).
The cost of the phone would also drop substantially within a couple of years. Right now, the flash manufacturers are barely able to meet demand. If people weren't over-buying flash initially and then re-buying flash every time they upgrade their phones, within a year or two, the demand would drop substantially, and flash prices would drop substantially, allowing Apple to reduce their manufacturing costs and, eventually, the price of the phones themselves.
That reduction in demand for flash would also have far-reaching impacts on the computer market. Instead of limiting their laptops to 1 TB because of lack of parts availability, Apple could sell laptops with multi-TB flash drives, and people would be able to afford them because of the reduced cost of flash.
I know the feeling. My understanding is that you're supposed to put everything "in the cloud" now. How that's supposed to work on location with no network connection is anyone's guess.
Yeah, that's a nice theory and all, but in practice, I neither trust cloud providers to be secure nor reliable. For example, last week, I was forced to move several hundred GB of public photos from a major ISP to my own server because they suddenly decided that unlimited storage for web pages wasn't unlimited, and wanted several times as much to store the photos in some idiotic object-oriented nightmare. And that's with an actual ISP hosting account. I can't imagine ever trusting any random cloud-based photo site when I'm publicly making available hundreds of GBs of photos. It is just laughable.
Not to mention that even if I could get past those problems, when you're talking about RAW files that are 25 MB apiece, over a 3 Mbps DSL connection, that comes out to more than a minute apiece just to retrieve them (and several times as long to upload them). The cloud is for tiny little bits of data that have low value. It is great if the only data you care about are your Facebook posts. It is tolerable for email messages. It's a joke for photos, and always will be until we have ubiquitous gigabit to the curb. Maybe even then.
For applications like this, a desktop unit is still somewhat necessary. On location, my laptop is a place where I can sort through the day's shots and get a leg up on post processing work. When I get home, the files get transferred to the desktop machine.
I'd love to go that route. Unfortunately, Apple's desktops are almost as bad as their laptops capacity-wise. So if I have to use an external hard drive anyway, I might as well just use it with my MacBook Pro. *sigh*
A valve opening outwards need not open in such a way that the explosion would blow it open. It could just as easily swing in and slide slightly upwards at the end of its swing, thus putting it underneath a lip that locks it into place. It is not the easiest design to build, mind, you, but it would be possible.
I can't imagine why you think a sliding valve would clog easily. Gate valves might not have ever been used in engines, but they're pretty common in other areas like plumbing, and AFAIK, they don't have any particular clogging problems in that domain.
And an iris is mechanically pretty simple. It isn't as simple as a valve that flaps, but it offers substantial advantages over such a valve, too, such as the ability to precisely control how far open it is. So the open question is whether those advantages would outweigh the complexity difference.
The company's terminology? You mean "secure enclave"? That's a fairly common, industry-standard term in the field of data security, though it is more commonly used in the context of networks rather than parts of a specific device. And Intel also uses that term. But if your hatred for Apple runs so deep that you can't stand to use Apple's terminology while talking about an Apple product, I suppose we could call it a Trusted Execution Environment....
Apparently, they are based on designs that they acquired from WD as part of WD's acquisition of HGST, but I wouldn't go so far as to say that they're rebadged Hitachi drives. After all, HGST is owned by WD, not Toshiba. So the answer is kind of convoluted.:-)
IMHO once these higher density SSD drives arrive, there will be little or nothing for me to recommend standard HDD, for any application. None. There is barely any reason to have spinning drives right now.
Except that we're nowhere near that point, at least from my perspective. A 1 TB laptop SSD starts at $240. $150 will buy you 3 TB of laptop-sized spinning storage (as long as your computer can handle a 15mm drive, otherwise 2 TB for $95). More importantly, no amount of money will buy you a laptop-sized 3 TB flash drive. (You can, however, buy them in a PCIe form factor for about $30,000.)
That last part is a major problem. Everybody is building flash drives to hit price points, ignoring the fact that flash drive capacity is still basically flailing around at 2010 hard drive capacities. In a decade, we've gone from flash drives being a decade behind hard drives capacity-wise to being six years behind. And we've gone from being 10x the price per TB to only 6x the price per TB. At this rate, the break-even point on capacity won't be until probably at least 2030, and maybe 2040, and the break-even point in cost-per terabyte is probably at least a decade away, too.
Of course, ten years ago, I'd have expected us to have passed the break-even point already. As best I can tell, the main reason we haven't is that the rise of smartphones has put a tremendous strain on the flash manufacturing industry, and they still haven't caught up. I'm starting to wonder if they ever will. If and when they actually do, then I would agree that spinning drives should go away. I'm not counting on that happening any time soon, though. I hope I'm wrong.
Unfortunately, the average user doesn't need much at all, but probably 5–10% of users want/need way more space than is currently available. So the options are either making radically different models with radically different capacity and annoying the high-end users with the price difference or using larger capacity everywhere so that economies of scale drive the price down for everyone. Hard drive manufacturers have always done the latter. For some reason, with flash, everybody is doing the former.
The worst part is that the difference between lightweight users and pro users is more than an order of magnitude. As a photographer, I desperately want to be able to buy a laptop with 5+ TB of capacity. That way, I could use the computer for 3–5 years before I fill it up completely. I could then clone everything onto an external drive, and start fresh with the next machine. Instead, my MacBook Pro has an appallingly inadequate 1 TB of flash storage, and I'm constantly having to micromanage things to keep my laptop in a usable state. I ran out of storage within the first six or seven months, and that was starting almost fresh, without cloning the contents of my previous laptop (which already had almost a terabyte of photos on it). I'm about to dump another ~700 GB of photos onto an external hard drive pretty soon, but that won't buy me more than a few months before I have to repeat the process all over again. And heaven help me if I actually want to write software on the thing. A single git checkout of WebKit takes something like forty or fifty gigabytes between the source and binaries. It is really easy to run out of space when you only have a terabyte to work with.
The current state of laptop storage is already way past annoying. Up until recently, each laptop I bought, I moved up to at least 4x the capacity through replacement hard drives by the time I stopped using it, then replaced it with a new model that had still greater capacity. Then, about 2010, all the improvements in capacity suddenly stopped. My black MacBook had a 1 TB hard drive in it by the time I replaced it. The largest drive I could get in the pre-retina MacBook Pro was also 1 TB, and the largest capacity I could get in the Retina MBP that replaced it was also 1 TB. I have literally not been able to upgrade my hard drive capacity for SIX YEARS. I could move back to the pre-retina laptop, swap out the optical drive, and end up with 4TB of capacity. I'm seriously considering it, because as nice as the retina screen is, I'd rather have enough storage to be able to function.
I'm seriously fed up with computers at this point. I haven't been this unhappy with the disk capacity of any Mac since the mid-1990s, and back then, I was unhappy because I couldn't afford more storage, not because it wasn't available. Now, I'm sitting here with disposable income, telling Apple "take my money!", and I still can't buy a new computer with enough storage to meet my needs at any price.
IMO, the computer industry is a great candidate for disruptive innovation. Too bad nobody seems interested in innovating anymore. Just saying.
Yeah, early Seagate perpendicular storage drives had serious problems, including (supposedly) some firmware bugs that made the problem worse. This was about the same time period where I lost five or six drives in the same year, all Seagate. I stopped using their hardware after that, and haven't looked back. Good to know that their reliability has gotten back to acceptable levels since then, but they should never have shipped that junk.
The things that stand out to me in that data are:
The Toshiba drives have such a wide confidence interval (presumably because of low device count) to make the data mostly useless.
Same goes for the newest HGST (8 TB) drive.
Same goes for the smallest WD (2 TB) drive, to some degree, but there's cause for concern there.
The 1.5 GB Seagate appears to be garbage, and should probably be recalled en masse.
The 3 TB WD line appears to be garbage and should probably be recalled en masse.
The rest of the WD line looks dubious and should be watched very carefully.
But the biggest takeaway is that HGST drives appear to be about an order of magnitude more reliable than any other manufacturer, on average, with the possible exception of Toshiba (for which the data is insufficient to render judgment), and ignoring HGST's 8 TB drive, for which there's still not enough data to judge its reliability.
It's hardware based encryption, where half of the key comes out of a value burned into the CPU during manufacturing (and not recorded anywhere) combined with a value burned into the "Secure Enclave" during manufacturing (and not recorded anywhere).
Not on the iPhone 5c. That was added in devices with the A7 CPU and later. The iPhone 5c uses an A6 and thus lacks a secure enclave. I'm pretty sure that you can also tell whether a device has a secure enclave by looking for the fingerprint reader. No fingerprint reader, no secure enclave.
That analysis would be correct if this were an iPhone 5s or later. It isn't. It's an iPhone 5c, which doesn't have a secure enclave, making the entire analysis incorrect.
That's true on the iPhone 5s and later. This is an iPhone 5c, which lacks a secure enclave. AFAIK, the key is stored on the external flash part, and I think the passcode is entangled with a hardware ID, but that can be obtained by the OS (by definition) without knowing the passcode. So all that is necessary to crack one of these is to remove the code that wipes the key upon a certain number of failed attempts, along with the code that progressively increases the time between failed attempts.
And really, you don't even need those changes, as long as you're willing to unsolder the flash parts, clone them, and resolder them, then repeat this process in the reverse direction every time it wipes the key or the delay gets longer than the time it takes to unsolder, reflash, and resolder the chips.
However, iOS is different. The Secure Enclave (SE) chip holds the key (literally), and no one, not even Apple, can cause the SE to disgorge said key.
This is an iPhone 5c. The secure enclave wasn't added until the iPhone 5s. Even if it existed, as I understand it, it is possible to read both the secure enclave's NAND flash and its hardware ID using electron microscopy.
Note, however, that obtaining that data would likely involve completely destroying the secure enclave (uncapping the chip and removing layers of wiring), which will render the device permanently unusable. This means that you would have to scrape the encrypted data off the external flash chips directly and use a separate computer for doing the actual decryption. More importantly, it means that the device would be a brick afterwards. For post-crime forensic purposes on a device that doesn't need to be returned to its owner, that's not necessarily a show-stopper, but it is probably impossible/infeasible to do this surreptitiously to a confiscated phone and then return the phone (functioning) to its owner.
Slashdot Mirror
And yet if you get out of the way when there's a moon falling towards your head, they call you loony. Ba-dump bump.
Thanks. I'll be here all night.
Uh, it isn't even similar. In a sting operation, the operation ends when the person buys the drugs. The purpose is to get them to buy the drugs, but at that point, they arrest the person. In this case, they download the fake porn that tattles on them, but they also continue to download real porn from systems run by the government, upload new porn to systems run by the government, and trade porn in a marketplace run by the government over an extended period of time.
This is more closely equivalent to a rogue CIA agent infiltrating a drug cartel and then continuing to run it for a decade, growing and selling drugs, killing members of competing cartels, and advertising the availability of drugs in an effort to entice (entrap) people who otherwise might not have bought them, all under the pretense that while his or her organization deals drugs, he or she will also periodically slip the police a list of some of the people who are buying so that they can eventually do a sting operation and bust them....
This is so far outside the bounds of what should legitimately be legal in a free society that it is downright terrifying. There's a bright line between pretending to be part of a criminal organization and participating in the operation of a criminal organization, and from what I've read about this operation, I would say that they crossed that line by a large enough distance that they couldn't even see it in their rearview mirrors....
They've already done that. The remaining problem is that the user's passcode is entangled with a hardware key that can only be obtained by either using a custom OS (which Apple would have to agree to produce) or by destroying evidence—specifically, uncapping the CPU and then using an electron microscope on its guts, if memory serves.
If the backups were disabled by the user at that point, then that would be a show-stopper. If it simply stopped backing up, Apple could always bump up the quota on that account, and the device would automatically detect that it had space, and then it would back itself up.
Unfortunately, somebody changed the password on the iCloud account, making that approach impossible.
Nobody sane is going to encode unicode data as UTF-32. I mean sure, you can do it, but it is horribly inefficient. In fact, with the current UTF-8 standard, most Chinese, Japanese, and Korean characters are encoded using 3 bytes, with the exceptions being characters that are relatively uncommonly used. So unless your text consists entirely of special mathematical symbols, emoji, and certain less-common Chinese, Japanese, and Korean characters, you're wasting a lot of space by storing it in that way. And there is no situation in which UTF-32 is more efficient than UTF-8, now that the five- and six-byte characters are considered illegal byte sequences.
... unless you just enjoy using about 33% more space for your Chinese characters and 4x as much space for your ASCII characters, in which case, be my guest. But in general, the situations in which it makes sense to inspect the character at a particular index without looking through the preceding characters are few and far between.
Only in MS SQL (and in Ingres if you're using UTF-16 for some reason). In MySQL, PostGres, and Ingres, the varchar data type handles UTF-8 just fine.
Neither. You make it a varchar so that the space it uses varies based on the actual size of the data, and then you don't worry about the length of a particular character.
Actually, I think I understand why there's a difference. The OTA updates look like they run inside iOS, similar to the way minor OS X updates work, whereas the non-OTA updates seem to involve booting a from a separate installer root like major OS X upgrades work. So if that install DMG's OS contained a bug, it would affect the non-OTA updates during the upgrade process itself, but would have no impact on OTA updates.
That also means that you ought to have been able to get around the problem (albeit without being able to complete the upgrade process) by doing whatever the iOS equivalent of an NVRAM reset is. Does entering and leaving DFU mode have that effect, by any chance?
Supposedly (though I can't imagine why this would be the case) updating OTA to the earlier 9.2.1 build didn't cause the error to appear. So there may be no need to rev the OTA update.
With that said, I seem to recall that over-the-air updates require additional carrier approval because they're big and they can be DLed over the cellular network (depending on the carrier and the size of the update). So for an emergency update revision like this one, the OTA update could lag behind the normal release. If that recollection is correct, then the OTA update might get revved to match the normal iTunes downloadable update in a few days.
No more so than they could without changing the scanner. This change doesn't enable fake fingerprint scanners. It just lets you continue to use the device with a passcode as though the fingerprint scanner weren't there.
Apple does this quite frequently when they make a minor mistake in an update, silently releasing a new build with the same version number. What this signifies is that for 99.9999% of users, there's no functional difference between the two builds, so they didn't feel the need to turn a new build number and force everyone to update over something that affects probably a single-digit number of users.
By turning the build, they're ensuring that no new users encounter the problem going forwards, and providing a mechanism for the few affected users to get their devices up and running again (by manually reinstalling the current update). It wouldn't be an automated update anyway, because the devices won't let you use them, so for affected users, bumping the version number gains them nothing. And bumping the version number for everyone else would have resulted in everybody downloading a patch that they really don't need, and worse, would have caused anybody upgrading from 9.2 to (for example) 9.2.1a to get hit with a much larger combo update because they skipped the quirky 9.2.1 build.
When the next OS release happens, everybody will be back in sync, and until then, the differences are minor enough that they really don't matter for the most part.
With that said, if you want to know which version you are running, go to Settings > General > About, and look at the Version field. If it says 9.2.1 (13D15), you're running the older build. If it says 9.2.1 (13D20), you're running the newer build.
I mostly agree with what you're saying, and for a primary drive, you're absolutely right. But your original post didn't limit it to primary drives. That was my objection.
Don't get me wrong. I love my SSD's performance. It makes the laptop much, much faster than my pre-retina MBP. I don't want a spinning drive in lieu of an SSD. I want an internal spinning drive in addition to the SSD. Basically, the frequently used data (OS, apps, current projects) would live on the SSD, and bulk data (photo library, audio and video recordings, etc.) would live on the spinning drive, which would be spun down most of the time, but still available inside the computer for when the data is needed, without the clumsiness of an external drive.
External drives on portable machines are a royal pain in the backside. That's what I'm doing now as a workaround for the inadequate storage on my laptop. It means that every few months, I offload half a terabyte of photos from my laptop to that hard drive, and whenever I want to do anything with the older photos, I end up having to grab that hard drive. It isn't unusable by any means, but it isn't nearly as convenient as it would be if I could actually buy several terabytes of flash storage, or even a modern Mac laptop with a 3+ TB fusion drive setup. If I had that setup, my photos would go onto the spinning storage from day one. As for battery life, Lightroom uses so much battery even with an SSD that the extra battery hit probably wouldn't be a big deal. So when I'm running that app, my battery life would suck, but that's the price of working with RAW photos. :-)
What really bugs me is that my current laptop (a retina MacBook Pro) has the same disk capacity as I had in my laptop back in 2010 (and a third less storage than I had at various points in between). While my storage has been stuck at 1 TB, desktop hard drives went from 3 TB to 8 TB, and laptop hard drives went from 1 TB to 2 TB (or 3 TB if you have room for a 15mm drive).
I love the performance of SSD, but hate with a fiery passion the fact that capacity growth, at least in the laptop world, has ground to a halt for the past six years, while my actual storage needs are increasing at an ever faster pace. I just want my laptop to be able to hold my entire personal photo collection without having to connect external hard drives, and to not have to constantly delete things that I would prefer to keep around because I'm running out of disk space.....
With that said, for non-laptop purposes, I wouldn't expect power to be a significant piece of the equation, at least if you compare SSDs against modern enterprise spinning storage. An 8 TB HGST helium hard drive draws about 9.1W typically, which is only 80 kWh per year, or ~$28 per year in power costs at PG&E's top tier (which is some of the most expensive power in the country). Cooling is typically about a third of the total power consumption, so that's still under $40 per year difference for an 8 TB drive even if we assume that the flash drive uses zero power.
I mean yes, power factors in, but until your cost per drive of comparable capacity is within tens of dollars, it shouldn't be enough to erase the difference in hardware costs. Other factors, like capacity per unit of physical space are likely to be far more important, because that determines when you need to add an additional machine. So it will be interesting to see what happens with these 10 TB SSDs in that regard.
Except they aren't doing nearly all that they can. One big change they could make that would both improve sales and reduce prices is to add a micro-SD card slot to the iPhone, and then stop selling phones with larger capacity built-in.
Apple tacks on an extra $100 to move from 16 GB to 64 GB, and another $100 to move to 128 GB. I can buy a 64 GB micro-SD card for $30 from Amazon, or a 128 GB card for $45. So Apple is charging about 4x the normal market price for iPhone flash upgrades. And nearly everybody has to pay that first $100 simply because 16 GB is too small for most people, and because they're afraid that they will run out of space, and then will suffer until they replace their phone in three years.
This requirement that users know how much space they're going to use before they buy the phone is a real burden, both on users and on the flash manufacturing industry. It causes people to over-buy (most of those folks would be fine with 32 GB), and worse, causes people to re-buy the flash components every time they replace their devices. They could easily move their photos and other large content to their new phone by moving the flash card were it not soldered to the phone's main board.
If every iPhone came with only 16 GB and stored additional content on the micro-SD card, then there would be only one model of iPhone per form factor, which means less customer confusion, fewer supply chain issues, and fewer problems for the companies selling the products (customers not being able to find the particular model that they want).
The cost of the phone would also drop substantially within a couple of years. Right now, the flash manufacturers are barely able to meet demand. If people weren't over-buying flash initially and then re-buying flash every time they upgrade their phones, within a year or two, the demand would drop substantially, and flash prices would drop substantially, allowing Apple to reduce their manufacturing costs and, eventually, the price of the phones themselves.
That reduction in demand for flash would also have far-reaching impacts on the computer market. Instead of limiting their laptops to 1 TB because of lack of parts availability, Apple could sell laptops with multi-TB flash drives, and people would be able to afford them because of the reduced cost of flash.
Yeah, that's a nice theory and all, but in practice, I neither trust cloud providers to be secure nor reliable. For example, last week, I was forced to move several hundred GB of public photos from a major ISP to my own server because they suddenly decided that unlimited storage for web pages wasn't unlimited, and wanted several times as much to store the photos in some idiotic object-oriented nightmare. And that's with an actual ISP hosting account. I can't imagine ever trusting any random cloud-based photo site when I'm publicly making available hundreds of GBs of photos. It is just laughable.
Not to mention that even if I could get past those problems, when you're talking about RAW files that are 25 MB apiece, over a 3 Mbps DSL connection, that comes out to more than a minute apiece just to retrieve them (and several times as long to upload them). The cloud is for tiny little bits of data that have low value. It is great if the only data you care about are your Facebook posts. It is tolerable for email messages. It's a joke for photos, and always will be until we have ubiquitous gigabit to the curb. Maybe even then.
I'd love to go that route. Unfortunately, Apple's desktops are almost as bad as their laptops capacity-wise. So if I have to use an external hard drive anyway, I might as well just use it with my MacBook Pro. *sigh*
A valve opening outwards need not open in such a way that the explosion would blow it open. It could just as easily swing in and slide slightly upwards at the end of its swing, thus putting it underneath a lip that locks it into place. It is not the easiest design to build, mind, you, but it would be possible.
I can't imagine why you think a sliding valve would clog easily. Gate valves might not have ever been used in engines, but they're pretty common in other areas like plumbing, and AFAIK, they don't have any particular clogging problems in that domain.
And an iris is mechanically pretty simple. It isn't as simple as a valve that flaps, but it offers substantial advantages over such a valve, too, such as the ability to precisely control how far open it is. So the open question is whether those advantages would outweigh the complexity difference.
The company's terminology? You mean "secure enclave"? That's a fairly common, industry-standard term in the field of data security, though it is more commonly used in the context of networks rather than parts of a specific device. And Intel also uses that term. But if your hatred for Apple runs so deep that you can't stand to use Apple's terminology while talking about an Apple product, I suppose we could call it a Trusted Execution Environment....
Apparently, they are based on designs that they acquired from WD as part of WD's acquisition of HGST, but I wouldn't go so far as to say that they're rebadged Hitachi drives. After all, HGST is owned by WD, not Toshiba. So the answer is kind of convoluted. :-)
Except that we're nowhere near that point, at least from my perspective. A 1 TB laptop SSD starts at $240. $150 will buy you 3 TB of laptop-sized spinning storage (as long as your computer can handle a 15mm drive, otherwise 2 TB for $95). More importantly, no amount of money will buy you a laptop-sized 3 TB flash drive. (You can, however, buy them in a PCIe form factor for about $30,000.)
That last part is a major problem. Everybody is building flash drives to hit price points, ignoring the fact that flash drive capacity is still basically flailing around at 2010 hard drive capacities. In a decade, we've gone from flash drives being a decade behind hard drives capacity-wise to being six years behind. And we've gone from being 10x the price per TB to only 6x the price per TB. At this rate, the break-even point on capacity won't be until probably at least 2030, and maybe 2040, and the break-even point in cost-per terabyte is probably at least a decade away, too.
Of course, ten years ago, I'd have expected us to have passed the break-even point already. As best I can tell, the main reason we haven't is that the rise of smartphones has put a tremendous strain on the flash manufacturing industry, and they still haven't caught up. I'm starting to wonder if they ever will. If and when they actually do, then I would agree that spinning drives should go away. I'm not counting on that happening any time soon, though. I hope I'm wrong.
Unfortunately, the average user doesn't need much at all, but probably 5–10% of users want/need way more space than is currently available. So the options are either making radically different models with radically different capacity and annoying the high-end users with the price difference or using larger capacity everywhere so that economies of scale drive the price down for everyone. Hard drive manufacturers have always done the latter. For some reason, with flash, everybody is doing the former.
The worst part is that the difference between lightweight users and pro users is more than an order of magnitude. As a photographer, I desperately want to be able to buy a laptop with 5+ TB of capacity. That way, I could use the computer for 3–5 years before I fill it up completely. I could then clone everything onto an external drive, and start fresh with the next machine. Instead, my MacBook Pro has an appallingly inadequate 1 TB of flash storage, and I'm constantly having to micromanage things to keep my laptop in a usable state. I ran out of storage within the first six or seven months, and that was starting almost fresh, without cloning the contents of my previous laptop (which already had almost a terabyte of photos on it). I'm about to dump another ~700 GB of photos onto an external hard drive pretty soon, but that won't buy me more than a few months before I have to repeat the process all over again. And heaven help me if I actually want to write software on the thing. A single git checkout of WebKit takes something like forty or fifty gigabytes between the source and binaries. It is really easy to run out of space when you only have a terabyte to work with.
The current state of laptop storage is already way past annoying. Up until recently, each laptop I bought, I moved up to at least 4x the capacity through replacement hard drives by the time I stopped using it, then replaced it with a new model that had still greater capacity. Then, about 2010, all the improvements in capacity suddenly stopped. My black MacBook had a 1 TB hard drive in it by the time I replaced it. The largest drive I could get in the pre-retina MacBook Pro was also 1 TB, and the largest capacity I could get in the Retina MBP that replaced it was also 1 TB. I have literally not been able to upgrade my hard drive capacity for SIX YEARS. I could move back to the pre-retina laptop, swap out the optical drive, and end up with 4TB of capacity. I'm seriously considering it, because as nice as the retina screen is, I'd rather have enough storage to be able to function.
I'm seriously fed up with computers at this point. I haven't been this unhappy with the disk capacity of any Mac since the mid-1990s, and back then, I was unhappy because I couldn't afford more storage, not because it wasn't available. Now, I'm sitting here with disposable income, telling Apple "take my money!", and I still can't buy a new computer with enough storage to meet my needs at any price.
IMO, the computer industry is a great candidate for disruptive innovation. Too bad nobody seems interested in innovating anymore. Just saying.
Yeah, early Seagate perpendicular storage drives had serious problems, including (supposedly) some firmware bugs that made the problem worse. This was about the same time period where I lost five or six drives in the same year, all Seagate. I stopped using their hardware after that, and haven't looked back. Good to know that their reliability has gotten back to acceptable levels since then, but they should never have shipped that junk.
The things that stand out to me in that data are:
But the biggest takeaway is that HGST drives appear to be about an order of magnitude more reliable than any other manufacturer, on average, with the possible exception of Toshiba (for which the data is insufficient to render judgment), and ignoring HGST's 8 TB drive, for which there's still not enough data to judge its reliability.
The iPhone 5c doesn't actually have a secure enclave, so yes, you can safely assume that the failed count is not stored in one.
Not on the iPhone 5c. That was added in devices with the A7 CPU and later. The iPhone 5c uses an A6 and thus lacks a secure enclave. I'm pretty sure that you can also tell whether a device has a secure enclave by looking for the fingerprint reader. No fingerprint reader, no secure enclave.
That analysis would be correct if this were an iPhone 5s or later. It isn't. It's an iPhone 5c, which doesn't have a secure enclave, making the entire analysis incorrect.
That's true on the iPhone 5s and later. This is an iPhone 5c, which lacks a secure enclave. AFAIK, the key is stored on the external flash part, and I think the passcode is entangled with a hardware ID, but that can be obtained by the OS (by definition) without knowing the passcode. So all that is necessary to crack one of these is to remove the code that wipes the key upon a certain number of failed attempts, along with the code that progressively increases the time between failed attempts.
And really, you don't even need those changes, as long as you're willing to unsolder the flash parts, clone them, and resolder them, then repeat this process in the reverse direction every time it wipes the key or the delay gets longer than the time it takes to unsolder, reflash, and resolder the chips.
This is an iPhone 5c. The secure enclave wasn't added until the iPhone 5s. Even if it existed, as I understand it, it is possible to read both the secure enclave's NAND flash and its hardware ID using electron microscopy.
Note, however, that obtaining that data would likely involve completely destroying the secure enclave (uncapping the chip and removing layers of wiring), which will render the device permanently unusable. This means that you would have to scrape the encrypted data off the external flash chips directly and use a separate computer for doing the actual decryption. More importantly, it means that the device would be a brick afterwards. For post-crime forensic purposes on a device that doesn't need to be returned to its owner, that's not necessarily a show-stopper, but it is probably impossible/infeasible to do this surreptitiously to a confiscated phone and then return the phone (functioning) to its owner.