Jailbreaking isn't really official, and it is becoming harder and harder for the Dev Teams to find a usable JB. For example, it took about two months for them to make jailbreakme.com when the iPhone 4 came out, then when the Greenpois0n exploit was found, that pretty much allowed any iPhone 4 (not 4s) to be jailbroken either tethered or semi-tethered. The 4s is a different beast altogether, and the gymnastics required to JB that device took a lot of effort.
It would be nice if Apple offered low level access to their devices, but realistically, it only is going to become longer and longer before a successful JB is achieved on future phones.
Plus, JB-ing isn't just getting something to run as root like on Android. It is installing a complete user environment, moving apps around, and many other workarounds that might change with each OS revision.
I've caught some apps by looking at the permissions asked for. For example, a game that asks for everything under the sun.
Then, when you look at the reviews, they are short and pithy, or consist of text like "App work[sic] great!".
That is when you know it isn't something you want on your device.
What is ironic is that I've yet to encounter an app that would request root permissions via su that isn't supposed to. I'm pretty sure it is because I refuse to install any app that requests irrelevant permissions, but it is sort of surprising that the baddies have not taken the tack of popping up a su prompt randomly. Maybe because users who root their phones would get very suspicious very fast.
I'm sure users who install it will find out it isn't up to snuff when all their contacts first get barraged by text messages from that device, then spam outlets as the contact data gets imported into the spammer databases.
Don't forget that anything in international waters that isn't under a country's flag is fair salvage for anyone who comes up to it. Even if it under a flag, the rule of the sea is often might makes right, so a data center on a barge might just get the flag of a white star on a blue background because the guys on the Zodiac boats with the AK-47s say so.
Then, if the user so chooses to set foot into Mordor, there can be a tier of apps that are downloadable almost immediately, and pulled if people justifiably report it as malicious.
This type of system has worked on jailbroken phones, where the App store serves one tier, and Cydia serves another. Since it takes a little bit of effort to JB an iPhone, generally someone is clued enough to be able to watch out for Trojans.
What this is protecting against, is arguably the biggest security hole of all; the user. Most smartphone users are not anywhere as savvy as a/. reader. The casual user will see an app that might offer "cool smilies", install it by reflex, and go on their merry way. On iOS, the damage a user can do is limited [1]. On Android, it is fairly easy to find apps that are malicious, and where a competent person would not install a fleshlight app that asks for full phone, GPS, contact, photos, and filesystem access (or even a prompt for a su), an inexperienced user will just click "install" nontheless, then scream that Android is insecure when they get bitten. iOS is designed to keep this from happening. Only beta code, Cydia apps, and enterprise apps are not coming through Apple's gateway. It is almost certain that the worst an iOS app can do is lighten the user's pocketbook due to its cost, or the cost of in-app transactions.
This isn't exactly the "dancing bunnies" security hole, but protecting the ignorant user from themselves is the difference between a platform having a rep as secure versus easily compromised.
I like both worlds. Have some barrier so a user doesn't exit the managed tier without a deliberate decision, then if they choose to, allow them to do what they want. This keeps the novices from footshooting while allowing people with a clue to use their device to the fullest.
[1]: Assuming the user doesn't JB, but generally if someone is clued enough to jailbreak, they will either know what they are doing, or end up having a clued friend DFU restoring their device and not do it again.
Doesn't BlackberryOS do this? Apple really should take a page from that PlayBook and have permissions for apps accessing the phone or text items, contacts, music, and photos. It wouldn't add that much clutter, and it would add a lot of protection.
On the cheap, maybe Apple should see about licensing the Cydia app Protect My Privacy and building that into the OS. That way, if an app does go and access stuff it shouldn't, it will get results, although it will just get a random UDID and garbage in the fields.
Contact list protection would help immensely if an app glitches like Facebook's and starts overwriting or appending contact fields without permission.
Once malware gets rooted out and Apple slams the banhammer down, it is a lot harder for a shady developer to get around closed accounts than on the Google Marketplace. This by itself keeps the bad guys on notice.
That is the main security mechanism of iOS which keeps the bad stuff at bay: As soon as Apple gets wind of something malicious or violating the rules, it gets tossed out immediately. The same action doesn't get repeated.
Now, once an app does get past the gatekeeper, it has a lot of room to play because only locations and alerts are granted/denied by the user. So, in theory, an app can copy pictures and contacts off, as well as send text messages all it wants. However, if users find something doing this, Apple squashes it.
Since Apple's reputation is on the line for security, the strong gatekeeper has shown that it is more secure than the weak gatekeeper/strong OS security of the Android ecosystem. Google needs to get with it and start having a tier of the Marketplace that requires apps to be actively approved, similar to what Amazon does.
On Android, I would recommend LBE Privacy Guard (requires root) to ensure FB keeps its sticky fingers out of the contacts.
On iOS, it requires jailbreaking, but there is a Cydia app called PMP or Protect My Privacy which will allow FB to have what it thinks is a contact list... when in reality, it is getting randomly generated garbage.
Either way, FB gets nothing that it shouldn't have if you know what you are doing.
A lot of the RF meters also have the capability of being shut off from remote. Having a wardriver see your electricity usage is one thing. Having someone be able to shut off electricity to people on a Friday before finals is something completely different.
I know Texas is a laughingstock, but from what I have experienced, the power grid here has been pretty reliable. There have not been any major blackouts other than one caused by a transformer oil fire about a decade ago, and the grid did remain up for the most part. The biggest danger here are ice storms. However, those are few and far between, especially with the warmer and drier climate.
The ironic thing is that most Texans also tend to have portable generators, so if power does go out, the generator gets dragged out of the garage, fired up [1], plugged into the transfer switch receptacle [2], and life goes on.
[1]: This is if one remembers to keep the fuel fresh and fire up the generator every few weeks. Otherwise, the carb bowl will be full of varnish and the jet will be clogged.
[2]: People can go to prison for criminal negligence if they use a suicide cord to plug their generator in (which backfeeds and can kill a lineman), so generally an automatic transfer switch plus a receptacle for the generator's cord gets installed.
There are also geographic issues as well. East of Houston to Florida is swamp. Good luck burying anything there. There is a reason why Louisiana is known for its elaborate crypts and morgues. There is just no way to bury the dead, so they have to remain above ground.
The US is a very disparate country. Some places the cities are as safe as Europe (Seattle, Portland, and chunks of NYC.) Other places, not so much. One of the main reason why some cities are burying cables now is because overhead lines tend to be a target for metal thieves so they can get their next meth fix.
To be specific, 6 volt, deep-cycle golf cart batteries. These used in pairs are a staple of RV boondocking because they hold a good amount of amp-hours.
I like using AGM batteries because they don't need water added, and can take a lot more incoming amperage than regular flooded cells can. However, they are more expensive.
Don't forget, even with the best solar system and wind turbine, there will be cases where one will need a generator. The staple for boondockers is usually two Honda 2000 watt inverters run with a parallel kit.
Don't cheap out on the generator, or else, and you might find yourself sans charger or refrigerator because the control board got fried. This is why I always recommend a Honda or Yamaha, since both will almost always outlive most RVs.
I'd just be happy with a battery that 1/2 to 1/10 of the energy volume density of gasoline, and can be scaled up.
That way, the Otto cycle engines can be chucked for electric motors which don't have energy loss due to exhaust or needless heat.
For RVs, it would allow for the rig to be completely electric. No loud generators, just use a high capacity inverter that can handle the 60 or so locked rotor amps from an A/C, and that is that. Then when you get to home or storage, plug the RV into shore power to trickle charge.
Technically, there are levels of currency that one can use.
At the lowest tier, where there is no civilization structure up enough to support validating the purity of gold, the only currency that would work would be ammo, since it is relatively small (assuming small arms and not tank shells), and useful.
The next tier up, where there is some infrastructure in place allowing for items that have less practical usefulness, but more fungible (one troy ounce of gold is one troy ounce of gold, assuming similar purity.)
Some tiers up from that would be a cryptographic based currency. Chaum and Timothy C. May have discussed that at length on the cypherpunks mailing list in the mid 1990s. In fact, Tim May wrote a long document called the Cyphernomicon which touches on a lot of points.
Yes, there are ways to make an anonymous currency. However, realistically, every government out there will step in to stop it. Bitcoins are not anonymous. I'm sure if someone kept doing over USD 10,000 worth of transactions into and out of the US in BitCoins, FinCEN will start taking notice and investigating.
An anonymous currency is a double-edged sword. On one hand, it keeps the guys who love to track every move a consumer makes out of the equation. On the other hand, it makes for a perfect way to exploit/extort/blackmail other people, similar to how e-Gold was used for payment when an intruder encrypted a victim's hard disk and demanded payment.
Even more ominous, it gives FB a lot of control over things it should have no business with.
Disclaimer: This is theory.
1: FB could in theory allow others access to that account, and thus allow people who had the cash to have access to any FB-derived authentication.
2: Someone compromising FB wouldn't just have that site, but a lot of others as well. Part of security is packing your own parachute. OpenID is good because it is distributed.
3: Relying on FB which has -zero- SLAs is about as bright as relying on using a high school student's cast off iBook on his Internet connection as the company critical E-mail server.
4: FB gains a lot of power. They boot someone, that person not just loses access to FB and the games on there, but Spotify and many other places. If push came to shove, people who FB chose to toss out could find themselves losing a lot of access to services.
I'd say people are becoming forced to in some ways.
For example authentication. A number of websites are using FB for their authentication mechanism, with no way to just create an account outside of that. This seems to be increasing because it seems to be easy to implement.
Right now, if a site demands FB authentication, they are avoidable. For example, Spotify. Easy change -- drop them, pick up Rdio which has a better selection anyway. However, I dread a time where if one wants to pay a phone bill online, order a pizza, or log onto a MMO, one must cough up a FB ID or else go elsewhere.
When recommending Macs to people [1], I get the virus question asked all the time. I try to clarify the difference between malware types.
Viruses are not really a viable infection vector on Macs because people don't share executables, and Word macro stuff is pretty much stomped out.
Trojan horses are a major threat. Especially when someone wants a pirated copy of something and finds that their copy of iWork has more than just an office suite in the.DMG file. Executable signing helps here, but the Dancing Bunnies security hole is still there. However, the biggest thing that closes this hole is to have a repository (which Apple has done), and dissuade users from installing from anywhere but that unless they are 100% sure of the source. Locking down the entry vector for applications with brutal gatekeepers is why iOS has such an excellent perceived security record while in reality, its whole security model pretty much depends on the BSD jail() command.
Probably the biggest threat these days are holes in Web browsers and add-ons. Even iOS had these, although it was used for extreme good with the jailbreakme.com website. However, on Windows, generally a hole in the Web browser will be used to insert malware at the user (or possibly system) level. The best protection here is having the OS sandbox with all writes redirected, or even run the browser in a virtual machine that rolls back all changes except to files stored in the browser's area.
[1]: For an individual, Apple has the best CS in town, unless someone buys a business line (Optiplex for example) with the premium service from Dell, HP, or another PC vendor. For someone who uses their personal machine for their livelihood, it isn't bad to have a one-stop-shop for the OS, hardware, and applications. That way, one doesn't spend time playing the "sorry, that's not our department" game. Apple's hardware is on par with other vendors, but they have a lot better support for individuals than anyone else out there.
With the requirement to store every single thing users do, it might be a good time to invest in EMC because it is going to require an enterprise (VNX level) SAN to record all what is going on, as well as the licenses for features like deduplication (since a bunch of troll posts are usually alike, the SAN can store one copy, and pointers to the others.)
As a user in the UK, I'd be looking to find the best always-on VPN service, one in the country (since some services are country-locked), and one situated somewhere less repressive but close by, network-wise, perhaps Sweden or Norway.
I'm sure that is going to be coming to the US really soon (if it isn't already present), so guess it is time to find a Canadian VPN provider.
I might be feeding the trolls here, but I see the "encrypt your data" thing often, like it is a magical switch I can flip and call it done.
The gotcha is the key management. For example, tape encryption. For a lot of companies, one can get adequate security by setting a long passphrase on the LTO-4 newer drives and library and using that for all tapes, perhaps changing it yearly with the old passphrases remaining for read compatibility with previous year media. However, some companies want to sell you a key device which makes every single tape have a different key. OK... what happens if the key device gets destroyed? Unless one constantly dumps the data out of it somehow, pretty much the tapes are unreadable. Of course, the key device vendor's solution is another key device and replication.
Or, perhaps take something as simple as BitLocker in the enterprise. This is as close as one can find to flipping on encryption. However, if the TPM on one of the machines gets zapped, the data is gone unless there is some recovery method, and that requires some type of infrastructure, be it storing recovery keys in Active Directory, using data recovery agent keys, or just having IT encrypt the volume, print out the key (so it can't be erased), and save the BEK file to some secure storage somewhere.
First, most machines don't possess one, so DRM can't really assume it is there.
Second, it is shipped off and disabled. A user has to explicitly flip the TPM on in BIOS setup, then allow the OS to take ownership.
Of course, this technology is a double-edged sword. Look at the PS3 is an example.
However, on laptops, it provides an additional security boost, especially with a full disk encryption utility like BitLocker. The enhanced security it provides (allowing the OS to boot without a password needed, as well as protecting against "evil maid" attacks) is a help.
I'm curious about something with the pentalobe screws. Other than being used for tamper-resistance, I wonder if they allow for additional torque to be applied without camming out (a la Robertson or Torx.) If this were the case, it would make sense for Apple to use it.
I forgot about one huge section of peripherals -- music items.
The TB bus would be extremely useful because it is not just high bandwidth, but (IIRC) it has a very low latency. If TB became common, one could see keyboards and synths sporting a TB bus. This would open a lot of abilities to add new features, far more than just mLAN or raw MIDI could ever do. I can see an application that allows one to use the screen on a Korg Kronos or other high end keyboard and allow one to use its onboard tools and MIDI editor, then send the produced soundfiles back as tracks ready to be mixed.
Similar with video. A low latency bus will help a lot when dealing with various devices.
I'm seeing some high end devices using TB. An example would be enterprise-level tape drives (LTO-5) that can be attached to a laptop. Last time I've seen a capacity that was relevant that was usable with one was around 2000 with Firewire drives.
I wonder TB can stand out enough for it not to be put to the side. Since USB is the lowest common denominator, most devices will have at least this unless they really need faster I/O such as a mini SAN or a tape drive.
What will make or break TB is if it winds up on every PC out there just like USB 3, even if it is just used as a replacement for the VGA/DVI out for low end machines. If the critical mass of TB ports does form, then the devices will be made, be it SAN attachments, very fast SSD devices, or the ability to use a render server that streams video back, so devices don't have to have large GPUs in order to play games at a fast framerate.
Time will tell. If TB stays an "Apple-only" item, it will die on the vine. However, if it is available on consumer level hardware in the future, it likely will find a permanent niche, if only just for a replacement for the VGA out connector.
My cynical self says that is true, but I remember in the past people saying that Internet censorship was impossible. Now, it is commonplace.
I wouldn't be surprised if there is a son-of-ACTA bill brewing, where it wouldn't just do encryption, but signatures, so if something detects an unencrypted item (music/book/video/program), it would shut the device down, phone home, and call the local popo on a "IP tampering" violation.
DRM is improving. It took a long time for the iPhone 4s to be jailbroken. It took almost five years for any type of action to crack the PS3. Blu-Ray is still a cat and mouse game.
With a law and reactive infrastructure in place that would not just disable devices that are tampered with (think XBL bans), but also accounts. Then add criminal penalties onto it, and it wouldn't be surprising to see something put into place that would be robust without any cracks.
Yes, in theory, having Bob and Charlie be the same person is wrong, but throw enough tamper-resistant hardware at the problem, and it will work, just like how deploying censorware has effectively worked.
Call me naiive, but I thought the biggest different between an internal network and a private cloud is abstraction, especially using technologies like vMotion, PowerHA and others so hardware failure won't knock down the virtual server.
This also might be combined with a technology like Citrix that also provides a layer of abstraction between app servers and users. Citrix isn't perfect, but it can provide a layer of security, as well as keeping the critical user apps (and their data) a department uses in one place, as opposed to on each desktop.
In any case, call it a data center or private cloud, it is inherently more secure than an offsite installation any day, just due to who physically possesses the computers.
Tablet PCs were pretty useful. They functioned as nice laptops, and if one did want the tablet functionality, it was a screen flip away.
The trick would be to have ultrabooks get the 180 degree rotating screen and a touch screen. It wouldn't be as light as an iPad by any means, but it would be useful as a tablet, but when one needed to do actual heavy duty typing, the screen could be flipped around and the device used as a light notebook.
All and all, a good idea, especially since Windows 8 will have touch ability native to it.
I don't get why PC makers are flipping out. They just need to adapt.
Digressing from tablets, there are a ton of markets that are untapped which PC makers could find a very lucrative niche in.
First off, the idea of a render server. Devices on a LAN send the render server the Direct3D commands, the server does the rendering on a powerful array of GPUs, then streams the output back. The advantage of this is that games can then be played on any platform that has enough CPU and RAM. Of course, one needs a fast enough network connection, but 1GB is common, and it is only a matter of time before 10GB starts becoming common on the consumer level. PC makers could easily make a render server and sell it for a premium.
Second is working on the Windows Home Server concept. Cloud storage is nice, but with the bandwidth fees getting jacked up, people will be going back to a home server. This would be a superset of the features in a Time Capsule. Perhaps if done right, it could function as a true SAN with FCoE (or even iSCSI) LUNs that are backed up either as snapshots or file by file, a CIFS file server, a decent firewall/wireless AP, and so on. Essentially take the DSL or cable "modem", good firewalling router that can have multiple subnets (wired and wireless), backups, and the ability to add drives which can be mirrored then stored offsite, and put this all in one device.
Even the humble desktop box can get some changes to make it useful, especially if PC makers get a deal with Microsoft. For example, a card similar to a SIM card that is on the motherboard that stores activations. That way, the PC only needs activated for a new Windows product once with the cert stored on the SIM card. That way, in the future, if a box is licensed for Windows Server 2012, it will install and run without ever needing to activate. The SIM card allows the licenses to be moved to another box, so someone's XP license obtained in 2002 will be able to be used on a new box. This isn't perfect, as it is a lot like Steinberg's license key dongle, but if this is used as an activation "cache" as opposed to having to reactivate on any install or significant hardware modification, it can be useful.
The plain desktop PC can also have some useful stuff added. A built in hypervisor would be useful, so people could run one OS just for work stuff that is locked down, another OS for gaming, and another OS dedicated to browsing the pr0n sites. That way, the barrier to completely owning the machine is a lot higher. With newer CPUs coming with 6+ cores, might as well use them. Add to this a disk controller that has block level deduplication functionality, and even if someone has a ton of VMs (one just for WoW, one for banking, etc.), it will only store one instance of Windows for all of those. The hypervisor could even be given different users, so Billy can run his OS without interfering with Jill's term paper.
As for security, it isn't hard to add a decent fingerprint scanner on a desktop machine. This combined with BitLocker and a TPM chip would provide excellent security for home users who need protection against thieves. Even better, add this on the hypervisor level, so all the VMs are protected.
The PC companies just need to start doing some R&D. Yes, desktops are not snazzy, but they do fulfill a need that nothing else really can. Laptops come close, but one isn't going to be able to upgrade to the latest video card in one that easily.
Slashdot Mirror
Jailbreaking isn't really official, and it is becoming harder and harder for the Dev Teams to find a usable JB. For example, it took about two months for them to make jailbreakme.com when the iPhone 4 came out, then when the Greenpois0n exploit was found, that pretty much allowed any iPhone 4 (not 4s) to be jailbroken either tethered or semi-tethered. The 4s is a different beast altogether, and the gymnastics required to JB that device took a lot of effort.
It would be nice if Apple offered low level access to their devices, but realistically, it only is going to become longer and longer before a successful JB is achieved on future phones.
Plus, JB-ing isn't just getting something to run as root like on Android. It is installing a complete user environment, moving apps around, and many other workarounds that might change with each OS revision.
I've caught some apps by looking at the permissions asked for. For example, a game that asks for everything under the sun.
Then, when you look at the reviews, they are short and pithy, or consist of text like "App work[sic] great!".
That is when you know it isn't something you want on your device.
What is ironic is that I've yet to encounter an app that would request root permissions via su that isn't supposed to. I'm pretty sure it is because I refuse to install any app that requests irrelevant permissions, but it is sort of surprising that the baddies have not taken the tack of popping up a su prompt randomly. Maybe because users who root their phones would get very suspicious very fast.
I'm sure users who install it will find out it isn't up to snuff when all their contacts first get barraged by text messages from that device, then spam outlets as the contact data gets imported into the spammer databases.
Don't forget that anything in international waters that isn't under a country's flag is fair salvage for anyone who comes up to it. Even if it under a flag, the rule of the sea is often might makes right, so a data center on a barge might just get the flag of a white star on a blue background because the guys on the Zodiac boats with the AK-47s say so.
One answer would likely be tiers:
The first tier would be actively approved apps.
Then, if the user so chooses to set foot into Mordor, there can be a tier of apps that are downloadable almost immediately, and pulled if people justifiably report it as malicious.
This type of system has worked on jailbroken phones, where the App store serves one tier, and Cydia serves another. Since it takes a little bit of effort to JB an iPhone, generally someone is clued enough to be able to watch out for Trojans.
What this is protecting against, is arguably the biggest security hole of all; the user. Most smartphone users are not anywhere as savvy as a /. reader. The casual user will see an app that might offer "cool smilies", install it by reflex, and go on their merry way. On iOS, the damage a user can do is limited [1]. On Android, it is fairly easy to find apps that are malicious, and where a competent person would not install a fleshlight app that asks for full phone, GPS, contact, photos, and filesystem access (or even a prompt for a su), an inexperienced user will just click "install" nontheless, then scream that Android is insecure when they get bitten. iOS is designed to keep this from happening. Only beta code, Cydia apps, and enterprise apps are not coming through Apple's gateway. It is almost certain that the worst an iOS app can do is lighten the user's pocketbook due to its cost, or the cost of in-app transactions.
This isn't exactly the "dancing bunnies" security hole, but protecting the ignorant user from themselves is the difference between a platform having a rep as secure versus easily compromised.
I like both worlds. Have some barrier so a user doesn't exit the managed tier without a deliberate decision, then if they choose to, allow them to do what they want. This keeps the novices from footshooting while allowing people with a clue to use their device to the fullest.
[1]: Assuming the user doesn't JB, but generally if someone is clued enough to jailbreak, they will either know what they are doing, or end up having a clued friend DFU restoring their device and not do it again.
Doesn't BlackberryOS do this? Apple really should take a page from that PlayBook and have permissions for apps accessing the phone or text items, contacts, music, and photos. It wouldn't add that much clutter, and it would add a lot of protection.
On the cheap, maybe Apple should see about licensing the Cydia app Protect My Privacy and building that into the OS. That way, if an app does go and access stuff it shouldn't, it will get results, although it will just get a random UDID and garbage in the fields.
Contact list protection would help immensely if an app glitches like Facebook's and starts overwriting or appending contact fields without permission.
Once malware gets rooted out and Apple slams the banhammer down, it is a lot harder for a shady developer to get around closed accounts than on the Google Marketplace. This by itself keeps the bad guys on notice.
That is the main security mechanism of iOS which keeps the bad stuff at bay: As soon as Apple gets wind of something malicious or violating the rules, it gets tossed out immediately. The same action doesn't get repeated.
Now, once an app does get past the gatekeeper, it has a lot of room to play because only locations and alerts are granted/denied by the user. So, in theory, an app can copy pictures and contacts off, as well as send text messages all it wants. However, if users find something doing this, Apple squashes it.
Since Apple's reputation is on the line for security, the strong gatekeeper has shown that it is more secure than the weak gatekeeper/strong OS security of the Android ecosystem. Google needs to get with it and start having a tier of the Marketplace that requires apps to be actively approved, similar to what Amazon does.
On Android, I would recommend LBE Privacy Guard (requires root) to ensure FB keeps its sticky fingers out of the contacts.
On iOS, it requires jailbreaking, but there is a Cydia app called PMP or Protect My Privacy which will allow FB to have what it thinks is a contact list... when in reality, it is getting randomly generated garbage.
Either way, FB gets nothing that it shouldn't have if you know what you are doing.
A lot of the RF meters also have the capability of being shut off from remote. Having a wardriver see your electricity usage is one thing. Having someone be able to shut off electricity to people on a Friday before finals is something completely different.
I know Texas is a laughingstock, but from what I have experienced, the power grid here has been pretty reliable. There have not been any major blackouts other than one caused by a transformer oil fire about a decade ago, and the grid did remain up for the most part. The biggest danger here are ice storms. However, those are few and far between, especially with the warmer and drier climate.
The ironic thing is that most Texans also tend to have portable generators, so if power does go out, the generator gets dragged out of the garage, fired up [1], plugged into the transfer switch receptacle [2], and life goes on.
[1]: This is if one remembers to keep the fuel fresh and fire up the generator every few weeks. Otherwise, the carb bowl will be full of varnish and the jet will be clogged.
[2]: People can go to prison for criminal negligence if they use a suicide cord to plug their generator in (which backfeeds and can kill a lineman), so generally an automatic transfer switch plus a receptacle for the generator's cord gets installed.
There are also geographic issues as well. East of Houston to Florida is swamp. Good luck burying anything there. There is a reason why Louisiana is known for its elaborate crypts and morgues. There is just no way to bury the dead, so they have to remain above ground.
The US is a very disparate country. Some places the cities are as safe as Europe (Seattle, Portland, and chunks of NYC.) Other places, not so much. One of the main reason why some cities are burying cables now is because overhead lines tend to be a target for metal thieves so they can get their next meth fix.
To be specific, 6 volt, deep-cycle golf cart batteries. These used in pairs are a staple of RV boondocking because they hold a good amount of amp-hours.
I like using AGM batteries because they don't need water added, and can take a lot more incoming amperage than regular flooded cells can. However, they are more expensive.
Don't forget, even with the best solar system and wind turbine, there will be cases where one will need a generator. The staple for boondockers is usually two Honda 2000 watt inverters run with a parallel kit.
Don't cheap out on the generator, or else, and you might find yourself sans charger or refrigerator because the control board got fried. This is why I always recommend a Honda or Yamaha, since both will almost always outlive most RVs.
I'd just be happy with a battery that 1/2 to 1/10 of the energy volume density of gasoline, and can be scaled up.
That way, the Otto cycle engines can be chucked for electric motors which don't have energy loss due to exhaust or needless heat.
For RVs, it would allow for the rig to be completely electric. No loud generators, just use a high capacity inverter that can handle the 60 or so locked rotor amps from an A/C, and that is that. Then when you get to home or storage, plug the RV into shore power to trickle charge.
Technically, there are levels of currency that one can use.
At the lowest tier, where there is no civilization structure up enough to support validating the purity of gold, the only currency that would work would be ammo, since it is relatively small (assuming small arms and not tank shells), and useful.
The next tier up, where there is some infrastructure in place allowing for items that have less practical usefulness, but more fungible (one troy ounce of gold is one troy ounce of gold, assuming similar purity.)
Some tiers up from that would be a cryptographic based currency. Chaum and Timothy C. May have discussed that at length on the cypherpunks mailing list in the mid 1990s. In fact, Tim May wrote a long document called the Cyphernomicon which touches on a lot of points.
Yes, there are ways to make an anonymous currency. However, realistically, every government out there will step in to stop it. Bitcoins are not anonymous. I'm sure if someone kept doing over USD 10,000 worth of transactions into and out of the US in BitCoins, FinCEN will start taking notice and investigating.
An anonymous currency is a double-edged sword. On one hand, it keeps the guys who love to track every move a consumer makes out of the equation. On the other hand, it makes for a perfect way to exploit/extort/blackmail other people, similar to how e-Gold was used for payment when an intruder encrypted a victim's hard disk and demanded payment.
Even more ominous, it gives FB a lot of control over things it should have no business with.
Disclaimer: This is theory.
1: FB could in theory allow others access to that account, and thus allow people who had the cash to have access to any FB-derived authentication.
2: Someone compromising FB wouldn't just have that site, but a lot of others as well. Part of security is packing your own parachute. OpenID is good because it is distributed.
3: Relying on FB which has -zero- SLAs is about as bright as relying on using a high school student's cast off iBook on his Internet connection as the company critical E-mail server.
4: FB gains a lot of power. They boot someone, that person not just loses access to FB and the games on there, but Spotify and many other places. If push came to shove, people who FB chose to toss out could find themselves losing a lot of access to services.
I'd say people are becoming forced to in some ways.
For example authentication. A number of websites are using FB for their authentication mechanism, with no way to just create an account outside of that. This seems to be increasing because it seems to be easy to implement.
Right now, if a site demands FB authentication, they are avoidable. For example, Spotify. Easy change -- drop them, pick up Rdio which has a better selection anyway. However, I dread a time where if one wants to pay a phone bill online, order a pizza, or log onto a MMO, one must cough up a FB ID or else go elsewhere.
When recommending Macs to people [1], I get the virus question asked all the time. I try to clarify the difference between malware types.
Viruses are not really a viable infection vector on Macs because people don't share executables, and Word macro stuff is pretty much stomped out.
Trojan horses are a major threat. Especially when someone wants a pirated copy of something and finds that their copy of iWork has more than just an office suite in the .DMG file. Executable signing helps here, but the Dancing Bunnies security hole is still there. However, the biggest thing that closes this hole is to have a repository (which Apple has done), and dissuade users from installing from anywhere but that unless they are 100% sure of the source. Locking down the entry vector for applications with brutal gatekeepers is why iOS has such an excellent perceived security record while in reality, its whole security model pretty much depends on the BSD jail() command.
Probably the biggest threat these days are holes in Web browsers and add-ons. Even iOS had these, although it was used for extreme good with the jailbreakme.com website. However, on Windows, generally a hole in the Web browser will be used to insert malware at the user (or possibly system) level. The best protection here is having the OS sandbox with all writes redirected, or even run the browser in a virtual machine that rolls back all changes except to files stored in the browser's area.
[1]: For an individual, Apple has the best CS in town, unless someone buys a business line (Optiplex for example) with the premium service from Dell, HP, or another PC vendor. For someone who uses their personal machine for their livelihood, it isn't bad to have a one-stop-shop for the OS, hardware, and applications. That way, one doesn't spend time playing the "sorry, that's not our department" game. Apple's hardware is on par with other vendors, but they have a lot better support for individuals than anyone else out there.
With the requirement to store every single thing users do, it might be a good time to invest in EMC because it is going to require an enterprise (VNX level) SAN to record all what is going on, as well as the licenses for features like deduplication (since a bunch of troll posts are usually alike, the SAN can store one copy, and pointers to the others.)
As a user in the UK, I'd be looking to find the best always-on VPN service, one in the country (since some services are country-locked), and one situated somewhere less repressive but close by, network-wise, perhaps Sweden or Norway.
I'm sure that is going to be coming to the US really soon (if it isn't already present), so guess it is time to find a Canadian VPN provider.
I might be feeding the trolls here, but I see the "encrypt your data" thing often, like it is a magical switch I can flip and call it done.
The gotcha is the key management. For example, tape encryption. For a lot of companies, one can get adequate security by setting a long passphrase on the LTO-4 newer drives and library and using that for all tapes, perhaps changing it yearly with the old passphrases remaining for read compatibility with previous year media. However, some companies want to sell you a key device which makes every single tape have a different key. OK... what happens if the key device gets destroyed? Unless one constantly dumps the data out of it somehow, pretty much the tapes are unreadable. Of course, the key device vendor's solution is another key device and replication.
Or, perhaps take something as simple as BitLocker in the enterprise. This is as close as one can find to flipping on encryption. However, if the TPM on one of the machines gets zapped, the data is gone unless there is some recovery method, and that requires some type of infrastructure, be it storing recovery keys in Active Directory, using data recovery agent keys, or just having IT encrypt the volume, print out the key (so it can't be erased), and save the BEK file to some secure storage somewhere.
That isn't really how a TPM works.
First, most machines don't possess one, so DRM can't really assume it is there.
Second, it is shipped off and disabled. A user has to explicitly flip the TPM on in BIOS setup, then allow the OS to take ownership.
Of course, this technology is a double-edged sword. Look at the PS3 is an example.
However, on laptops, it provides an additional security boost, especially with a full disk encryption utility like BitLocker. The enhanced security it provides (allowing the OS to boot without a password needed, as well as protecting against "evil maid" attacks) is a help.
I'm curious about something with the pentalobe screws. Other than being used for tamper-resistance, I wonder if they allow for additional torque to be applied without camming out (a la Robertson or Torx.) If this were the case, it would make sense for Apple to use it.
I forgot about one huge section of peripherals -- music items.
The TB bus would be extremely useful because it is not just high bandwidth, but (IIRC) it has a very low latency. If TB became common, one could see keyboards and synths sporting a TB bus. This would open a lot of abilities to add new features, far more than just mLAN or raw MIDI could ever do. I can see an application that allows one to use the screen on a Korg Kronos or other high end keyboard and allow one to use its onboard tools and MIDI editor, then send the produced soundfiles back as tracks ready to be mixed.
Similar with video. A low latency bus will help a lot when dealing with various devices.
I'm seeing some high end devices using TB. An example would be enterprise-level tape drives (LTO-5) that can be attached to a laptop. Last time I've seen a capacity that was relevant that was usable with one was around 2000 with Firewire drives.
I wonder TB can stand out enough for it not to be put to the side. Since USB is the lowest common denominator, most devices will have at least this unless they really need faster I/O such as a mini SAN or a tape drive.
What will make or break TB is if it winds up on every PC out there just like USB 3, even if it is just used as a replacement for the VGA/DVI out for low end machines. If the critical mass of TB ports does form, then the devices will be made, be it SAN attachments, very fast SSD devices, or the ability to use a render server that streams video back, so devices don't have to have large GPUs in order to play games at a fast framerate.
Time will tell. If TB stays an "Apple-only" item, it will die on the vine. However, if it is available on consumer level hardware in the future, it likely will find a permanent niche, if only just for a replacement for the VGA out connector.
My cynical self says that is true, but I remember in the past people saying that Internet censorship was impossible. Now, it is commonplace.
I wouldn't be surprised if there is a son-of-ACTA bill brewing, where it wouldn't just do encryption, but signatures, so if something detects an unencrypted item (music/book/video/program), it would shut the device down, phone home, and call the local popo on a "IP tampering" violation.
DRM is improving. It took a long time for the iPhone 4s to be jailbroken. It took almost five years for any type of action to crack the PS3. Blu-Ray is still a cat and mouse game.
With a law and reactive infrastructure in place that would not just disable devices that are tampered with (think XBL bans), but also accounts. Then add criminal penalties onto it, and it wouldn't be surprising to see something put into place that would be robust without any cracks.
Yes, in theory, having Bob and Charlie be the same person is wrong, but throw enough tamper-resistant hardware at the problem, and it will work, just like how deploying censorware has effectively worked.
Call me naiive, but I thought the biggest different between an internal network and a private cloud is abstraction, especially using technologies like vMotion, PowerHA and others so hardware failure won't knock down the virtual server.
This also might be combined with a technology like Citrix that also provides a layer of abstraction between app servers and users. Citrix isn't perfect, but it can provide a layer of security, as well as keeping the critical user apps (and their data) a department uses in one place, as opposed to on each desktop.
In any case, call it a data center or private cloud, it is inherently more secure than an offsite installation any day, just due to who physically possesses the computers.
Tablet PCs were pretty useful. They functioned as nice laptops, and if one did want the tablet functionality, it was a screen flip away.
The trick would be to have ultrabooks get the 180 degree rotating screen and a touch screen. It wouldn't be as light as an iPad by any means, but it would be useful as a tablet, but when one needed to do actual heavy duty typing, the screen could be flipped around and the device used as a light notebook.
All and all, a good idea, especially since Windows 8 will have touch ability native to it.
I don't get why PC makers are flipping out. They just need to adapt.
Digressing from tablets, there are a ton of markets that are untapped which PC makers could find a very lucrative niche in.
First off, the idea of a render server. Devices on a LAN send the render server the Direct3D commands, the server does the rendering on a powerful array of GPUs, then streams the output back. The advantage of this is that games can then be played on any platform that has enough CPU and RAM. Of course, one needs a fast enough network connection, but 1GB is common, and it is only a matter of time before 10GB starts becoming common on the consumer level. PC makers could easily make a render server and sell it for a premium.
Second is working on the Windows Home Server concept. Cloud storage is nice, but with the bandwidth fees getting jacked up, people will be going back to a home server. This would be a superset of the features in a Time Capsule. Perhaps if done right, it could function as a true SAN with FCoE (or even iSCSI) LUNs that are backed up either as snapshots or file by file, a CIFS file server, a decent firewall/wireless AP, and so on. Essentially take the DSL or cable "modem", good firewalling router that can have multiple subnets (wired and wireless), backups, and the ability to add drives which can be mirrored then stored offsite, and put this all in one device.
Even the humble desktop box can get some changes to make it useful, especially if PC makers get a deal with Microsoft. For example, a card similar to a SIM card that is on the motherboard that stores activations. That way, the PC only needs activated for a new Windows product once with the cert stored on the SIM card. That way, in the future, if a box is licensed for Windows Server 2012, it will install and run without ever needing to activate. The SIM card allows the licenses to be moved to another box, so someone's XP license obtained in 2002 will be able to be used on a new box. This isn't perfect, as it is a lot like Steinberg's license key dongle, but if this is used as an activation "cache" as opposed to having to reactivate on any install or significant hardware modification, it can be useful.
The plain desktop PC can also have some useful stuff added. A built in hypervisor would be useful, so people could run one OS just for work stuff that is locked down, another OS for gaming, and another OS dedicated to browsing the pr0n sites. That way, the barrier to completely owning the machine is a lot higher. With newer CPUs coming with 6+ cores, might as well use them. Add to this a disk controller that has block level deduplication functionality, and even if someone has a ton of VMs (one just for WoW, one for banking, etc.), it will only store one instance of Windows for all of those. The hypervisor could even be given different users, so Billy can run his OS without interfering with Jill's term paper.
As for security, it isn't hard to add a decent fingerprint scanner on a desktop machine. This combined with BitLocker and a TPM chip would provide excellent security for home users who need protection against thieves. Even better, add this on the hypervisor level, so all the VMs are protected.
The PC companies just need to start doing some R&D. Yes, desktops are not snazzy, but they do fulfill a need that nothing else really can. Laptops come close, but one isn't going to be able to upgrade to the latest video card in one that easily.