You do not have to be Google to run all your content over HTTPS
Indeed you do not. The company I work for has several public facing websites that sustain > 200 Mbps during peak times. These sites are all SSL enabled and we don't have any issues keeping up with it. We use an SSL offload engine in our DC which sits in front of the web cluster and does all the SSL stuff for us. These are widely available and aren't as expensive as they used to be.
You don't even need a browser extension. It's very simple to alter the search settings to point to the HTTPS version of the site. I've been using encrypted.google.com since they made it available a year or so back with no issues. The only problem is that all of their content is not available over https. If you do a search on the https site and compare it to a search on the www site, you'll see that there are more service links (images, shopping, etc) on the www site.
And to the folks complaining that https slows down your connection - get real. Unless you live in a datacenter and are connected to the Internet vie GigE, any modern computer can encrypt and decrypt data far faster than your Internet connection can transmit it...
the rest of the drop is because there weren't many big pay-per-view events like boxing matches last quarter, and because regular movie rentals are down, too
Sure, but WHY are movie rentals down? Because people are smart and realize that instead of paying $4 (or whatever it costs these days) to rent a movie from Blockbuster or watch it on VOD, they can pay $8 and watch as many movies as they want on Netflix. You'd think that if Netflix can pay the applicable licensing fees to let you watch a movie or two every day for $8/mo, that the cable companies are making huge margins on their VOD rentals. If it were me, I'd sure as heck be figuring out how to lower my VOD pricing to something like $1-2 per rental to put themselves in the same price point as companies like redbox. Bam, you would probably instantly increase VOD sales by a factor of 10.
And while we're talking about VOD, folks know that you the cable co isn't the only game in town when it comes to watching UFC or other big ticket PPV shows. Roku and several other streaming services have them too. It's still the same OLD mentality that just doesn't get it...
You forget, this law probably only applies to the private sector. The government is obviously exempt from these rules because nobody would ever have reason to suspect the government doesn't have your best interests in mind.
Seriously? Granted, I didn't RTFM, but it sounds to me like it could bring a lot to the table, especially in rural areas. But even in areas where cable/dsl is readily available, this could be the third option that keeps the others in check. Currently, I can get cable or DSL but the DSL is so much slower than cable, it's really not a viable option. If this could match my current standard cable Internet speed (10 mbps), it might cause the cable company to realize they're not the only game in town and cause them to start adding additional value, either in terms of more speed or less cost.
DFS namespaces and DFS replication would solve this problem. It's what I use. It does have the disadvantage that you need Windows servers to make it work, but it's totally transparent and awesome once you get it going. I have a Windows server at home and another colo'ed (these are actually just VMs running on ESXi). I use a DFS namespace and have a target folder on each server. So, I can work on either set of files, and Windows keeps them in sync transparently, so for instance if I work on the files at the colo, then go home, I have a local copy of the files so access to them is nice and fast.
I am stuck in the unfortunate position of being responsible for my company's wireless account, which means anytime someone gets a new phone or has a problem with their current phone, they come see me. We have about 150 lines on our account, and I can definitely tell you that company issues iPhones have the highest failure rate. From what I've seen, the failure rate is about 4-5x that of what you would expect to see for a phone someone has paid for out of their own pocket. When you pay for it yourself and are on the hook for that 2 year agreement, you take care of your shit. If the company gives it to you, apparently you don't care because you know if you break it, you'll get another one.
These people were previously given Blackberries, so maybe they've come to expect their devices to take a certain level of abuse, but I will agree that Blackberry's quality has gone to shit over the past 3-4 years. If you take a 5 year old BB and put it next to a current model - well, it would make for an interesting torture test.
If anything - I would say that owning an iPhone can sometimes be exactly the opposite - a 'tard symbol.
Back when iPhone first came out, it *was* a status symbol and now all the unwashed masses buy them because they think it makes them cool.
I'd almost say that a high end Android device has replaced the iPhone as the "status symbol" - at least for people who know anything about anything. The iPhone 4 has been out for over a year, and as such is quite dated compared to some of the newer Android handsets.
The idea that most people cannot afford iPhones is wrong. You can go to AT&T and buy a 3GS for $49. I know people who are unemployed, have trouble buying diapers for their kids, but they're damn sure going to make sure they have the new iPhone when it comes out. It's sad to see what advertising and the perception of a phone as a social symbol does to people....
In terms of their streaming selection, it's not missing "oddball movies", it's missing a ton of movies. Try searching for the Lethal Weapon series, I think they had one of them...I wanted to watch "Under Siege" and found they didn't have any Steven Seagal movies. Please don't interpret that as my taste in movies, just making a point. They're missing a ton of movies I'd expect them to have. If I had to guess, I'd surmise that they have maybe 5% of their DVD movies available for streaming...
A quick Google tells me that there were about 400 movies released in theaters in 2010. If we assume that's about average, then we can determine that over the last 30 years, there have been about 12000 movies that were good enough to come out in theaters that people might want to stream. Figure about 2-3 GB of storage per movie and we arrive at a figure of approx 24-36 TB to store every movie that came out in a theater in the US in the last 30 years. Maybe it's just me, but that does not seem to be an unreasonable amount of data for Netflix to be storing. If we also factor in that there are movies in that list that Netflix doesn't have licensing to stream, and that there are also movies in there that totally bombed the box office so probably aren't worth having, and the figure comes down some.
So, megacorp Netflix could have every movie from the last 30 years for, let's be generous and call it 50 TB. I realize that means 50 TB of data at a whole bunch of different CDNs, but still, that doesn't seem that unreasonable to me.
The situation I've had to deal with is video storage for a small FCP production team. They have approx 30 TB of disk storage and it's all set up as RAID 5 to maximize the storage capacity of the disks. I know we really ought to be using RAID 6, but unfortunately that was viewed as costing too much and consuming too much of the available storage for redundancy. When we buy storage, we buy it in doubles, that is to say if we buy an array to add 10 TB, we actually buy two. One is production, the other is backup. We use a simple utility called Chronosync to push any incremental changes from the production storage to the backup storage nightly.
We don't have any offline storage at all. We know that if the building burns down or a tornado rips the roof off, we're SOL. It's not a matter of not wanting to do it, it's just that there's no practical and financially feasible way for us to keep that much data backed up on offline, offsite storage. We'd essentially have to hire someone to handle that extra workload, and then we're talking about 50-60k/yr in salary, plus the storage medium (LTO tape, etc).
With that being said, this is the most cost effective solution I've found, and it works well for what it's designed to do, which is protect against disk failure.
If you need WiFi, and want security too, stop trying to implement layer 2 security and move on to layer 3. It is much more practical to set up your WiFi network so that it has no route to your network, the Internet, or anything else, and then use VPN software to establish a secure tunnel, which in turns gets you access to these things. IPSec VPN with AES-256 encryption has been around for quite some time, is freely available, and isn't in the news for being cracked on a weekly basis like WiFi was/is.
Toms Hardware had an article a few weeks ago about cracking encryption using GPUs. They concluded that as long as you're using a secure password, AES-256 encryption will keep your data safe well beyond the time you die, even against big multi-GPU clusters purpose built for password cracking. Of course, Moore's law has implications here, but as of *right now*, you would be hard pressed to find a method providing better wireless security.
For anyone else wondering how to get the jump list to work, make sure you have saved sessions setup, and double click to connect. After you do this, that saved session should show up in the jump list. Rock!
I.E. how Microsoft is actually doing this password analysis, because we would presume that they're smart enough not to store them in clear text so anonymous/lulzsec/etc can come steal them. I wouldn't be surprised if they just popped themselves up on the radar of hacker groups, "Hey guys, M$ must be storing about 50 million passwords in clear text!!1"
Certainly, you can compare hashes to get a count of identical passwords, but then how do you know what those passwords actually are in order to ban them?
Parent title should be "never use RAID as a backup", unless the RAID in question IS a backup copy of data living elsewhere. In his example, the two failing drives in the RAID 10 set wouldn't matter, unless the primary array has also failed. Sure, you'd lose the RAID set, but RAID was never intended to be used as a backup mechanism in and of itself. By maintaining two separate arrays, you would have to have complete failure of two arrays simultaneously for your data to be lost. Very unlikely.
Did they consider that many of these might just be hotspots that were intentionally left open? 2.5% doesn't strike me as an unreasonable percentage of people that intentionally leave their wifi open. It'd be relatively simple to modify the test procedure to attempt to connect to the networks and 1) see if they can lease an IP address, which would tell if they at least have MAC-based authentication enabled, and if they can get an IP 2) see if they get the expected text when trying to connect to a predetermined website, which would tell them if there's a captive portal operating. Both of those things could be seen as valid reasons for leaving wifi networks unencrypted, but I wouldn't necessarily call either of those scenarios an "open and unsecured" network.
Am I the only one that thinks it's really slow? It seems to me that both the robot, and the computing resources available at Stanford ought to not be the limiting factors here, so why is it moving so slowly?
There are already a bunch of textbook rental companies out there, CourseSmart being the one I've used most often. The concept is good, but because they lock the content down with so much DRM, it severely limits the usability. I want a simple PDF file that I can easily search. I'd even be willing to install some sort of Acrobat DRM control (in my Windows VM, mind you), and I'D PAY MORE if I could actually get a regular PDF file that simply stopped working on a certain date.
I don't have any desire to try and search stuff on my iPad using the Kindle app, and the ridiculously locked-down stuff other companies have isn't much better.
JUST GIVE ME A PDF!
While there are products that do this (dual WAN firewalls, etc) none of them are particularly great. If it were me, I'd repurpose an old PC, or a dedicated board such as a Soekris 4501 (http://soekris.com/net4501.html) and roll your own. It should be pretty simple to do it with iptables and a few bash scripts. Off the top of my head, I'd do something like ping a device a few hops upstream on each providers network every 60 seconds or so, if the device isn't responding, then failover and use another script to failback when the device is reachable again.
Using linux would allow.you to incorporate traffic monitoring, QoS, etc and even a file/web/dns server if you want to. In short, the linux route keeps you from getting locked in to a proprietary system that may not meet all your needs. There's load of documentation on doing stuff like this available online, so you don't even really need any prior linux experience as long as you have a will to learn.
Sorry Mr. Bar Bouncer Guy. I don't have an ID because carrying a wallet is so 2011. But I do have this digital ID picture on my smartphone, and I promise it's not fake even thought you would have absolutely no way to tell if it was.
I wonder if they're taking into account all the little indian-owned gas stations that will illegally refuse to accept a credit/debit card and require you to pay cash if you're just buying a 99 cent soda?
And I totally agree re: eBay/Paypal. I wish them a fiery death. I sold my old iPhone on eBay earlier this year, it got damaged in shipping, and Paypal gave the guy his money back. Even though the package was insured. They didn't even make him send the phone back so I could at least file an insurance claim with UPS and get my money back. So...shafted out of $200. So, after they stole my $200, I sent them an inquiry through their website, a couple days later I got an email basically saying sorry we haven't gotten back to you yet, then another few days later, another email saying gosh, we're real sorry but we just don't have time to answer your e-mail, so we consider the matter closed. If you still have a problem contact our level 1 phone support lackeys who have absolutely no authority to do anything that will help you and cannot escalate your request to anyone else that can. I tried submitting the form again as well as e-mailing my complaint to every conceivable e-mail alias @paypal.com and got the exact same response, so I'm guessing that's just their SOP. Give customers the shaft, ignore their complaints, lock every eBay transaction into using Paypal, PROFIT!
I am a hobbyist woodworker, and would like to think that we are perhaps some of the leading authorities on dealing with containment of dust and other airborne particulates. Sanding wood can generate particulates smaller than 1 micron and that stuff is very bad for your lungs.
With that being said, there is a whole industry devoted to dust collection for woodworkers and I think you would do well to investigate it. It sounds like a respirator is what you need. You can get units that will filter down to 1 micron off the shelf or by mail order from Amazon, Woodcraft, Rockler, etc for well under $100. If you're using a standard painters mask, you're probably breathing in a lot more of the dust than you realize. They don't filter to anywhere near 1 micron, and the poor fit to your face is letting unfiltered air go right around the mask into your lungs.
I would recommend doing your actual cutting outdoors or in as well a ventilated space as possible, and then going elsewhere for a while after you finish to let the dust settle. It can take several hours for airborne particulates to fall out of suspension, especially if they're really tiny.
If you need to remain in the area, there are other methods of containing the dust. Google "dust collector" to get an idea as to the amount of options out there. A shopvac is definitely the wrong solution for what you're trying to do, they don't filter small enough particles or have enough CFMs for adequate dust collection. Festool makes dust vacs that have HEPA filters and are actually designed for this sort of thing. They're pricey, but dirt cheap when compared to the cost of new lungs.
I didn't read TFA to see if they're talking about residential PSTN connections only, but in terms of businesses, I don't see this happening anytime soon.
If you're a business, you basically have three and a half choices when it comes to voice communication with the rest of the world:
1) POTS, most small businesses still rely on this, but it's impractical for larger businesses with more than about 50 employees at a given location
2) ISDN - many medium and most large businesses use ISDN for connectivity to the PSTN because it just plain works
3A) On-net VoIP - I.E. phone service from your cable company, or SIP over a dedicated link (i.e. from a CLEC over a T1, etc / not over the Internet)
3B) Off-net VoIP - I.E. Skype, Vonage, Ring Central, Google Voice, etc.
Of these, off-net VoIP really has no QoS capability, so it's not especially suitable for business use. I know tons of people use it and are happy with it (myself included, for home phone service), but when the rubber meets the road, businesses are more than willing to pay for ISDN or on-net VoIP for the QoS and reliability.
On-net VoIP is where I see the traditional PSTN losing most of their business over the next few years. But, you have to realize that the majority of small businesses that are using it are doing so over channelized T1 circuits, which still require the PSTN (T1 circuits = PSTN). You can also do it over fiber, and that's where I see the bulk of Ma Bell's business going. Although the phone companies are in some cases still providing the fiber, I don't consider that to be the PSTN.
So when you look at it, the phone companies are losing business, but I don't see the PSTN going away any time soon.
I think what the article meant to say is that POTS is going away, but POTS != PSTN. I think the article is just poorly summarized.
This is a very valid point. One of the big strengths of traditional PSTN service is that it works even when your power is out. There's a reason that IT guys refer to extremely high reliability gear as 'carrier' or 'telco' grade. The PSTN just doesn't go down all that often. I'm sure this has changed somewhat over the past 10-15 years as more of the 'non last-mile' portion of the telco network is becoming digital. I.E. my neighborhood has analog copper from the houses, but that only goes less than a mile to a multiplexer where that signal is carried further on digital circuits, and those do need power to work.
With regard to cell phones, I would imagine that carriers have standby power at most if not all cell sites. If the FCC doesn't already mandate at least a certain level of standby power, I think they should start. I.E. every cell site needs to stay alive for at least XXX minutes in the event of a power failure.
As far as VoIP solutions, all of the solutions the local cable co sells come with a combo modem / ATA that has a built in battery backup, so your phone and Internet stay up when the power goes out. At least for a while.
I recently bought my dad an Apple TV for father's day and forgot to pick up an HDMI cable for it, so I can to Best Buy and the cheapest HDMI cable they had in the entire store was $40. I understand you pay more at brick and mortars, but compare this to newegg where you can get the same thing for $4, and you see that this 1000% markup is a bit excessive.
The whole point of an internship program is that an intern is not ready to go out and stand on their own in the job force. Could be because they don't have enough experience to get their foot in the door, or maybe they're technically saavy but don't have enough business sense to apply it effectively.
Either way, yes, an intern can provide valuable service to a business, but they're also taking the time of other employees to mentor them, i.e. this is what you learned in school and these are the reasons why it doesn't work that way in the real world. Most technical programs don't teach you anything about the basic stuff that you actually need. I.E. instead of an AS program teaching students about what IRQs are used by what devices and how token ring works (seriously, who's needed that info in the past 10 years?), they should substitute that with a course about how to effectively communicate with non-technical people - i.e. the people you have to do work for, and the people you have to report to. It'd also be great if they could at least give people a summary overview of operating systems and operating system history. It's great that you've been trained on Windows 7 and Server 2008 because your school subscribes to the Microsoft dog food, but how does that help businesses that are still running XP, Server 2003, Win2000 and even NT4 and DOS? You might think they're obsolete, but in my local area, I know of at least three large IT employers that are still running applications on NT4 because of compatibility issues with Win2XXX. I'm not saying that they should spend a bunch of time teaching kids about WinNT, but it'd be nice if they at least know that it exists when they show up for an internship.
But, they don't, and that's why they're not ready to stand on their own in the workforce. Remember that most people graduating from IT programs have no prior IT experience, and all the little nuggets of knowledge you've picked up from reading Slashdot and other tech news over the past 15 years, none of these people have any of that. It's not that reading tech news makes you an expert on any given subject, but at least you have a passing familiarity with stuff. I.E. I've never used FreeBSD, but I know what it is.
Maybe I've just been lucky, but it is quite easy to run your own SMTP server here in Florida on Road Runner. They don't restrict in/outbound SMTP, so you can pretty do much whatever you want.
Even if they did block outbound SMTP (which I think is a halfway decent idea), there are plenty of services out there that will provide SMTP relay service for you, either on nonstandard ports so you can get by firewall rules, or via VPN, so the firewall can't even see the traffic. Last I checked, these services were a few bucks a month or something like $100/year. If you need to run your own mail server for business purposes, this shouldn't be too burdensome, plus you get the benefit of redundant smart hosts, so you don't have to worry about your ISP's mail server going offline.
You do not have to be Google to run all your content over HTTPS
Indeed you do not. The company I work for has several public facing websites that sustain > 200 Mbps during peak times. These sites are all SSL enabled and we don't have any issues keeping up with it. We use an SSL offload engine in our DC which sits in front of the web cluster and does all the SSL stuff for us. These are widely available and aren't as expensive as they used to be.
You don't even need a browser extension. It's very simple to alter the search settings to point to the HTTPS version of the site. I've been using encrypted.google.com since they made it available a year or so back with no issues. The only problem is that all of their content is not available over https. If you do a search on the https site and compare it to a search on the www site, you'll see that there are more service links (images, shopping, etc) on the www site.
And to the folks complaining that https slows down your connection - get real. Unless you live in a datacenter and are connected to the Internet vie GigE, any modern computer can encrypt and decrypt data far faster than your Internet connection can transmit it...
the rest of the drop is because there weren't many big pay-per-view events like boxing matches last quarter, and because regular movie rentals are down, too
Sure, but WHY are movie rentals down? Because people are smart and realize that instead of paying $4 (or whatever it costs these days) to rent a movie from Blockbuster or watch it on VOD, they can pay $8 and watch as many movies as they want on Netflix. You'd think that if Netflix can pay the applicable licensing fees to let you watch a movie or two every day for $8/mo, that the cable companies are making huge margins on their VOD rentals. If it were me, I'd sure as heck be figuring out how to lower my VOD pricing to something like $1-2 per rental to put themselves in the same price point as companies like redbox. Bam, you would probably instantly increase VOD sales by a factor of 10. And while we're talking about VOD, folks know that you the cable co isn't the only game in town when it comes to watching UFC or other big ticket PPV shows. Roku and several other streaming services have them too. It's still the same OLD mentality that just doesn't get it...
You forget, this law probably only applies to the private sector. The government is obviously exempt from these rules because nobody would ever have reason to suspect the government doesn't have your best interests in mind.
Seriously? Granted, I didn't RTFM, but it sounds to me like it could bring a lot to the table, especially in rural areas. But even in areas where cable/dsl is readily available, this could be the third option that keeps the others in check. Currently, I can get cable or DSL but the DSL is so much slower than cable, it's really not a viable option. If this could match my current standard cable Internet speed (10 mbps), it might cause the cable company to realize they're not the only game in town and cause them to start adding additional value, either in terms of more speed or less cost.
DFS namespaces and DFS replication would solve this problem. It's what I use. It does have the disadvantage that you need Windows servers to make it work, but it's totally transparent and awesome once you get it going. I have a Windows server at home and another colo'ed (these are actually just VMs running on ESXi). I use a DFS namespace and have a target folder on each server. So, I can work on either set of files, and Windows keeps them in sync transparently, so for instance if I work on the files at the colo, then go home, I have a local copy of the files so access to them is nice and fast.
I am stuck in the unfortunate position of being responsible for my company's wireless account, which means anytime someone gets a new phone or has a problem with their current phone, they come see me. We have about 150 lines on our account, and I can definitely tell you that company issues iPhones have the highest failure rate. From what I've seen, the failure rate is about 4-5x that of what you would expect to see for a phone someone has paid for out of their own pocket. When you pay for it yourself and are on the hook for that 2 year agreement, you take care of your shit. If the company gives it to you, apparently you don't care because you know if you break it, you'll get another one.
These people were previously given Blackberries, so maybe they've come to expect their devices to take a certain level of abuse, but I will agree that Blackberry's quality has gone to shit over the past 3-4 years. If you take a 5 year old BB and put it next to a current model - well, it would make for an interesting torture test.
If anything - I would say that owning an iPhone can sometimes be exactly the opposite - a 'tard symbol.
Back when iPhone first came out, it *was* a status symbol and now all the unwashed masses buy them because they think it makes them cool.
I'd almost say that a high end Android device has replaced the iPhone as the "status symbol" - at least for people who know anything about anything. The iPhone 4 has been out for over a year, and as such is quite dated compared to some of the newer Android handsets.
The idea that most people cannot afford iPhones is wrong. You can go to AT&T and buy a 3GS for $49. I know people who are unemployed, have trouble buying diapers for their kids, but they're damn sure going to make sure they have the new iPhone when it comes out. It's sad to see what advertising and the perception of a phone as a social symbol does to people....
In terms of their streaming selection, it's not missing "oddball movies", it's missing a ton of movies. Try searching for the Lethal Weapon series, I think they had one of them...I wanted to watch "Under Siege" and found they didn't have any Steven Seagal movies. Please don't interpret that as my taste in movies, just making a point. They're missing a ton of movies I'd expect them to have. If I had to guess, I'd surmise that they have maybe 5% of their DVD movies available for streaming... A quick Google tells me that there were about 400 movies released in theaters in 2010. If we assume that's about average, then we can determine that over the last 30 years, there have been about 12000 movies that were good enough to come out in theaters that people might want to stream. Figure about 2-3 GB of storage per movie and we arrive at a figure of approx 24-36 TB to store every movie that came out in a theater in the US in the last 30 years. Maybe it's just me, but that does not seem to be an unreasonable amount of data for Netflix to be storing. If we also factor in that there are movies in that list that Netflix doesn't have licensing to stream, and that there are also movies in there that totally bombed the box office so probably aren't worth having, and the figure comes down some. So, megacorp Netflix could have every movie from the last 30 years for, let's be generous and call it 50 TB. I realize that means 50 TB of data at a whole bunch of different CDNs, but still, that doesn't seem that unreasonable to me.
The situation I've had to deal with is video storage for a small FCP production team. They have approx 30 TB of disk storage and it's all set up as RAID 5 to maximize the storage capacity of the disks. I know we really ought to be using RAID 6, but unfortunately that was viewed as costing too much and consuming too much of the available storage for redundancy. When we buy storage, we buy it in doubles, that is to say if we buy an array to add 10 TB, we actually buy two. One is production, the other is backup. We use a simple utility called Chronosync to push any incremental changes from the production storage to the backup storage nightly.
We don't have any offline storage at all. We know that if the building burns down or a tornado rips the roof off, we're SOL. It's not a matter of not wanting to do it, it's just that there's no practical and financially feasible way for us to keep that much data backed up on offline, offsite storage. We'd essentially have to hire someone to handle that extra workload, and then we're talking about 50-60k/yr in salary, plus the storage medium (LTO tape, etc).
With that being said, this is the most cost effective solution I've found, and it works well for what it's designed to do, which is protect against disk failure.
If you need WiFi, and want security too, stop trying to implement layer 2 security and move on to layer 3. It is much more practical to set up your WiFi network so that it has no route to your network, the Internet, or anything else, and then use VPN software to establish a secure tunnel, which in turns gets you access to these things. IPSec VPN with AES-256 encryption has been around for quite some time, is freely available, and isn't in the news for being cracked on a weekly basis like WiFi was/is.
Toms Hardware had an article a few weeks ago about cracking encryption using GPUs. They concluded that as long as you're using a secure password, AES-256 encryption will keep your data safe well beyond the time you die, even against big multi-GPU clusters purpose built for password cracking. Of course, Moore's law has implications here, but as of *right now*, you would be hard pressed to find a method providing better wireless security.
For anyone else wondering how to get the jump list to work, make sure you have saved sessions setup, and double click to connect. After you do this, that saved session should show up in the jump list. Rock!
I.E. how Microsoft is actually doing this password analysis, because we would presume that they're smart enough not to store them in clear text so anonymous/lulzsec/etc can come steal them. I wouldn't be surprised if they just popped themselves up on the radar of hacker groups, "Hey guys, M$ must be storing about 50 million passwords in clear text!!1" Certainly, you can compare hashes to get a count of identical passwords, but then how do you know what those passwords actually are in order to ban them?
Parent title should be "never use RAID as a backup", unless the RAID in question IS a backup copy of data living elsewhere. In his example, the two failing drives in the RAID 10 set wouldn't matter, unless the primary array has also failed. Sure, you'd lose the RAID set, but RAID was never intended to be used as a backup mechanism in and of itself. By maintaining two separate arrays, you would have to have complete failure of two arrays simultaneously for your data to be lost. Very unlikely.
Did they consider that many of these might just be hotspots that were intentionally left open? 2.5% doesn't strike me as an unreasonable percentage of people that intentionally leave their wifi open. It'd be relatively simple to modify the test procedure to attempt to connect to the networks and 1) see if they can lease an IP address, which would tell if they at least have MAC-based authentication enabled, and if they can get an IP 2) see if they get the expected text when trying to connect to a predetermined website, which would tell them if there's a captive portal operating. Both of those things could be seen as valid reasons for leaving wifi networks unencrypted, but I wouldn't necessarily call either of those scenarios an "open and unsecured" network.
Am I the only one that thinks it's really slow? It seems to me that both the robot, and the computing resources available at Stanford ought to not be the limiting factors here, so why is it moving so slowly?
There are already a bunch of textbook rental companies out there, CourseSmart being the one I've used most often. The concept is good, but because they lock the content down with so much DRM, it severely limits the usability. I want a simple PDF file that I can easily search. I'd even be willing to install some sort of Acrobat DRM control (in my Windows VM, mind you), and I'D PAY MORE if I could actually get a regular PDF file that simply stopped working on a certain date. I don't have any desire to try and search stuff on my iPad using the Kindle app, and the ridiculously locked-down stuff other companies have isn't much better. JUST GIVE ME A PDF!
While there are products that do this (dual WAN firewalls, etc) none of them are particularly great. If it were me, I'd repurpose an old PC, or a dedicated board such as a Soekris 4501 (http://soekris.com/net4501.html) and roll your own. It should be pretty simple to do it with iptables and a few bash scripts. Off the top of my head, I'd do something like ping a device a few hops upstream on each providers network every 60 seconds or so, if the device isn't responding, then failover and use another script to failback when the device is reachable again.
Using linux would allow.you to incorporate traffic monitoring, QoS, etc and even a file/web/dns server if you want to. In short, the linux route keeps you from getting locked in to a proprietary system that may not meet all your needs. There's load of documentation on doing stuff like this available online, so you don't even really need any prior linux experience as long as you have a will to learn.
Sorry Mr. Bar Bouncer Guy. I don't have an ID because carrying a wallet is so 2011. But I do have this digital ID picture on my smartphone, and I promise it's not fake even thought you would have absolutely no way to tell if it was.
I wonder if they're taking into account all the little indian-owned gas stations that will illegally refuse to accept a credit/debit card and require you to pay cash if you're just buying a 99 cent soda?
And I totally agree re: eBay/Paypal. I wish them a fiery death. I sold my old iPhone on eBay earlier this year, it got damaged in shipping, and Paypal gave the guy his money back. Even though the package was insured. They didn't even make him send the phone back so I could at least file an insurance claim with UPS and get my money back. So...shafted out of $200. So, after they stole my $200, I sent them an inquiry through their website, a couple days later I got an email basically saying sorry we haven't gotten back to you yet, then another few days later, another email saying gosh, we're real sorry but we just don't have time to answer your e-mail, so we consider the matter closed. If you still have a problem contact our level 1 phone support lackeys who have absolutely no authority to do anything that will help you and cannot escalate your request to anyone else that can. I tried submitting the form again as well as e-mailing my complaint to every conceivable e-mail alias @paypal.com and got the exact same response, so I'm guessing that's just their SOP. Give customers the shaft, ignore their complaints, lock every eBay transaction into using Paypal, PROFIT!
I am a hobbyist woodworker, and would like to think that we are perhaps some of the leading authorities on dealing with containment of dust and other airborne particulates. Sanding wood can generate particulates smaller than 1 micron and that stuff is very bad for your lungs.
With that being said, there is a whole industry devoted to dust collection for woodworkers and I think you would do well to investigate it. It sounds like a respirator is what you need. You can get units that will filter down to 1 micron off the shelf or by mail order from Amazon, Woodcraft, Rockler, etc for well under $100. If you're using a standard painters mask, you're probably breathing in a lot more of the dust than you realize. They don't filter to anywhere near 1 micron, and the poor fit to your face is letting unfiltered air go right around the mask into your lungs.
I would recommend doing your actual cutting outdoors or in as well a ventilated space as possible, and then going elsewhere for a while after you finish to let the dust settle. It can take several hours for airborne particulates to fall out of suspension, especially if they're really tiny.
If you need to remain in the area, there are other methods of containing the dust. Google "dust collector" to get an idea as to the amount of options out there. A shopvac is definitely the wrong solution for what you're trying to do, they don't filter small enough particles or have enough CFMs for adequate dust collection. Festool makes dust vacs that have HEPA filters and are actually designed for this sort of thing. They're pricey, but dirt cheap when compared to the cost of new lungs.
I didn't read TFA to see if they're talking about residential PSTN connections only, but in terms of businesses, I don't see this happening anytime soon.
If you're a business, you basically have three and a half choices when it comes to voice communication with the rest of the world:
1) POTS, most small businesses still rely on this, but it's impractical for larger businesses with more than about 50 employees at a given location
2) ISDN - many medium and most large businesses use ISDN for connectivity to the PSTN because it just plain works
3A) On-net VoIP - I.E. phone service from your cable company, or SIP over a dedicated link (i.e. from a CLEC over a T1, etc / not over the Internet)
3B) Off-net VoIP - I.E. Skype, Vonage, Ring Central, Google Voice, etc.
Of these, off-net VoIP really has no QoS capability, so it's not especially suitable for business use. I know tons of people use it and are happy with it (myself included, for home phone service), but when the rubber meets the road, businesses are more than willing to pay for ISDN or on-net VoIP for the QoS and reliability.
On-net VoIP is where I see the traditional PSTN losing most of their business over the next few years. But, you have to realize that the majority of small businesses that are using it are doing so over channelized T1 circuits, which still require the PSTN (T1 circuits = PSTN). You can also do it over fiber, and that's where I see the bulk of Ma Bell's business going. Although the phone companies are in some cases still providing the fiber, I don't consider that to be the PSTN.
So when you look at it, the phone companies are losing business, but I don't see the PSTN going away any time soon.
I think what the article meant to say is that POTS is going away, but POTS != PSTN. I think the article is just poorly summarized.
This is a very valid point. One of the big strengths of traditional PSTN service is that it works even when your power is out. There's a reason that IT guys refer to extremely high reliability gear as 'carrier' or 'telco' grade. The PSTN just doesn't go down all that often. I'm sure this has changed somewhat over the past 10-15 years as more of the 'non last-mile' portion of the telco network is becoming digital. I.E. my neighborhood has analog copper from the houses, but that only goes less than a mile to a multiplexer where that signal is carried further on digital circuits, and those do need power to work.
With regard to cell phones, I would imagine that carriers have standby power at most if not all cell sites. If the FCC doesn't already mandate at least a certain level of standby power, I think they should start. I.E. every cell site needs to stay alive for at least XXX minutes in the event of a power failure.
As far as VoIP solutions, all of the solutions the local cable co sells come with a combo modem / ATA that has a built in battery backup, so your phone and Internet stay up when the power goes out. At least for a while.
I recently bought my dad an Apple TV for father's day and forgot to pick up an HDMI cable for it, so I can to Best Buy and the cheapest HDMI cable they had in the entire store was $40. I understand you pay more at brick and mortars, but compare this to newegg where you can get the same thing for $4, and you see that this 1000% markup is a bit excessive.
The whole point of an internship program is that an intern is not ready to go out and stand on their own in the job force. Could be because they don't have enough experience to get their foot in the door, or maybe they're technically saavy but don't have enough business sense to apply it effectively. Either way, yes, an intern can provide valuable service to a business, but they're also taking the time of other employees to mentor them, i.e. this is what you learned in school and these are the reasons why it doesn't work that way in the real world. Most technical programs don't teach you anything about the basic stuff that you actually need. I.E. instead of an AS program teaching students about what IRQs are used by what devices and how token ring works (seriously, who's needed that info in the past 10 years?), they should substitute that with a course about how to effectively communicate with non-technical people - i.e. the people you have to do work for, and the people you have to report to. It'd also be great if they could at least give people a summary overview of operating systems and operating system history. It's great that you've been trained on Windows 7 and Server 2008 because your school subscribes to the Microsoft dog food, but how does that help businesses that are still running XP, Server 2003, Win2000 and even NT4 and DOS? You might think they're obsolete, but in my local area, I know of at least three large IT employers that are still running applications on NT4 because of compatibility issues with Win2XXX. I'm not saying that they should spend a bunch of time teaching kids about WinNT, but it'd be nice if they at least know that it exists when they show up for an internship. But, they don't, and that's why they're not ready to stand on their own in the workforce. Remember that most people graduating from IT programs have no prior IT experience, and all the little nuggets of knowledge you've picked up from reading Slashdot and other tech news over the past 15 years, none of these people have any of that. It's not that reading tech news makes you an expert on any given subject, but at least you have a passing familiarity with stuff. I.E. I've never used FreeBSD, but I know what it is.
Maybe I've just been lucky, but it is quite easy to run your own SMTP server here in Florida on Road Runner. They don't restrict in/outbound SMTP, so you can pretty do much whatever you want. Even if they did block outbound SMTP (which I think is a halfway decent idea), there are plenty of services out there that will provide SMTP relay service for you, either on nonstandard ports so you can get by firewall rules, or via VPN, so the firewall can't even see the traffic. Last I checked, these services were a few bucks a month or something like $100/year. If you need to run your own mail server for business purposes, this shouldn't be too burdensome, plus you get the benefit of redundant smart hosts, so you don't have to worry about your ISP's mail server going offline.