Good advice. In my case, Gandi is my registar and web host (I rather like their hosting plans, as they tend not to suck), but the DNS is provided by CloudFlare (who sits in front of the site as a reverse proxy). If needed, I can have CloudFlare point the DNS elsewhere until the S stops HTF.
I have some domains with NameCheap (which others have pointed out is quite good) and others with Gandi.net.
NameCheap has somewhat better pricing, but I prefer Gandi for most things. I rather like that Gandi's whois privacy is included by default in the cost of registration with compatible domains (.us, for example, does not allow private registration). The only public information they display is your full name and NIC handle -- the address and phone number are changed to that of their Paris headquarters and the email address is a spam-protected alias. I like that they display your full name, as their policy is that regardless of if you have the privacy service enabled or not your domain is still yours and you own it. Other registrars have the policy that they own (for as long as the whois privacy is enabled) your domain and are leasing it back to you.
If you're already with Gandi, check out their Simple Hosting. It's pretty slick, as far as basic hosting goes: you get your own Apache/MySQL/PHP processes, the web server runs with the same permissions as your user account (so setting up stuff like WordPress is trivial as there's no permissions-related issues), can host multiple separate sites on a single instance, etc.
Their VPSs are pretty standard paravirtualized Xen systems which work out pretty well (I ran a Team Fortress 2 gameserver for a while on one and it was stable and reliable).
As a domain customer you get a 50% discount code for the first year ($30/year rather than $60/year).
Disclaimer: Gandi customer, not employed by them in any way.
Also FYI, I typoed the URL for Google's ad preferences. Here is the correct URL: http://www.google.com/ads/preferences/ -- I left off the "s" at the end of "preferences". Mea culpa.
It's already starting to bother me. I'm seeing these advertisements here on Slashdot too. After I've searched for something on Google, the related advertisements start to come up EVERYWHERE on the internet. Seriously, they come after you. If you search for specific flights you start to see ads for that everyone. It'll haunt you and there's nothing you can do.
Similarly, you can use the NAI's opt-out page to opt-out of Google and other ad network tracking.
There's plenty of browser plugins that work to block ads entirely (such as AdBlock) and ones that ensure that the "opt-out" cookies stay in existence even if you clear your other cookies.
All the other browsers than Safari and IE are in bed with advertisers because both Firefox and Opera get revenue directly from Google.
The default search box in those browsers comes configured to use Google, yes. They do get income from ad revenue stemming from searches from the box. You're not forced to use that search box, nor are you forced to use the default settings -- you can add other search providers (like DuckDuckGo, ixquick, etc.) -- Firefox, for one, doesn't have ad agreements with anyone other than Google.
So for the love of god Apache Project, stop taking bribes from Google and doing evil things like this!
Is there evidence that the Apache project is "taking bribes from Google"?
My understanding from the article is that an individual contributed a patch to the the Apache httpd.conf source code and does not reflect the official viewpoint of the Apache Foundation, nor that the patch has been approved for inclusion. Naturally, I welcome any corrections.
South America, much like Africa, is relatively poorly served in regards to local datacenters and backbone network connectivity compared to North America, Europe, and East Asia.
There's a large market there, so it makes sense that companies would start investing in local resources to serve that market.
Now factor in DLC, in which they take their cut again and developers HAVE to use Steam for DLC when a game is sold on Steam. EA wanted to sell DLC directly to customers via an in-game store and Valve said no no no, we want a piece of that pie too.
Not exactly: DLC for Mass Effect 2 (an EA game) is not available through Steam and is only available through the BioWare website. That's no problem for Steam, as far as I can tell. Of course, they don't benefit from the auto-updates and other things which help out players.
The problem comes about when EA wanted to require Origin for ME3 -- Valve said "You can sell your game on Steam or you can use Origin. You can't have Steam install Origin." EA took their ball and went home, which annoyed a lot of customers (including myself).
Bethesda (and other game companies) seems to have no problem with using Steam for their game and DLC sales: Fallout 3 and Fallout New Vegas (and their DLC) were both sold on Steam and that worked out great. They even had a "game of the year" edition that included all the DLC for a slightly lower price than buying everything separately.
I like having everything all in one place and nicely organized. Steam does that for me and keeps everything up to date for me. Because EA insisted on doing things their way, I have to get the DLC for ME1/ME2 directly from their website and ME3 and it's DLC from Origin. This is annoying.
Steam actually mentions when third-party DRM is required. For example, with Anno 2070 they mention in the description "3rd-party DRM: Solidshield Tages SAS 3 machine activation limit". I intentionally avoid games that have this.
"Casual" in the "here's a DVD-R or ISO of the game disc and a keygen, just install it" sense.
Sure, various cracking groups have cracked Steam DRM for quite a few games. They always will, and I'm sure Valve (and others) take it into consideration. There's always some group of people who will never pay for anything, but there's nothing effective that one can do about that.
Steam provides a decently-priced integrated marketplace for games from a variety of producers. They have well-connected download servers all over the world (I often can download from Steam faster than I can from well-populated torrent swarms), and I can buy a game legitimately, download it, and have it up and running through Steam faster than I can get it illicitly. There's also no worries about unwanted malware with games from Steam -- that's always a concern from torrent sites.
Yes, the game being locked to one's Steam account can be a bit annoying, but I generally buy games and replay them for years rather than selling them so it's not a huge concern for me. Other than GOG.com (which tends to have more older games rather than new releases), Steam is pretty much the least-evil distribution system for games. I'd much rather deal with the relatively minor drawbacks of Steam (and benefit from its advantages) than deal with the crap from Ubisoft: Securom, StarForce, etc. are all horrible.
PC gamer has an article about the subject in which they quote Gabe Newell, CEO of Valve:
It seems other games companies who had tried to penetrate Russia hadn’t done so due to a the infamous levels of piracy. “The people who are telling you that Russians pirate everything are the people who wait six months to localize their product into Russia,” said Newell. “It doesn’t take much in terms of providing a better service to make pirates a non-issue.”
[...]
Of course, a lack of internet access could have contributed to Russia’s sky-high piracy levels, with cracked black-market DVDs offering many an offline gaming experience. But the inevitable, interminable spread of internet access across the country comes with improved anti-piracy measures on most titles.
This wasn’t Valve’s approach, though – it offered the carrot instead of the crowbar. “The easiest way to stop piracy is not by putting antipiracy technology to work,” Newell said. “It’s by giving those people a service that’s better than what they’re receiving from the pirates.”
DRM serves to inconvenience legitimate users and does little to stop pirates: all it takes is one smart cow to open the gate and all the other cows can follow.
Steam seems to provide a good service to game sellers and players: reasonable DRM to reduce casual piracy while not being hideously obnoxious (you only need to be online once to activate the game, after that you can play offline), fast downloads, decent anti-cheating protection for multiplayer games, frequent sales, millions of regular viewers (so promotions are more effective), automatic updates, very simple click-to-buy procedure without any hassle, etc. Why wouldn't game developers sell games on Steam rather than creating their own obnoxious systems?
Jordan is "anti-US"? Yeah, right next to Saudi Arabia there...
Indeed. Jordan is quite a US-friendly place.
I'm also curious to know why Jiro thinks that the Bulletin of the Atomic Scientists is a "left-wing" publication. I've heard that being mentioned recently, which makes me suspect that it's on some list of talking points.
Yes, but with excellent filtering and easy filter-training spam becomes less economical.
Take, for example, Gmail's spam filters: I receive thousands of spams per month (down from tens of thousands a month from year or two ago) at my personal address hosted on Google Apps. Out of all those messages, maybe one or two a month slip by the filters. I select the messages and click "mark as spam" and they're gone from my mailbox and help train the filter. This is trivial work for the user and benefits the entire community. Every single one of the "common" spams (e.g. pills, 419 scams, etc.) is caught -- the rare ones that slip by are using some new gimmick to elude filters. Once the filters are trained to detect them, that gimmick becomes useless.
I'm an exception as receive a huge amount of spam due to having my own domain and a very trivial, widely-published-on-the-internet email address. I suspect most users have far less pre-filter spam hitting their mailbox and even less making it through.
Same thing with blog spam: Akismet catches a similar ratio of spam, making blog spam pretty much useless. Bloggers can submit spam that was missed or innocent postings that were mistakenly flagged and the system learns, benefiting everyone. Training this filter is basically a one-click operation for site admins. I can always identify blogs that don't use Akismet because the comments at those sites are flooded with spam.
Yes, to truly be stopped spam needs to be stopped at the source. When all it takes to send out spam is a cheap, anonymous SIM card from a mobile ISP in Nigeria, it's unlikely that there's any practical means of stopping spam at its source. When there's dozens of sites scanning the internet to find lists of open proxies ripe for abuse, it's possible to send spam with essentially no risk. Using a stolen credit card to rent a VPS allows spammers to send mail from "legitimate" IP addresses at high rates of speed with very little in the way of information that traces back to them.
Still, having filtering "communities" like Gmail or Akismet or other similar services (I presume that Hotmail and Yahoo do something similar) can stop the huge majority of spam and make it less worthwhile for the spammers -- already it's at the point where essentially no legitimate service sends spam, unlike the situation a decade ago. Indeed, several articles I've read suggested that some spammers are migrating away from email spam to social networking spam on services like Facebook and Twitter as their filtering methods are less advanced.
I don't mind if they all use a compatible OTP system, so that I can just have the one Google Authenticator app for my iOS device (or a compatible J2ME program on my non-smartphone). The services that annoy me are the ones that use different methods that I can't integrate with code generating programs I already have.
The nice thing with TOTP/RFC 6238 is that it's an open standard and not subject to the whims of a particular company. It's also completely independent of third-parties: I can set up my own TOTP system on my own systems and not have it be dependent on the availability or security of any third party.
Yes, you can have freedom without net neutrality. We've had net freedom since forever, and no-one started talking about net neutrality until a couple of years ago.
Because net neutrality was the de facto standard for the internet up until a few years ago when certain providers thought they could make more money by penalizing their competitors (e.g. Comcast imposing bandwidth caps, but not counting their own streaming video service [ala Netflix] towards that cap).
I still have a Q6600 in my gaming system. It's a solid CPU. Replacing it would require replacing the motherboard and I can't really justify it at this point -- things run really well (I have a GeForce 550 Ti graphics card, which handles essentially the games I want to play, including modern ones, with aplomb).
Once I start running into performance issues, I may upgrade, but that'll probably be in another year or two.
This can still leak a lot of information that doesn't go through the SOCKS proxy.
I prefer OpenVPN: it's easy to setup on both the client and server, and it can proxy *all* traffic, even for applications that don't otherwise support proxying.
If it's protection from local snooping (e.g. wifi snoopers at the coffee shop), any provider will be reasonable, though it may be better to get a reasonably local provider for better performance. As you say, Australia is on the arse end of the planet and there's very long links to get to North America or Europe.
If you want to be able to access geographically-limited sites (e.g. Netflix in the US), again, any provider with endpoints there should be adequate.
If you want anonymity for Serious Purposes (e.g. whistleblowing, or any other activity where you or your family could be at risk), you'd probably be better off using Tor or some other system that doesn't require user registration. Of course, considerable amounts of services have ended up blocking Tor due to extensive abuse being emitted from their network, but that may be something you're willing to put up with. Pretty much any commercial VPN provider logs basic stuff about users (e.g. which user is assigned what IP address at what time) so they can shut down accounts being used for abuse.
IP addresses are less portable. One can easily point a domain name at a new IP address, but this is not as easily possible with just an IP address. They're also quite a bit less memorable.
What I meant by "dedicated communications networks" was "radios used only by emergency services for voice", in that they weren't using public mobile phone bands for police radios but rather had their own radio infrastructure. I apologize for the lack of clarity.
My point was that data networks may make things a bit more efficient, particularly with routine stuff like dispatching resources, looking up license plates, etc., all of that can be done over voice radios. Most of the police departments I'm familiar with in the US (I have a few cops in my extended family) have mobile data terminals connected to commercial mobile phone providers but the police voice radios are owned and operated by the department (or the city/county/state). The MDTs use a minimal amount of bandwidth, the providers already give them priority access in the event of emergencies, and all of what the MDTs do can be done over voice radios, so why would they be remotely concerned about mooching off of privately-owned wifi networks in an emergency?
Don't emergency services have their own dedicated communications networks?
Sure, most police departments that I've seen use mobile data terminals for dispatching calls but the dispatcher, officer, and car itself still have two-way radios. They can still communicate among themselves and dispatch calls using radios, albeit slightly less efficiently than they can with mobile data terminals.
In the US at least, emergency services have priority access to telecommunication networks like mobile and landline phone networks. So long as the network itself is intact (but merely overwhelmed by non-emergency calls) they will be able to get access.
Why would any emergency service worth their salt even remotely consider using home networks, particularly with no assurance of service quality or availability?
Lastly, what prevents bad guys from operating this "switch" to gain access to home networks? Even if they can't access the internal network itself, they'd be able to piggyback on the connection to browse the internet (likely for nefarious purposes if they're activating this sort of access switch).
Indeed. I do part-time abuse desk work for a medium-sized email provider. We have basic no-automated-signup methods, such as only allowing one new account per day per/24 (we're only doing IPv4 right now, but are working on transitioning to dual-stack) and a captcha. It's reasonably effective at stopping bot signups, though we do see a lot of spammers creating new accounts from a bunch of open proxies to avoid the subnet blocks.
It's clear that they're the same spammer, as they create accounts with distinctive patterns and their post-account-creation behavior is very similar. It's really annoying. You'd think they'd give up after a while as it's clear that we're onto them, as their accounts get nuked within minutes of creation before they can abuse them.
I've proposed doing something similar to IRC servers where our server checks the user's IP address to see if it's running any open proxies on common ports, then checks various public open proxy blacklists. Alas, developer resources are constrained and they're working on other stuff.
In regards to forums and blogs, I've found reasonably good luck with Akismet and the WordPress plugin "conditional captcha" -- all comments/posts are sent to Akismet to see if the message is spammy. If it is, the plugin comes into play and presents the user with a captcha. If they solve the captcha within 10 minutes, the message gets placed in the moderation queue for approval (I figure that if it's spammy enough to trip Akismet, it is likely to warrant human moderation). If the captcha isn't solved within that time period then the message is deleted. It keeps out a ton of spam and saves the administrator time from having to go through a queue with hundreds of messages. Over the last few years, this method has caught tens of thousands of spam comments on one of my blogs and only about 10 messages made it into the moderation queue (only one of which was actually legitimate). I'm sure a similar method can be implemented for web forums.
Slashdot Mirror
Good advice. In my case, Gandi is my registar and web host (I rather like their hosting plans, as they tend not to suck), but the DNS is provided by CloudFlare (who sits in front of the site as a reverse proxy). If needed, I can have CloudFlare point the DNS elsewhere until the S stops HTF.
I have some domains with NameCheap (which others have pointed out is quite good) and others with Gandi.net.
NameCheap has somewhat better pricing, but I prefer Gandi for most things. I rather like that Gandi's whois privacy is included by default in the cost of registration with compatible domains (.us, for example, does not allow private registration). The only public information they display is your full name and NIC handle -- the address and phone number are changed to that of their Paris headquarters and the email address is a spam-protected alias. I like that they display your full name, as their policy is that regardless of if you have the privacy service enabled or not your domain is still yours and you own it. Other registrars have the policy that they own (for as long as the whois privacy is enabled) your domain and are leasing it back to you.
If you're already with Gandi, check out their Simple Hosting. It's pretty slick, as far as basic hosting goes: you get your own Apache/MySQL/PHP processes, the web server runs with the same permissions as your user account (so setting up stuff like WordPress is trivial as there's no permissions-related issues), can host multiple separate sites on a single instance, etc.
Their VPSs are pretty standard paravirtualized Xen systems which work out pretty well (I ran a Team Fortress 2 gameserver for a while on one and it was stable and reliable).
As a domain customer you get a 50% discount code for the first year ($30/year rather than $60/year).
Disclaimer: Gandi customer, not employed by them in any way.
Also FYI, I typoed the URL for Google's ad preferences. Here is the correct URL: http://www.google.com/ads/preferences/ -- I left off the "s" at the end of "preferences". Mea culpa.
It's already starting to bother me. I'm seeing these advertisements here on Slashdot too. After I've searched for something on Google, the related advertisements start to come up EVERYWHERE on the internet. Seriously, they come after you. If you search for specific flights you start to see ads for that everyone. It'll haunt you and there's nothing you can do.
Not true: you can change your Google Ad Preferences or opt-out.
Similarly, you can use the NAI's opt-out page to opt-out of Google and other ad network tracking.
There's plenty of browser plugins that work to block ads entirely (such as AdBlock) and ones that ensure that the "opt-out" cookies stay in existence even if you clear your other cookies.
All the other browsers than Safari and IE are in bed with advertisers because both Firefox and Opera get revenue directly from Google.
The default search box in those browsers comes configured to use Google, yes. They do get income from ad revenue stemming from searches from the box. You're not forced to use that search box, nor are you forced to use the default settings -- you can add other search providers (like DuckDuckGo, ixquick, etc.) -- Firefox, for one, doesn't have ad agreements with anyone other than Google.
So for the love of god Apache Project, stop taking bribes from Google and doing evil things like this!
Is there evidence that the Apache project is "taking bribes from Google"?
My understanding from the article is that an individual contributed a patch to the the Apache httpd.conf source code and does not reflect the official viewpoint of the Apache Foundation, nor that the patch has been approved for inclusion. Naturally, I welcome any corrections.
South America, much like Africa, is relatively poorly served in regards to local datacenters and backbone network connectivity compared to North America, Europe, and East Asia.
There's a large market there, so it makes sense that companies would start investing in local resources to serve that market.
Now factor in DLC, in which they take their cut again and developers HAVE to use Steam for DLC when a game is sold on Steam. EA wanted to sell DLC directly to customers via an in-game store and Valve said no no no, we want a piece of that pie too.
Not exactly: DLC for Mass Effect 2 (an EA game) is not available through Steam and is only available through the BioWare website. That's no problem for Steam, as far as I can tell. Of course, they don't benefit from the auto-updates and other things which help out players.
The problem comes about when EA wanted to require Origin for ME3 -- Valve said "You can sell your game on Steam or you can use Origin. You can't have Steam install Origin." EA took their ball and went home, which annoyed a lot of customers (including myself).
Bethesda (and other game companies) seems to have no problem with using Steam for their game and DLC sales: Fallout 3 and Fallout New Vegas (and their DLC) were both sold on Steam and that worked out great. They even had a "game of the year" edition that included all the DLC for a slightly lower price than buying everything separately.
I like having everything all in one place and nicely organized. Steam does that for me and keeps everything up to date for me. Because EA insisted on doing things their way, I have to get the DLC for ME1/ME2 directly from their website and ME3 and it's DLC from Origin. This is annoying.
Steam actually mentions when third-party DRM is required. For example, with Anno 2070 they mention in the description "3rd-party DRM: Solidshield Tages SAS
3 machine activation limit". I intentionally avoid games that have this.
"Casual" in the "here's a DVD-R or ISO of the game disc and a keygen, just install it" sense.
Sure, various cracking groups have cracked Steam DRM for quite a few games. They always will, and I'm sure Valve (and others) take it into consideration. There's always some group of people who will never pay for anything, but there's nothing effective that one can do about that.
Steam provides a decently-priced integrated marketplace for games from a variety of producers. They have well-connected download servers all over the world (I often can download from Steam faster than I can from well-populated torrent swarms), and I can buy a game legitimately, download it, and have it up and running through Steam faster than I can get it illicitly. There's also no worries about unwanted malware with games from Steam -- that's always a concern from torrent sites.
Yes, the game being locked to one's Steam account can be a bit annoying, but I generally buy games and replay them for years rather than selling them so it's not a huge concern for me. Other than GOG.com (which tends to have more older games rather than new releases), Steam is pretty much the least-evil distribution system for games. I'd much rather deal with the relatively minor drawbacks of Steam (and benefit from its advantages) than deal with the crap from Ubisoft: Securom, StarForce, etc. are all horrible.
PC gamer has an article about the subject in which they quote Gabe Newell, CEO of Valve:
It seems other games companies who had tried to penetrate Russia hadn’t done so due to a the infamous levels of piracy. “The people who are telling you that Russians pirate everything are the people who wait six months to localize their product into Russia,” said Newell. “It doesn’t take much in terms of providing a better service to make pirates a non-issue.”
[...]
Of course, a lack of internet access could have contributed to Russia’s sky-high piracy levels, with cracked black-market DVDs offering many an offline gaming experience. But the inevitable, interminable spread of internet access across the country comes with improved anti-piracy measures on most titles.
This wasn’t Valve’s approach, though – it offered the carrot instead of the crowbar. “The easiest way to stop piracy is not by putting antipiracy technology to work,” Newell said. “It’s by giving those people a service that’s better than what they’re receiving from the pirates.”
DRM serves to inconvenience legitimate users and does little to stop pirates: all it takes is one smart cow to open the gate and all the other cows can follow.
Steam seems to provide a good service to game sellers and players: reasonable DRM to reduce casual piracy while not being hideously obnoxious (you only need to be online once to activate the game, after that you can play offline), fast downloads, decent anti-cheating protection for multiplayer games, frequent sales, millions of regular viewers (so promotions are more effective), automatic updates, very simple click-to-buy procedure without any hassle, etc. Why wouldn't game developers sell games on Steam rather than creating their own obnoxious systems?
Jordan is "anti-US"? Yeah, right next to Saudi Arabia there...
Indeed. Jordan is quite a US-friendly place.
I'm also curious to know why Jiro thinks that the Bulletin of the Atomic Scientists is a "left-wing" publication. I've heard that being mentioned recently, which makes me suspect that it's on some list of talking points.
Yes, but with excellent filtering and easy filter-training spam becomes less economical.
Take, for example, Gmail's spam filters: I receive thousands of spams per month (down from tens of thousands a month from year or two ago) at my personal address hosted on Google Apps. Out of all those messages, maybe one or two a month slip by the filters. I select the messages and click "mark as spam" and they're gone from my mailbox and help train the filter. This is trivial work for the user and benefits the entire community. Every single one of the "common" spams (e.g. pills, 419 scams, etc.) is caught -- the rare ones that slip by are using some new gimmick to elude filters. Once the filters are trained to detect them, that gimmick becomes useless.
I'm an exception as receive a huge amount of spam due to having my own domain and a very trivial, widely-published-on-the-internet email address. I suspect most users have far less pre-filter spam hitting their mailbox and even less making it through.
Same thing with blog spam: Akismet catches a similar ratio of spam, making blog spam pretty much useless. Bloggers can submit spam that was missed or innocent postings that were mistakenly flagged and the system learns, benefiting everyone. Training this filter is basically a one-click operation for site admins. I can always identify blogs that don't use Akismet because the comments at those sites are flooded with spam.
Yes, to truly be stopped spam needs to be stopped at the source. When all it takes to send out spam is a cheap, anonymous SIM card from a mobile ISP in Nigeria, it's unlikely that there's any practical means of stopping spam at its source. When there's dozens of sites scanning the internet to find lists of open proxies ripe for abuse, it's possible to send spam with essentially no risk. Using a stolen credit card to rent a VPS allows spammers to send mail from "legitimate" IP addresses at high rates of speed with very little in the way of information that traces back to them.
Still, having filtering "communities" like Gmail or Akismet or other similar services (I presume that Hotmail and Yahoo do something similar) can stop the huge majority of spam and make it less worthwhile for the spammers -- already it's at the point where essentially no legitimate service sends spam, unlike the situation a decade ago. Indeed, several articles I've read suggested that some spammers are migrating away from email spam to social networking spam on services like Facebook and Twitter as their filtering methods are less advanced.
woosh.
I don't mind if they all use a compatible OTP system, so that I can just have the one Google Authenticator app for my iOS device (or a compatible J2ME program on my non-smartphone). The services that annoy me are the ones that use different methods that I can't integrate with code generating programs I already have.
The nice thing with TOTP/RFC 6238 is that it's an open standard and not subject to the whims of a particular company. It's also completely independent of third-parties: I can set up my own TOTP system on my own systems and not have it be dependent on the availability or security of any third party.
Seconded. It's simple, easy to setup, and easy to integrate into a variety of services.
Yes, you can have freedom without net neutrality. We've had net freedom since forever, and no-one started talking about net neutrality until a couple of years ago.
Because net neutrality was the de facto standard for the internet up until a few years ago when certain providers thought they could make more money by penalizing their competitors (e.g. Comcast imposing bandwidth caps, but not counting their own streaming video service [ala Netflix] towards that cap).
I still have a Q6600 in my gaming system. It's a solid CPU. Replacing it would require replacing the motherboard and I can't really justify it at this point -- things run really well (I have a GeForce 550 Ti graphics card, which handles essentially the games I want to play, including modern ones, with aplomb).
Once I start running into performance issues, I may upgrade, but that'll probably be in another year or two.
This can still leak a lot of information that doesn't go through the SOCKS proxy.
I prefer OpenVPN: it's easy to setup on both the client and server, and it can proxy *all* traffic, even for applications that don't otherwise support proxying.
If it's protection from local snooping (e.g. wifi snoopers at the coffee shop), any provider will be reasonable, though it may be better to get a reasonably local provider for better performance. As you say, Australia is on the arse end of the planet and there's very long links to get to North America or Europe.
If you want to be able to access geographically-limited sites (e.g. Netflix in the US), again, any provider with endpoints there should be adequate.
If you want anonymity for Serious Purposes (e.g. whistleblowing, or any other activity where you or your family could be at risk), you'd probably be better off using Tor or some other system that doesn't require user registration. Of course, considerable amounts of services have ended up blocking Tor due to extensive abuse being emitted from their network, but that may be something you're willing to put up with. Pretty much any commercial VPN provider logs basic stuff about users (e.g. which user is assigned what IP address at what time) so they can shut down accounts being used for abuse.
IP addresses are less portable. One can easily point a domain name at a new IP address, but this is not as easily possible with just an IP address. They're also quite a bit less memorable.
It's not unheard of to run a website off a 10 Mbit line, but transferring 50 GB over a 10 Mbit line is going to take over 113 hours.
You're off by a factor of ten: 50GB / 10Mbps = 11.37 hours.
Still, point taken.
What I meant by "dedicated communications networks" was "radios used only by emergency services for voice", in that they weren't using public mobile phone bands for police radios but rather had their own radio infrastructure. I apologize for the lack of clarity.
My point was that data networks may make things a bit more efficient, particularly with routine stuff like dispatching resources, looking up license plates, etc., all of that can be done over voice radios. Most of the police departments I'm familiar with in the US (I have a few cops in my extended family) have mobile data terminals connected to commercial mobile phone providers but the police voice radios are owned and operated by the department (or the city/county/state). The MDTs use a minimal amount of bandwidth, the providers already give them priority access in the event of emergencies, and all of what the MDTs do can be done over voice radios, so why would they be remotely concerned about mooching off of privately-owned wifi networks in an emergency?
Don't emergency services have their own dedicated communications networks?
Sure, most police departments that I've seen use mobile data terminals for dispatching calls but the dispatcher, officer, and car itself still have two-way radios. They can still communicate among themselves and dispatch calls using radios, albeit slightly less efficiently than they can with mobile data terminals.
In the US at least, emergency services have priority access to telecommunication networks like mobile and landline phone networks. So long as the network itself is intact (but merely overwhelmed by non-emergency calls) they will be able to get access.
Why would any emergency service worth their salt even remotely consider using home networks, particularly with no assurance of service quality or availability?
Lastly, what prevents bad guys from operating this "switch" to gain access to home networks? Even if they can't access the internal network itself, they'd be able to piggyback on the connection to browse the internet (likely for nefarious purposes if they're activating this sort of access switch).
Indeed. I do part-time abuse desk work for a medium-sized email provider. We have basic no-automated-signup methods, such as only allowing one new account per day per /24 (we're only doing IPv4 right now, but are working on transitioning to dual-stack) and a captcha. It's reasonably effective at stopping bot signups, though we do see a lot of spammers creating new accounts from a bunch of open proxies to avoid the subnet blocks.
It's clear that they're the same spammer, as they create accounts with distinctive patterns and their post-account-creation behavior is very similar. It's really annoying. You'd think they'd give up after a while as it's clear that we're onto them, as their accounts get nuked within minutes of creation before they can abuse them.
I've proposed doing something similar to IRC servers where our server checks the user's IP address to see if it's running any open proxies on common ports, then checks various public open proxy blacklists. Alas, developer resources are constrained and they're working on other stuff.
In regards to forums and blogs, I've found reasonably good luck with Akismet and the WordPress plugin "conditional captcha" -- all comments/posts are sent to Akismet to see if the message is spammy. If it is, the plugin comes into play and presents the user with a captcha. If they solve the captcha within 10 minutes, the message gets placed in the moderation queue for approval (I figure that if it's spammy enough to trip Akismet, it is likely to warrant human moderation). If the captcha isn't solved within that time period then the message is deleted. It keeps out a ton of spam and saves the administrator time from having to go through a queue with hundreds of messages. Over the last few years, this method has caught tens of thousands of spam comments on one of my blogs and only about 10 messages made it into the moderation queue (only one of which was actually legitimate). I'm sure a similar method can be implemented for web forums.
The "g" stands for "generic", not "global".