You know, I recall Netgear has had a Linux-based "Open Source" router (In fact, this very model IIRC.) for at least a year or two.
But there was little to no support for it, no community, and no information. It seemed then to be just lip service to the whole open source concept. Not sure what's different now.'
Given my past experience with Netgear hardware (I was a fan for a while, but I have since seen the light), I'll stick with Buffalo, or while Buffalo is dealing with patent trolls, Linksys.
Simple, don't do the DPI. Sandvine's appliance doesn't, it just does a basic traffic analysis. Not even the "more than N destinations" criteria mentioned in another response to your post - This is why there was a lot of collateral damage, such as some people being unable to send Lotus Notes mail with attachments larger than a certain size - Attempts to send such a message would trigger the Sandvine RST response 100% of the time, and unlike BT, Notes would have to restart from the beginning of the message.
Also, it's likely easier to have a device that monitors traffic only and analyzes it, sometimes injecting crap into the network in response with some not-too-stringent latency requirements, than to have a device that has to be inline and make drop/no-drop decisions with minimum latency.
Actually, encrypted or not, the way the Sandvine (I think that was the name?) system used by Comcast worked was it just did a traffic analysis - If your upload connection was more than X% saturated for N seconds, the Sandvine appliance would start spoofed RST injection to kill off connections. The only way around this would be a full blown VPN that used an encrypted transport layer. (Encrypted BitTorrent, SSH, and nearly all encrypted protocols except the various VPN systems are an encrypted application stream over an unencrypted TCP session. Even some VPNs use an unencrypted TCP session to tunnel through, making them vulnerable to RST injection.)
"The thing most people don't understand about cable modem is that it has virtually no buffer for outbound traffic (e.g. the traffic you do control) so subsequently it is almost a given that you will overrun the transmit buffer on your cable modem doing the simplest of things. This in turn will destroy all your throughput because..."
Actually this is not true. Cable modems are notorious for having large transmit buffers and slow transmit speed.
The end result is that when you saturate your connection, the problem is not necessarily packet loss, but latency goes through the roof (because stuff is staying in that huge buffer).
This kills your download speeds in addition to your uploads because ACK packets get delayed in the buffer.
This is exactly how I have my Buffalo WHR-G54S (not available currently due to patent trolls) running DD-WRT set up.
The active connections and connection tracking timeout values are critical if you're torrenting - Without them BT will kill the router. With them, the router handles BT easily.
Similarly, packet prioritization at the router doesn't help unless you cap bandwidth at something somewhat below your connection's maximum, like you did.
To the original person that asked the question - Unless you do extreme customization, there is never any need to touch the commandline with many of the Linux-based router packages. DD-WRT is easy to install and the configuration interface is 100% web-based unless you do some really crazy and nonstandard stuff.
Yup. I wish Facebook would implement some way of performing metadata sharing with external photo sites. I can understand not wanting to store the originals of images unless users paid (and I don't think anyone would pay for photo hosting from Facebook...), but it would be nice if there could be a way for a photo posted on an external site (like Flickr or SmugMug) could appear on a person's Facebook profile *in an integrated manner*.
There are Facebook apps that put galleries from other sites (flickr, SmugMug, etc.) onto a user's Facebook profile, but they are not integrated with FB's photo system at all, namely the ability to tag people in photos.
I'd need to test it again, but I'm fairly certain FB had a function that let you share albums with non-users by having FB generate a special link you'd give to the user.
Sadly, Mythic has a pretty bad track record here. With DAoC, a given engine (most expansions came with an improved graphics engine) would not be supported by WINE or even Cedega until it was 1-2 expansions out of date.
I think most DAoC fans will agree that the basic fundamentals of keeps, RAs, and such were pretty good.
The problem was that Mythic screwed up some of the little details, in some cases in ways that killed the playerbase before Mythic fixed it. For the most part, it wasn't any particular part of RvR design even - it was a PvE expansion.
For those who played DAoC, I'm sure I don't need to say more and you know exactly which one I'm referring to.
For everyone else - Mythic's second expansion was a pure PvE grinding expansion that included some of Everquest's worst aspects (camping mobs with multi-hour respawn times for good loot), but it had MASSIVE impacts on Realm vs. Realm warfare. This combination led to a LOT of players quitting the game. Mythic eventually fixed ToA, but by then it was too late - the playerbase was gone.
I am praying that WAR doesn't get FUBARed by EA (like every other MMO EA has been involved in).
DAoC had a lot of good aspects of its fundamental design, but a few screwups (namely Trials of Atlantis) were able to kill the playerbase before Mythic could fix them.
Many friends of mine and I were hoping for WAR = DAoC V2, but now that EA is involved it's far more likely that WAR = crap.:(
Hopefully I'll be wrong, but EA's MMO track record is awful.
In addition - In most cases the suspicion is not that someone randomly walked in and tampered with the election, but that an election official (such as the person who submitted this to Ask Slashdot) tampered with the election.
Sad to say it to the submitter, but for many people "Republican" and "E-Voting" instantly casts suspicions of tampering for a wide variety of reasons, including but not limited to significant monetary connections between Diebold (now PES) and the Republican party and a claim from someone from Diebold that they would "deliver the election" or something like that.
So, sadly, the question as a Republican election coordinator is not "How do I prevent tampering", but it is "How can I prove that the election was not tampered with" and "How can I prove that *I* did not tamper with the election"
How? I'm not sure, but based on other requirements a voter-verifiable paper trail is the first step, and not using Diebold/PES is the second (although there is lots of evidence that most other E-Voting vendors are not giving any thought to security whatsoever.)
You don't exactly provide details of what apps you're using.
In most if not all cases, you will find applications that are as good as if not better than the PalmOS version.
TomTom - Exists for PalmOS and WM. In my opinion the WM version seems to work MUCH better and is far easier to set up. TCPMP media player - Exists for both platforms, has somewhat extended codec support compared to PalmOS on a Windows Mobile device Web browser - The PalmOS web browser Just Plain Sucks in every way possible. Even Pocket IE is better and it isn't that hot (there are other options for WM) Java - The Java environment for PalmOS is utter and total crap and when I tried it on my Treo 650 not a single app I tried would work. GMail's Java app works great on my Tilt Google Maps - Don't think there's a PalmOS client that can come anywhere close to Google's Windows Mobile native app Instant Messaging - Half of the PalmOS IM app vendors seem to have gone out of business or stopped supporting the app. I've found FAR more choices for WM.
OK, yeah, there's PalmOS too but I don't count it any more since it's painful and users/developers are leaving it in droves because it and the devices it runs on are obsolete.
(I'm a former Treo 650 owner. My Tilt is such a breath of fresh air after that limited and buggy POS.)
As to Linux - Believe it or not most of the existing Linux-on-mobile-phone deployments are the MOST locked down (hence my comment about Android catering to the Lords of Lockdown). See MOTOMAGX for an example.
Actually I know quite a few people who love Windows Mobile, including myself.
Many of those people are hardcore Linux users on the desktop, too.
The iPhone is a toy. It's shiny and cool but it isn't very flexible. My AT&T Tilt blows it out of the water in every aspect except user interface, and the UI of the Tilt is good enough for me, especially considering the significantly better functionality.
Android looks like it's going to cater to the Lords of Lockdown (carriers).
It's really sad that the most open mobile phone platform out there is Windows Mobile. Everything else is a nightmare of signed applications and lockdown.
(Yes, Windows Mobile has application signing, but every WM device I know uses this for warning purposes only, not lockout. In addition, WM will remember when you say "yes, I want to run this unsigned app" and not bother you again.)
Works great for me. You probably didn't set it up properly.
Giving VOIP top priority won't do squat unless you prevent your cable modem from buffering data. To do this, you need to set DD-WRT to throttle traffic to about 10% below the upstream and downstream of the router cable modem. (e.g. if it can do 512 kbps up, throttle to around 450.)
It's interesting that autos have an expensive price premium in Europe - In the U.S. it has gotten to the point where a manual commands a significant price premium, if you can even get it.
Most of the studies on algae as fuel have focused on the fact that since it needs to be grown in a special structure and is effectively going to be hydroponically grown, it can be grown anywhere.
Where is the cheapest land? Sunny deserts where almost nothing can grow (hence no farming) and hence not much is growing. I know the UNH algal biodiesel studies have said that something like only 1/4 of the Sonora Desert would be enough to fuel the entire country's transportation needs.
Nokia has almost no cdmaOne/CDMA2000 phones. They're a GSM/UMTS shop.
In addition, previously only Sprint had an ESN whitelist that would prevent them from activating phones not sold by them. I have heard that Verizon also does this now in order to prevent feature-superior uncrippled phones from using their network (bypassing the overpriced replacements for the features removed from Verizon's version of said phones - see Treo 650 Bluetooth DUN as an example.)
You know, I recall Netgear has had a Linux-based "Open Source" router (In fact, this very model IIRC.) for at least a year or two.
But there was little to no support for it, no community, and no information. It seemed then to be just lip service to the whole open source concept. Not sure what's different now.'
Given my past experience with Netgear hardware (I was a fan for a while, but I have since seen the light), I'll stick with Buffalo, or while Buffalo is dealing with patent trolls, Linksys.
I believe this has been the basis of 1-2 lawsuit attempts against Comcast.
Simple, don't do the DPI. Sandvine's appliance doesn't, it just does a basic traffic analysis. Not even the "more than N destinations" criteria mentioned in another response to your post - This is why there was a lot of collateral damage, such as some people being unable to send Lotus Notes mail with attachments larger than a certain size - Attempts to send such a message would trigger the Sandvine RST response 100% of the time, and unlike BT, Notes would have to restart from the beginning of the message.
Also, it's likely easier to have a device that monitors traffic only and analyzes it, sometimes injecting crap into the network in response with some not-too-stringent latency requirements, than to have a device that has to be inline and make drop/no-drop decisions with minimum latency.
Actually, encrypted or not, the way the Sandvine (I think that was the name?) system used by Comcast worked was it just did a traffic analysis - If your upload connection was more than X% saturated for N seconds, the Sandvine appliance would start spoofed RST injection to kill off connections. The only way around this would be a full blown VPN that used an encrypted transport layer. (Encrypted BitTorrent, SSH, and nearly all encrypted protocols except the various VPN systems are an encrypted application stream over an unencrypted TCP session. Even some VPNs use an unencrypted TCP session to tunnel through, making them vulnerable to RST injection.)
"The thing most people don't understand about cable modem is that it has virtually no buffer for outbound traffic (e.g. the traffic you do control) so subsequently it is almost a given that you will overrun the transmit buffer on your cable modem doing the simplest of things. This in turn will destroy all your throughput because..."
Actually this is not true. Cable modems are notorious for having large transmit buffers and slow transmit speed.
The end result is that when you saturate your connection, the problem is not necessarily packet loss, but latency goes through the roof (because stuff is staying in that huge buffer).
This kills your download speeds in addition to your uploads because ACK packets get delayed in the buffer.
This is exactly how I have my Buffalo WHR-G54S (not available currently due to patent trolls) running DD-WRT set up.
The active connections and connection tracking timeout values are critical if you're torrenting - Without them BT will kill the router. With them, the router handles BT easily.
Similarly, packet prioritization at the router doesn't help unless you cap bandwidth at something somewhat below your connection's maximum, like you did.
To the original person that asked the question - Unless you do extreme customization, there is never any need to touch the commandline with many of the Linux-based router packages. DD-WRT is easy to install and the configuration interface is 100% web-based unless you do some really crazy and nonstandard stuff.
Yup. I wish Facebook would implement some way of performing metadata sharing with external photo sites. I can understand not wanting to store the originals of images unless users paid (and I don't think anyone would pay for photo hosting from Facebook...), but it would be nice if there could be a way for a photo posted on an external site (like Flickr or SmugMug) could appear on a person's Facebook profile *in an integrated manner*.
There are Facebook apps that put galleries from other sites (flickr, SmugMug, etc.) onto a user's Facebook profile, but they are not integrated with FB's photo system at all, namely the ability to tag people in photos.
I'd need to test it again, but I'm fairly certain FB had a function that let you share albums with non-users by having FB generate a special link you'd give to the user.
#3 will probably be the most convincing.
Sadly, Mythic has a pretty bad track record here. With DAoC, a given engine (most expansions came with an improved graphics engine) would not be supported by WINE or even Cedega until it was 1-2 expansions out of date.
I think most DAoC fans will agree that the basic fundamentals of keeps, RAs, and such were pretty good.
The problem was that Mythic screwed up some of the little details, in some cases in ways that killed the playerbase before Mythic fixed it. For the most part, it wasn't any particular part of RvR design even - it was a PvE expansion.
For those who played DAoC, I'm sure I don't need to say more and you know exactly which one I'm referring to.
For everyone else - Mythic's second expansion was a pure PvE grinding expansion that included some of Everquest's worst aspects (camping mobs with multi-hour respawn times for good loot), but it had MASSIVE impacts on Realm vs. Realm warfare. This combination led to a LOT of players quitting the game. Mythic eventually fixed ToA, but by then it was too late - the playerbase was gone.
I am praying that WAR doesn't get FUBARed by EA (like every other MMO EA has been involved in).
:(
DAoC had a lot of good aspects of its fundamental design, but a few screwups (namely Trials of Atlantis) were able to kill the playerbase before Mythic could fix them.
Many friends of mine and I were hoping for WAR = DAoC V2, but now that EA is involved it's far more likely that WAR = crap.
Hopefully I'll be wrong, but EA's MMO track record is awful.
In addition - In most cases the suspicion is not that someone randomly walked in and tampered with the election, but that an election official (such as the person who submitted this to Ask Slashdot) tampered with the election.
Sad to say it to the submitter, but for many people "Republican" and "E-Voting" instantly casts suspicions of tampering for a wide variety of reasons, including but not limited to significant monetary connections between Diebold (now PES) and the Republican party and a claim from someone from Diebold that they would "deliver the election" or something like that.
So, sadly, the question as a Republican election coordinator is not
"How do I prevent tampering", but it is
"How can I prove that the election was not tampered with" and "How can I prove that *I* did not tamper with the election"
How? I'm not sure, but based on other requirements a voter-verifiable paper trail is the first step, and not using Diebold/PES is the second (although there is lots of evidence that most other E-Voting vendors are not giving any thought to security whatsoever.)
You don't exactly provide details of what apps you're using.
In most if not all cases, you will find applications that are as good as if not better than the PalmOS version.
TomTom - Exists for PalmOS and WM. In my opinion the WM version seems to work MUCH better and is far easier to set up.
TCPMP media player - Exists for both platforms, has somewhat extended codec support compared to PalmOS on a Windows Mobile device
Web browser - The PalmOS web browser Just Plain Sucks in every way possible. Even Pocket IE is better and it isn't that hot (there are other options for WM)
Java - The Java environment for PalmOS is utter and total crap and when I tried it on my Treo 650 not a single app I tried would work. GMail's Java app works great on my Tilt
Google Maps - Don't think there's a PalmOS client that can come anywhere close to Google's Windows Mobile native app
Instant Messaging - Half of the PalmOS IM app vendors seem to have gone out of business or stopped supporting the app. I've found FAR more choices for WM.
From what I've heard, a lot of Symbian phones do not allow you to disable the signing check.
Whereas I have yet to see a WM device that had forced signing checks.
OK, yeah, there's PalmOS too but I don't count it any more since it's painful and users/developers are leaving it in droves because it and the devices it runs on are obsolete.
(I'm a former Treo 650 owner. My Tilt is such a breath of fresh air after that limited and buggy POS.)
As to Linux - Believe it or not most of the existing Linux-on-mobile-phone deployments are the MOST locked down (hence my comment about Android catering to the Lords of Lockdown). See MOTOMAGX for an example.
Yes and Yes.
Plus I can tether it via USB which is much faster. (Bluetooth 2.0+EDR is a bottleneck in HSDPA service areas.)
Actually I know quite a few people who love Windows Mobile, including myself.
Many of those people are hardcore Linux users on the desktop, too.
The iPhone is a toy. It's shiny and cool but it isn't very flexible. My AT&T Tilt blows it out of the water in every aspect except user interface, and the UI of the Tilt is good enough for me, especially considering the significantly better functionality.
Android looks like it's going to cater to the Lords of Lockdown (carriers).
It's really sad that the most open mobile phone platform out there is Windows Mobile. Everything else is a nightmare of signed applications and lockdown.
(Yes, Windows Mobile has application signing, but every WM device I know uses this for warning purposes only, not lockout. In addition, WM will remember when you say "yes, I want to run this unsigned app" and not bother you again.)
Plenty of ways to do that already, nothing new...
Works great for me. You probably didn't set it up properly.
Giving VOIP top priority won't do squat unless you prevent your cable modem from buffering data. To do this, you need to set DD-WRT to throttle traffic to about 10% below the upstream and downstream of the router cable modem. (e.g. if it can do 512 kbps up, throttle to around 450.)
It's interesting that autos have an expensive price premium in Europe - In the U.S. it has gotten to the point where a manual commands a significant price premium, if you can even get it.
The Sonora Desert is close to a lot of populated areas.
The problem is not that there is no water available, it is that it is all runoff/waste from those areas.
The theory is that algae can grow in this "grey water" that has already been used.
Clearing what land?
Most of the studies on algae as fuel have focused on the fact that since it needs to be grown in a special structure and is effectively going to be hydroponically grown, it can be grown anywhere.
Where is the cheapest land? Sunny deserts where almost nothing can grow (hence no farming) and hence not much is growing. I know the UNH algal biodiesel studies have said that something like only 1/4 of the Sonora Desert would be enough to fuel the entire country's transportation needs.
Nokia has almost no cdmaOne/CDMA2000 phones. They're a GSM/UMTS shop.
In addition, previously only Sprint had an ESN whitelist that would prevent them from activating phones not sold by them. I have heard that Verizon also does this now in order to prevent feature-superior uncrippled phones from using their network (bypassing the overpriced replacements for the features removed from Verizon's version of said phones - see Treo 650 Bluetooth DUN as an example.)