Slashdot Mirror


User: Shakrai

Shakrai's activity in the archive.

Stories
0
Comments
12,853
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 12,853

  1. Re:Not really an issue on US Control of Internet Remains an Issue · · Score: 2, Insightful

    or some kind of line about how Libya is in charge of the UN council on human rights, whatever that has to do with it

    It has to do with the respectability of that institution in the eyes of a lot of people. To a lot of people the UN hasn't done anything morally just since the intervention in the Korean War. Many of us are leery of ceding control over anything to an organization that comes off as quite hypocritical in many areas (the human rights council being one of them).

    Now I'm not looking for a flame-fest over the pros and cons of the UN. Just trying to explain the viewpoints of a lot of people. Personally, I also think this is a manufactured crisis, because I have yet to see anything that ICANN or the United States has done to the internet that harmed the interests of the rest of the world. And please don't throw '.xxx' out there as your example, as there are many legitimate reasons not to setup yet another TLD and I'd hardly think that a disagreement over porn of all things should be something to get nation-states upset about.

  2. Re:Australia on US Control of Internet Remains an Issue · · Score: 4, Insightful

    It would certainly be nice for Australia to have some part of the infrastructure here

    Uhh, what do you think you are using to connect to the internet? Tin-cans with string run all the way back to the US and plugged into our infrastructure?

    Define infrastructure? Because there's an internet backbone in Australia. There's also at least three DNS root servers in Australia. What's the problem here?

  3. Re:packet masquerading on Vuze Petitions FCC To Restrict Traffic Throttling · · Score: 1

    -- but this hasn't happened yet in any of the mainstream p2p protocols that I'm aware of and there are other ways around it. The only ssh packets that I prioritize on my network are those

    ..... under 200 bytes in size, otherwise scp transfers get priority and kill my other interactive (other ssh sessions, telnet, voip, etc) stuff.

    Blah, used the < symbol and forgot to encode it for html ;)

  4. Re:packet masquerading on Vuze Petitions FCC To Restrict Traffic Throttling · · Score: 1

    If ssh packets are legally exempt from intentional delays, guess what protocol high-bandwidth users will disguise their packets as?

    *shrug*, you missed the point.

    A) Under a properly designed traffic shaping scheme there is no such thing as an "intentional delay". They might be delayed if the pipe is maxed out and something else (voip/ssh/etc) needs it. There's a world of difference between this and intentionally interfering with and/or slowing down p2p protocols. Also, no p2p user is going to notice any slowdown caused by prioritizing ssh or voip, because neither one of them uses that much bandwidth compared to p2p.

    B) Deep packet inspection would be able to tell the difference between actual ssh and just randomly encrypted data. Fire up wireshark if you don't believe me. It easily decodes ssh requests and responses. I suppose the p2p protocols could actually go so far as to adopt the ssh protocol and mask their traffic that way -- but this hasn't happened yet in any of the mainstream p2p protocols that I'm aware of and there are other ways around it. The only ssh packets that I prioritize on my network are those

    Next you'll say that the p2p application will just set it's MTU to a really small number. All the power to them, because this will slow their transfers down with IP and TCP overhead more then any traffic shaping would.

    Besides which, I don't think the goal of most p2p users is to monopolize their pipes. The goal is to get a decent amount of performance. Most people aren't going to get upset if their neighbors WoW or VoIP packets move ahead of their bittorrent packets in the queue during times when the pipe is maxed out. They are going to get upset if you take measures to slow them down/actively interfere with them regardless of how low or high the network utilization is.

  5. Re:I agree its wrong on Wi-Fi Piggybacking Widespread · · Score: 2, Interesting

    I'd do it if I could easily limit the bandwidth of unreckognized guests and guarantee theu were was isolated from my network

    There are ways to isolate them and limit their bandwidth. Whether or not you'd call them "easy" is up to you.

  6. Re:I agree its wrong on Wi-Fi Piggybacking Widespread · · Score: 4, Insightful

    broadcasting the SSID is offering access. the purpose of the SSID is to say "hey i'm here, connect to me"

    Actually, I thought the purpose of the SSID was to serve as the service set identifier to differentiate between networks. The SSID is also broadcast on an encrypted network, and anyone would agree that an encrypted network is not exactly saying "hey, I'm here, connect to me"

  7. Re:I agree its wrong on Wi-Fi Piggybacking Widespread · · Score: 4, Informative

    By that definition, my operating system is in violation of the law whenever it scans for an available network and presents it to me for connection.

    New York's definition is a lot better. Of course, I can't pull it up right now, because section of the Assembly site with our laws seems to be down, but it basically requires that you have to bypass a "password or code system" in order to commit the crime of "unauthorized use of a computer".

    That's actually quite logical. Connecting to an open wi-fi network is not a crime in New York State. Bypassing someones WEP key in order to use his wi-fi however, is.

  8. Re:Finally on Vuze Petitions FCC To Restrict Traffic Throttling · · Score: 1

    erhaps in an earlier paradigm, but there's really a small distinction today between "interactive stuff"

    Umm, it's not really that hard. Here's some of my iptables rules:

    Flag traffic from my T-Mobile UMA phone:
    iptables -t mangle -A POSTROUTING -o eth0 -s 172.25.42.100 -j TMOBILE

    Flag TCP ACKs:
    iptables -t mangle -A OUTPUT -o eth0 -m length --length 0:52 -p tcp --tcp-flags ALL ACK -j ACKS

    Flag SSH and telnet (only small packets so to avoid giving priority to scp transfers):
    iptables -t mangle -A OUTPUT -o eth0 -p tcp --sport 22:23 -m length --length 0:200 -j INTERACTIVE

    Flag NTP:
    iptables -t mangle -A OUTPUT -o eth0 -p udp --sport 123 -j INTERACTIVE
    iptables -t mangle -A OUTPUT -o eth0 -p udp --dport 123 -j INTERACTIVE

    Flag echos and echo-replies to ensure good ping times:
    iptables -t mangle -A OUTPUT -o eth0 -p icmp --icmp-type echo-request -m length --length 0:128 -j INTERACTIVE
    iptables -t mangle -A OUTPUT -o eth0 -p icmp --icmp-type echo-reply -m length -- length 0:128 -j INTERACTIVE

    All those targets are custom chains that classify the packets with --set-class, but the reality is you could do whatever you want with them. There's also -t mangle POSTROUTING equivs for all of these rules, to classify traffic being forwarded from my LAN.

    Now, I don't pretend that the above is a perfect solution. For one it would be open to abuse by somebody that went out of his way to make his bandwidth hogging application match my rules. But it would seem to me that a company with the resources of Comcast and access to routers that can do deep packet inspections could find a way to classify and prioritize most interactive traffic in a such way that wouldn't be disruptive to anyone.

    It shouldn't be able throttling a bandwidth heavy application down -- it should be about giving priority to more important traffic. Do you really think a bittorrent user or web surfer is going to notice his download rate drop by 1kb/s because his neighbor is pounding away in that ssh session? He will notice if you rate-limit that one protocol down to some arbitrary number even though the network isn't maxed out.

  9. Re:Fair trade on Vuze Petitions FCC To Restrict Traffic Throttling · · Score: 1

    Prevent shaping to offer quality-of-service tiers, provided that any data was eligible to travel on any tier if the customer wants to pay for it. I may want to pay for guaranteed low-latency and throughput of all traffic. My neighbor may want to pay for that service but only for traffic from YouTube.

    I agreed with everything else you wrote, but I'm somewhat leery of this. Offering different tiers at the ISP level is all well and good -- nobody is suggesting that the 1,500 byte packet from the $29.95/mo residential customer should have priority over the 1,500 byte packet from the $500/mo business customer. In the ideal world they would also give priority to interactive traffic (ssh/telnet packets, ntp packets, TCP ACKs, gaming, voip, etc) regardless of what tier the customer is in, as such traffic tends not to consume very much bandwidth at all and the extra costs to them are nil.

    It's when they start talking about having a end-to-end tier'ed internet that I start to have a serious problem with it. My ISP has the right, within reason, to ensure that everybody receives a fair slice of a limited (the last mile) resource. AT&T does not have the right (IMHO) to charge Google, Yahoo or Bob's website extra for the right of ensuring that their packets reach me in a timely manner.

    All packets should be treated equally on the internet backbone. And, ideally, barring natural disasters/war/acts of god/equipment failures/etc, QoS on the internet backbone should be a moot point, because the backbone links shouldn't reach 100% utilization.

  10. Re:The answering machine on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 1

    Err, I'll be damned. And I was already over on IMDB looking for the link to prove you wrong. I always swore that role was Tasha Yar......

    Learn something new everyday.

  11. Re:The answering machine on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 1

    Yeah, but she was pretty fucking cute ;)

    Err, wait, wrong link... *duck*

  12. Re:One wonders what we can ever do right on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 1, Offtopic

    But in a dog fight the F-15 does quite well

    Yes, the F-15 has never been defeated in air to air combat. It's also never faced an opponent remotely close to it's own technological level. Nor has it ever faced a foe as well trained as the typical American or Israeli pilot. The F-15 has been "defeated" during exercises with allied powers, flying planes that are it's equal in technology, with pilots as well trained as ours.

    Understand that I'm not bad mouthing it, because it's a beautiful and effective aircraft. I just don't think it's very fair to say it's never been shot down and use that as an example of how great American engineering is, when it's never faced a foe on equal terms.

  13. Re:One wonders what we can ever do right on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 1

    I think he meant in air to air combat, which is a true statement.

    Is it?

    (And that's not what he said anyway)

  14. Re:One wonders what we can ever do right on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 1

    My bad F-15.

    Wrong again. Two F-15Es were shot down by ground fire during the Gulf War.

    (I'm not trying to give you too much shit, and I generally agree with American engineering being among the best in the World, but our technology isn't invulnerable either...)

  15. Re:One wonders what we can ever do right on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 0, Offtopic

    How about the F16 it has never been shot down.

    Uhh, ya wanna rethink that?

  16. Re:The answering machine on New NSA-Approved Encryption Standard May Contain Backdoor · · Score: 4, Insightful

    Anyone else reminded of the little Black Box from Sneakers? The one that used a mathematical backdoor to break any encryption based on a certain algorithm that was only used in the USA?

    More to the point, anyone else remember the premise of that movie? That said black box was utterly useless for doing anything other then spying on Americans, which (prior to Dubya anyway) was outside of the NSAs mandate.

  17. Re:Government-granted monopoly leads to no alt. IS on Comcast Sued Over P2P Blocking · · Score: 1

    And the one your using now is called, what, exactly?

  18. Re:Prosecute them. on Wikileaks Releases Sensitive Guantanamo Manual · · Score: 1

    Why does death as a punishment only serve one purpose (revenge) but locking someone away in prison doesn't also serve that purpose? What purpose is served by locking someone away that isn't also served by the death penalty?

    The only "purpose" that I see that the death penalty doesn't provide for is rehabilitation. But in the case of a multiple murderer, rehabilitation isn't really a goal anyway, otherwise they wouldn't be serving life sentences with no parole.

    I'm sorry, I don't condone the death penalty simply because I'm afraid of innocents being executed. But I have no problem with it on a moral level for particularly heinous crimes like murder, treason and genocide. You call it revenge, I call it punishment.

  19. Re:Government-granted monopoly leads to no alt. IS on Comcast Sued Over P2P Blocking · · Score: 1

    Yeah, you said this in the last Comcast discussion, but you couldn't name a single ISP that actually does it as I recall.

    It's also common in PPP or PPPoE setups

    With whom? I've used PPPoE on two different DSL providers (Frontier and Verizon) and neither of them did it. As far as AOL goes, can anybody actually confirm it? My last experiences with them they gave out globally valid IPs for internet activities. Does anybody even use AOL anymore? ;)

    Name me a halfway mainstream ISP that does this. Until then I'm calling bullshit.

  20. Re:About time on Comcast Sued Over P2P Blocking · · Score: 1

    While I agree that it sucks they are blocking P2P apps, I will have to admit their service is pretty darn good. A few case examples:

    Your story is pretty interesting and an example of how customer service is SUPPOSED to work, but you'd have to agree that a story like yours seems to be the exception and not the rule with Comcast. The overwhelming majority of posts about them seem to be negative. While I'm sure there's some selection bias there (people with bad stories are more likely to write about them), I'd still tend to think that by and large their customer service stinks. Hell, I can think of one little old lady with a claw hammer that would probably agree with me ;)

    I subscribe to their 8Mb/768Kb plan and consistantly receive 8Mb plus transfer rates. The Speedboost to 16Mb is AMAZING! I purchased TF2 over Steam and started the 7GB download. To my suprise I was receiving it at 1.5MB-2.0MB/sec and it was completed in 60min!!!! The same couldn't be said for ATT's DSL.

    Is that speed really worth the money though? I have two broadband options: 1.5/384 from Verizon or 5.0/384 from Time Warner. I'll admit that the 5.0 is nice, but I never had a problem when I had DSL. Anything I wanted to download that was large enough that the bandwidth difference actually mattered was usually not something that I wanted right away anyway. And as far as torrents go, it's a pretty rare torrent that gets over 2.0 - 2.5 in my experience (not counting heavily seeded stuff like Linux ISOs).

    What I'd really like is more upload. If my loop length was 2,000 feet shorter I'd qualify for Verizon's 3.0/768 tier. I'd take that in a heartbeat over TW's 5.0/384.

  21. Re:Government-granted monopoly leads to no alt. IS on Comcast Sued Over P2P Blocking · · Score: 2, Informative

    Seriously?

    He made this same argument in another story about Comcast and stopped replying to posts when people asked him to name a few ISPs that do this.

    While I'm sure there are small remote ISPs that NAT their customers by default (and by remote I mean remote... think Alaskan wilderness), it's not even close to being a standard practice in the United States and the number of people affected by it are so small that it hardly bears mentioning.

    A few people have claimed that AOL does it. They didn't used to (over a decade ago I had them... always had globally valid IPs when I went outside of AOL and used internet apps), but it might have changed for all I know. In any case, I'd hardly call AOL an "ISP".

  22. Re:About time on Comcast Sued Over P2P Blocking · · Score: 1

    I've instead coughed up the small fee to use dyndns's mailhop relaying on nonstandard ports. Yay! Reliable mail again!

    Can you provide any information on that service? Link, costs, how well it works, etc?

    I used to host my own mail when I had service with a local WISP (fixed ip, no blocked ports), but now I'm stuck on a cable provider with a dynamic IP that filters port 25, so I'm forced to pay someone to host my e-mail for me.

  23. Re:Government-granted monopoly leads to no alt. IS on Comcast Sued Over P2P Blocking · · Score: 1

    Then you don't get out much. I take it you've never used AOL as a primary ISP?

    Uhh, not in a LONG time (read: over a decade ago), but when I did use them I was using IRC. I had a globally valid IP address that could accept incoming (DCC chat/DCC send/IM file transfers/etc) connections.

    Did that change?

  24. Re:Pay to steal on Comcast Sued Over P2P Blocking · · Score: 1

    In most cases, my guess is these people just don't have the disposable income to pay for music and games over and above the hardware they bought.

    Well, yes and no. If a friend of mine suggests that I try this really cool game, I'm probably going to go download it somewhere.

    If I really like the game, then I'm likely to find the cash to go buy it. Then I have a real CD, a game manual, a box, etc, etc. Off the top of my head, one player games that I didn't need to buy but bought anyway: Civ2, Hearts of Iron, Worms World Party and Defcon.

    I realize this is the tired old "I just wanna try it before I buy it" excuse, and it probably doesn't hold water because I'm not going to fork over the cash unless I really like it. If it holds my attention for two days and then gets deleted I'm not dropping cash on it. If I fall in love with the game, spend hours and days playing it and forget about my friends..... yeah I'll fork over $30-$50 for it. And sometimes I've even bought follow-up games without even pirating them first.... like the entire Civilization series and Hearts of Iron 2.

  25. Re:About time on Comcast Sued Over P2P Blocking · · Score: 4, Insightful

    I take it you didn't read the fine print.

    Show me where in the fine print it says they have the right to perform a man-in-the-middle attack on my communications. In fact, it's even more ironic, because their AUP doubtless has the same clause that my ISPs AUP has: You will not forge any IP header or datagram to make it appear as though it came from someone else.

    There are any of number of solutions to the problem of p2p traffic they could have taken. Like traffic shaping, QoS prioritization, canceling the accounts of massive bandwidth users, etc, etc. They crossed the line when they started forging packets in an attempt to disrupt communications.