NNSquad - Network Neutrality Squad
[ NNSquad ] Re: ISPs and the "Secret Service"
All, Can we get beyond the politics and policy issues? If Brett, or any other ISP, wants to fully disclose how they treat end-user traffic, to include the specific metrics and policies that are enforced (policy maps if it is Cisco gear, or equivalent configuration for other gear) then it is all welcome. That is, after all, exactly what the goal of this group is. We are not here to pass judgment on a particular policy or, to be more exact, device configuration, rather we are here to discover, and document, exactly what ISP's are doing to our packets. Are ISP's treating certain types of traffic differently? The end-users certainly don't know. Are they treating all traffic the same? As I expressed before, I personally would not have a problem with treating certain types of traffic differently. BGP traffic SHOULD have a higher priority than "normal" user traffic. End users SHOULD have the ability to tag traffic of a certain type as more important than others (VoIP traffic should have a higher precedence than web (HTTP/HTTPS) traffic, which should have a higher precedence than P2P traffic). These are all policy issues that are outside of the mandate of this group. However, these are all my opinions, and revolve around how traffic should be treated. If we knew in detail how ISP's prioritized their traffic it would alleviate the need for this group completely. Sadly, that is not the case. So I call out to any ISP's to share their detailed information as to how they treat customer traffic. I realize there will be a great reluctance to share this information, but the call must still be made. I await with baited breath a response... Thanks, Fred Reimer, CISSP, CCNP, CQS-VPN, CQS-ISS Senior Network Engineer Coleman Technologies, Inc. 954-298-1697 > -----Original Message----- > From: nnsquad-bounces+freimer=ctiusa.com@nnsquad.org > [mailto:nnsquad-bounces+freimer=ctiusa.com@nnsquad.org] On > Behalf Of Brett Glass > Sent: Saturday, November 24, 2007 9:39 PM > To: Lauren Weinstein; nnsquad@nnsquad.org > Cc: lauren@vortex.com > Subject: [ NNSquad ] Re: ISPs and the "Secret Service" > > > [ I'm going to run this one in the interests of fair play > since I did quote > from Brett's item that didn't go out on the list. But > responses should go > direct to Brett and/or to the NNSquad Forum. As far as > Brett's > specific comments below are concerned, I'll stand on my > previous > statements. > > -- Lauren Weinstein > NNSquad Moderator ] > > > At 07:11 PM 11/24/2007, Lauren Weinstein wrote: > >Greetings. I'm not sending to the list several messages > that > >threatened to spin us off into Neverland again, despite an > amusing > >anecdote from Brett Glass, where he speculates that he was > apparently > >being mischaracterized as a likely P2P user (!) which > caused his > >attempts to do a large download of GPS software via > RoadRunner to > >fail repeatedly. He ended up having to go to a cybercafe, > and since > >their router was having problems he updated its firmware > and finally > >got his download. A happy ending. > > The details of the story would probably interest the list. I > think > that Lauren should have posted it. If anyone wants to see > the message, > please e-mail me privately. > > >Apropos, this might be a good time to note again a key > reason why I > >instigated this project in the first place. In many ways, > most > >consumer and low-end business Internet access packages are > something > >of "secret services" from the standpoint of individual > customers. > > I firmly believe that they are "secret" only to the extent > that the > providers feel that it is necessary to retard the "arms > race" between > writers of abusive software and themselves. > > >To an extent that is really quite remarkable, people really > don't know > >what they're getting for their money, > > They do. They're paying the ISP to exercise its judgment, > not to tell > them every detail of the ways in which they are fighting the > latest > worm, P2P software, or other malicious creation. > > >and most ISPs seem perfectly > >happy to let their subscribers assume that any observed > service > >oddities are due to problems in subscribers' hardware or > software, > >not related to how the ISP network is provisioned. > > Actually, most of the time they are. We know, because we get > called. > We've helped users to remove nasty Trojan horses like > "Storm" more and > more. > > >Note that Internet access services at this level are sold > almost > >entirely based on theoretical speed claims, with usually > nary a > >mention of traffic shaping, throttling, blocking, jitter, > warping, > >morphing, or other "active" ISP data management procedures > that > >really can have dramatic impacts on end-user applications > >performance. > > I disagree. There is a mention of these things in the fine > print in > the ad. Granted, it'd be nice to see it in the large print, > but many > if not most customers would misinterpret it as meaning that > the > service was inferior rather than superior. > > >The specific decisions of how these various actions will be > applied > >to customer circuits are generally made by (from the > customer point > >of view) faceless entities deep in giant corporations (yes, > Brett, I > >know *you* are not a faceless entity in a giant > corporation, but > >you're the exception). > > I do not believe that I'm an exception. System > administrators are real > people with a real problem: Stopping abuse whilst making > their > companies' customers happy. > > I also believe that Lauren, consciously or not, may have > withheld > my posting so as to deny those administrators a human face > and preserve > the illusion that such measures are undertaken by faceless > corporations > rather than real people trying to do the right thing. > > >Typically, only the most general description > >of such activities will be buried in the Terms of Service, > and ISPs > >often consider the details to be proprietary. Subscribers > pay > >anyway, because (1) they don't really know what's going on > in these > >respects and (2) they don't have much choice anyway. > > > >In Brett's case, was the behavior he saw the result of > purposeful > >decisions by RoadRunner, or was a misconfiguration or other > technical > >problem to blame? Hard to really know for sure, and trying > to dig > >out info like that could easily become a long-term hobby > for the > >average consumer. > > Actually, it's more a pursuit for those who want to defeat > it. > > >We can argue forever (but not on the list!) about which (if > any) of > >these data "management" procedures are appropriate and > reasonable. > >But my take is that keeping them secret is not acceptable. > > Unfortunately, there is no way to expose them without > tipping the > playing field strongly toward those who would degrade the > service. > > --Brett Glass
Attachment:
smime.p7s
Description: S/MIME cryptographic signature