Re: Verizon Public Policy on Netflix
On Fri, Jul 11, 2014 at 09:50:22AM -0700, Owen DeLong wrote: [...] I'm always surprised that folks at smaller exchanges don't form consortiums to build a mutually beneficial transit AS that connects to a larger remote exchange. In my experience, the price of buying transit from established players has always been close to the combined price of buying a circuit and establishing some form of presence at a remote exchange. Close enough that everyone was willing to just pay for transit without the added administrative overhead of the transit consortium. I've seen such transit consortiums that pretend to be exchange points as well -- but that's a slightly different beast. I've also seen where the folks that should peer don't because they all have mutual transit providers, and the cost of interconnection is higher than the incremental transit costs for their cross-ASN traffic. You can't argue increased route splay when the circuit costs dominate the equation. Internet in the hinterlands is a tough ride compared to fiber-rich areas... But it keeps getting better, so there is hope.
Re: Verizon Public Policy on Netflix
On Jul 11, 2014, at 10:31 PM, Owen DeLong o...@delong.com wrote: On Jul 11, 2014, at 8:18 PM, Randy Bush ra...@psg.com wrote: And, for the record, it's pretty widely acknowledge that The World (Barry Shein) was the world's first commercial ISP - offering shell access in 1989, and at some point started offering PPP dial-up services. As I recall, they were a UUnet POP. yep. and uunet and psi were hallucinations. can we please not rewrite well-known history? or are you equating shell access with isp? that would be novel. unix shell != internet. btw, not do denigrate what barry did. a commercial unix bbs connected to the real internet was significant. the left coasties were doing free stuff, the well, community memory, ... and barry created a viable bbs commercial service which still survives (i presume). a significant achievement. randy Not to take away from Barry, but around that same time, some of us left coasts were also helping to build Netcom as a viable commercial entity providing shell and later PPP and dedicated line access (DS0, T1). Owen ...and CRL, and shortly after Netcom came Scruznet, and ... (Still giggling at how many times CRL got the intersection of Market/Geary/Kearny dug up in the early 90s bringing fiber in...). George William Herbert Sent from my iPhone
Re: Verizon Public Policy on Netflix
Hi Randy, Randy Bush wrote: And, for the record, it's pretty widely acknowledge that The World (Barry Shein) was the world's first commercial ISP - offering shell access in 1989, and at some point started offering PPP dial-up services. As I recall, they were a UUnet POP. yep. and uunet and psi were hallucinations. can we please not rewrite well-known history? umm what history am I re-writing? http://www.zakon.org/robert/internet/timeline/ - is as good a source as any for Internet history, which says this under 1990 The World comes on-line (world.std.com), becoming the first commercial provider of Internet dial-up access says ok - one can quibble 1989 (what Barry states on World's home page) PSInet was very late 1989, so there was that, I believe UUnet was 1990 What I did forget was NEARnet - which embarrasses me, since I was at BBN at the time. But, at first, NEARnet limited access to the NSFnet backbone to it's non-commercial customers (at least that was the policy - I'm not sure that filtering was ever really turned on in the gateways). I don't recall whether CSnet had any commercial members. or are you equating shell access with isp? that would be novel. unix shell != internet. well now we get to rehash to very old definitional distinction between Internet Access Provider and Internet Service Provider and yes, if a service provider takes money, to provide access to the Internet in some way, shape, manner, or form, yes - that's providing Internet access or service - and as soon as dial-up included PPP, then that's a non-issue btw, not do denigrate what barry did. a commercial unix bbs connected to the real internet was significant. the left coasties were doing free stuff, the well, community memory, ... and barry created a viable bbs commercial service which still survives (i presume). a significant achievement. The other service Barry provided was pushing the whole issue of commercial access to the backbone. That was kind of epic. And yes, they're still going strong. I still maintain an account - it's my backup for the rare case that I need a separate site for diagnosing issues with our cluster. Cheers, Miles -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
On Fri, 11 Jul 2014 19:22:52 -0700, Matthew Petach said: ISP until you're blue in the face, for all the good it does you; the incontrovertible point I'm making is that you don't exist as a recognizably separate entity from your upstream provider from the network perspective. If there's a problem, you're welcome to insist on calling his upstream's NOC and listen to them say that address is properly SWIP'ed to a customer until they're blue in the face because you claim the customer doesn't exist. The rest of us will go ahead and call the customer about the errant host on their network. pgpwd7kz34JTz.pgp Description: PGP signature
Re: Verizon Public Policy on Netflix
- Original Message - From: Matthew Petach mpet...@netflight.com I'm sorry. This is a networking mailing list, not a feel-good-about-yourself mailing list. From the perspective of the internet routing table, if you don't have your own AS number, you are completely indistinguishable from your upstream. Period. As far as BGP is concerned, you don't exist. Only the upstream ISP exists. Those things are all true, Matt. But they are orthogonal to are you an ISP (for any definition of ISP). Cheers, -- jra -- Jay R. Ashworth Baylink j...@baylink.com Designer The Things I Think RFC 2100 Ashworth Associates http://www.bcp38.info 2000 Land Rover DII St Petersburg FL USA BCP38: Ask For It By Name! +1 727 647 1274
Re: The Cidr Report
Well, probably 512k, but...
- Original Message -
From: Patrick W. Gilmore patr...@ianai.net
To: cidr-rep...@potaroo.net
Cc: NANOG list nanog@nanog.org
Sent: Friday, July 11, 2014 6:07:54 PM
Subject: Re: The Cidr Report
Does the CIDR report have a 510K prefix limit and crashed or
something?
:)
--
TTFN,
patrick
On Jul 11, 2014, at 18:00 , cidr-rep...@potaroo.net wrote:
This report has been generated at Fri Jul 11 21:10:32 2014 AEST.
The report analyses the BGP Routing Table of AS2.0 router
and generates a report on aggregation potential within the table.
Check http://www.cidr-report.org/2.0 for a current version of this
report.
Recent Table History
Date Prefixes CIDR Agg
04-07-14 507546 284271
05-07-14 508097 284317
06-07-14 508095 284519
07-07-14 508243 284914
08-07-14 508764 284695
09-07-14 508685 284695
10-07-14 0 284695
11-07-14 0 284695
AS Summary
0 Number of ASes in routing system
0 Number of ASes announcing only one prefix
3792 Largest number of prefixes announced by an AS
AS28573: NET Serviços de Comunicação S.A.,BR
0 Largest address span announced by an AS (/32s)
ÖØÿÿÿ : NET Serviços de Comunicação S.A.,BR
Aggregation Summary
The algorithm used in this report proposes aggregation only
when there is a precise match using the AS path, so as
to preserve traffic transit policies. Aggregation is also
proposed across non-advertised address space ('holes').
--- 11Jul14 ---
ASnum NetsNow NetsAggr NetGain % Gain Description
Table 508685 284695 223990 44.0% All ASes
AS28573 3792 139 3653 96.3% NET Serviços de Comunicação
S.A.,BR
AS6389 2951 80 2871 97.3% BELLSOUTH-NET-BLK -
BellSouth.net Inc.,US
AS17974 2789 186 2603 93.3% TELKOMNET-AS2-AP PT
Telekomunikasi
Indonesia,ID
AS22773 2664 191 2473 92.8% ASN-CXA-ALL-CCI-22773-RDC -
Cox Communications
Inc.,US
AS7029 2565 435 2130 83.0% WINDSTREAM - Windstream
Communications Inc,US
AS4766 2969 933 2036 68.6% KIXS-AS-KR Korea Telecom,KR
AS18881 2071 41 2030 98.0% Global Village Telecom,BR
AS18566 2046 565 1481 72.4% MEGAPATH5-US - MegaPath
Corporation,US
AS7303 1774 435 1339 75.5% Telecom Argentina S.A.,AR
AS7545 2322 996 1326 57.1% TPG-INTERNET-AP TPG Telecom
Limited,AU
AS10620 2901 1583 1318 45.4% Telmex Colombia S.A.,CO
AS4755 1866 591 1275 68.3% TATACOMM-AS TATA
Communications
formerly VSNL
is Leading ISP,IN
AS4323 1654 433 1221 73.8% TWTC - tw telecom holdings,
inc.,US
AS7552 1269 166 1103 86.9% VIETEL-AS-AP Viettel
Corporation,VN
AS36998 1114 37 1077 96.7% SDN-MOBITEL,SD
AS6983 1381 314 1067 77.3% ITCDELTA - Earthlink, Inc.,US
AS22561 1302 241 1061 81.5% AS22561 - CenturyTel Internet
Holdings, Inc.,US
AS6147 1020 145 875 85.8% Telefonica del Peru S.A.A.,PE
AS4788 1027 156 871 84.8% TMNET-AS-AP TM Net, Internet
Service Provider,MY
AS24560 1149 332 817 71.1% AIRTELBROADBAND-AS-AP Bharti
Airtel Ltd., Telemedia
Services,IN
AS7738 979 170 809 82.6% Telemar Norte Leste S.A.,BR
AS4808 1216 408 808 66.4% CHINA169-BJ CNCGROUP IP
network China169
Beijing
Province Network,CN
AS9829 1592 825 767 48.2% BSNL-NIB National Internet
Backbone,IN
AS11492 1247 490 757 60.7% CABLEONE - CABLE ONE, INC.,US
AS18101 942 186 756 80.3% RELIANCE-COMMUNICATIONS-IN
Reliance
Communications
Ltd.DAKC MUMBAI,IN
AS8151 1451 698 753 51.9% Uninet S.A. de C.V.,MX
AS26615 863 128 735 85.2% Tim Celular S.A.,BR
AS855 774 58 716 92.5% CANET-ASN-4 - Bell Aliant
Regional
Communications,
Re: Verizon Public Policy on Netflix
- Original Message - From: Matthew Petach mpet...@netflight.com To the core of the internet, if you do not have an AS number, you do not exist. If your business does not have an AS number *as far as the BGP speaking core of the internet is concerned, there is no representation for your entity, no matter what acronym you attach to it.* There. Confusion over. You can call yourself an ISP until you're blue in the face, for all the good it does you; the incontrovertible point I'm making is that you don't exist as a recognizably separate entity from your upstream provider from the network perspective. Ok. Correct. From the viewpoint of the context of this thread... why was that pertinent again? :-) Cheers, -- jra -- Jay R. Ashworth Baylink j...@baylink.com Designer The Things I Think RFC 2100 Ashworth Associates http://www.bcp38.info 2000 Land Rover DII St Petersburg FL USA BCP38: Ask For It By Name! +1 727 647 1274
Re: Verizon Public Policy on Netflix
On Sat, Jul 12, 2014 at 11:12 AM, Jay Ashworth j...@baylink.com wrote: - Original Message - From: Matthew Petach mpet...@netflight.com To the core of the internet, if you do not have an AS number, you do not exist. If your business does not have an AS number *as far as the BGP speaking core of the internet is concerned, there is no representation for your entity, no matter what acronym you attach to it.* There. Confusion over. You can call yourself an ISP until you're blue in the face, for all the good it does you; the incontrovertible point I'm making is that you don't exist as a recognizably separate entity from your upstream provider from the network perspective. Ok. Correct. From the viewpoint of the context of this thread... why was that pertinent again? :-) I totally don't remember. I just hit a stubborn streak. Now we're so far off in the weeds, I can't even see where we started from. ^_^;; Matt Cheers, -- jra -- Jay R. Ashworth Baylink j...@baylink.com Designer The Things I Think RFC 2100 Ashworth Associates http://www.bcp38.info 2000 Land Rover DII St Petersburg FL USA BCP38: Ask For It By Name! +1 727 647 1274
Re: Verizon Public Policy on Netflix
Now we're so far off in the weeds, I can't even see where we started from. ^_^;; What I'd like to know is 1) when does a terminating network become a transit network, and.. 2 )are there, should there, be different peering standards for each, and 3) if so some kind of functional if not structural separation 4) by regulation? j -- --- Joly MacFie 218 565 9365 Skype:punkcast WWWhatsup NYC - http://wwwhatsup.com http://pinstand.com - http://punkcast.com VP (Admin) - ISOC-NY - http://isoc-ny.org -- -
Re: Verizon Public Policy on Netflix
Joly MacFie wrote: Now we're so far off in the weeds, I can't even see where we started from. ^_^;; What I'd like to know is 1) when does a terminating network become a transit network, and.. 2 )are there, should there, be different peering standards for each, and 3) if so some kind of functional if not structural separation 4) by regulation? Ditto. These questions really get to the nub of the current issues! Miles Fidelman -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
One thing I've noted from those that support Verizon in this thread is that they often talk about Netflix's policy being unfair on small ISPs. Verizon is not a small ISP. Small ISPs seem happy peering with Netflix when they can (in fact they seem happy peering with anyone given there costs of transit) or getting a cache if they're big enough. My way of thinking it always has been that you are an ISP. An INTERNET service provider. As such you must make a best effort attempt to connect your customers to the internet at the speed you advertise. Let's cut the crap, Verizon is not irritated by Netflix's policies. They're irritated by Netflix and friends cutting into their far more lucrative content market.
Re: The Cidr Report
I've asked Geoff Huston to check, but no answer until now...
Am Samstag, 12. Juli 2014, 14:11:13 schrieb Jay Ashworth:
Well, probably 512k, but...
- Original Message -
From: Patrick W. Gilmore patr...@ianai.net
To: cidr-rep...@potaroo.net
Cc: NANOG list nanog@nanog.org
Sent: Friday, July 11, 2014 6:07:54 PM
Subject: Re: The Cidr Report
Does the CIDR report have a 510K prefix limit and crashed or
something?
:)
--
TTFN,
patrick
On Jul 11, 2014, at 18:00 , cidr-rep...@potaroo.net wrote:
This report has been generated at Fri Jul 11 21:10:32 2014 AEST.
The report analyses the BGP Routing Table of AS2.0 router
and generates a report on aggregation potential within the table.
Check http://www.cidr-report.org/2.0 for a current version of this
report.
Recent Table History
Date Prefixes CIDR Agg
04-07-14 507546 284271
05-07-14 508097 284317
06-07-14 508095 284519
07-07-14 508243 284914
08-07-14 508764 284695
09-07-14 508685 284695
10-07-14 0 284695
11-07-14 0 284695
AS Summary
0 Number of ASes in routing system
0 Number of ASes announcing only one prefix
3792 Largest number of prefixes announced by an AS
AS28573: NET Serviços de Comunicação S.A.,BR
0 Largest address span announced by an AS (/32s)
ÖØÿÿÿ : NET Serviços de Comunicação S.A.,BR
Aggregation Summary
The algorithm used in this report proposes aggregation only
when there is a precise match using the AS path, so as
to preserve traffic transit policies. Aggregation is also
proposed across non-advertised address space ('holes').
--- 11Jul14 ---
ASnum NetsNow NetsAggr NetGain % Gain Description
Table 508685 284695 223990 44.0% All ASes
AS28573 3792 139 3653 96.3% NET Serviços de Comunicação
S.A.,BR
AS6389 2951 80 2871 97.3% BELLSOUTH-NET-BLK -
BellSouth.net Inc.,US
AS17974 2789 186 2603 93.3% TELKOMNET-AS2-AP PT
Telekomunikasi
Indonesia,ID
AS22773 2664 191 2473 92.8% ASN-CXA-ALL-CCI-22773-RDC -
Cox Communications
Inc.,US
AS7029 2565 435 2130 83.0% WINDSTREAM - Windstream
Communications Inc,US
AS4766 2969 933 2036 68.6% KIXS-AS-KR Korea Telecom,KR
AS18881 2071 41 2030 98.0% Global Village Telecom,BR
AS18566 2046 565 1481 72.4% MEGAPATH5-US - MegaPath
Corporation,US
AS7303 1774 435 1339 75.5% Telecom Argentina S.A.,AR
AS7545 2322 996 1326 57.1% TPG-INTERNET-AP TPG Telecom
Limited,AU
AS10620 2901 1583 1318 45.4% Telmex Colombia S.A.,CO
AS4755 1866 591 1275 68.3% TATACOMM-AS TATA
Communications
formerly VSNL
is Leading ISP,IN
AS4323 1654 433 1221 73.8% TWTC - tw telecom holdings,
inc.,US
AS7552 1269 166 1103 86.9% VIETEL-AS-AP Viettel
Corporation,VN
AS36998 1114 37 1077 96.7% SDN-MOBITEL,SD
AS6983 1381 314 1067 77.3% ITCDELTA - Earthlink, Inc.,US
AS22561 1302 241 1061 81.5% AS22561 - CenturyTel Internet
Holdings, Inc.,US
AS6147 1020 145 875 85.8% Telefonica del Peru S.A.A.,PE
AS4788 1027 156 871 84.8% TMNET-AS-AP TM Net, Internet
Service Provider,MY
AS24560 1149 332 817 71.1% AIRTELBROADBAND-AS-AP Bharti
Airtel Ltd., Telemedia
Services,IN
AS7738 979 170 809 82.6% Telemar Norte Leste S.A.,BR
AS4808 1216 408 808 66.4% CHINA169-BJ CNCGROUP IP
network China169
Beijing
Province Network,CN
AS9829 1592 825 767 48.2% BSNL-NIB National Internet
Backbone,IN
AS11492 1247 490 757 60.7% CABLEONE - CABLE ONE, INC.,US
AS18101 942 186 756 80.3% RELIANCE-COMMUNICATIONS-IN
Reliance
Re: Verizon Public Policy on Netflix
I've only been 1/2 paying attention, did I miss the sarcasm tag are are people really looking for those answers. -jim Sent from my BlackBerry 10 smartphone on the Rogers network. Original Message From: Miles Fidelman Sent: Saturday, July 12, 2014 6:11 PM Cc: NANOG Subject: Re: Verizon Public Policy on Netflix Joly MacFie wrote: Now we're so far off in the weeds, I can't even see where we started from. ^_^;; What I'd like to know is 1) when does a terminating network become a transit network, and.. 2 )are there, should there, be different peering standards for each, and 3) if so some kind of functional if not structural separation 4) by regulation? Ditto. These questions really get to the nub of the current issues! Miles Fidelman -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
On July 12, 2014 at 12:08 ra...@psg.com (Randy Bush) wrote: And, for the record, it's pretty widely acknowledge that The World (Barry Shein) was the world's first commercial ISP - offering shell access in 1989, and at some point started offering PPP dial-up services. As I recall, they were a UUnet POP. yep. and uunet and psi were hallucinations. can we please not rewrite well-known history? or are you equating shell access with isp? that would be novel. unix shell != internet. You mean when you sat at a unix shell using a dumb terminal on a machine attached to the internet in, say, 1986 you didn't think you were on the internet? The shell machines were connected to the internet. You could FTP, email, telnet, etc etc etc. Back in 1989 that was on the internet. Heck, in 2014 it means on the internet. Right this minute I'm in a shell on a Linux machine connected to the internet and I'm pretty sure I have access to the internet. Consider the difference if you unplug that shell machine from the internet. Internet Service Provider. You got internet services. What hair are you trying to split? That you were using a shared address? Are people behind a NAT wall not on the internet? -- -Barry Shein The World | b...@theworld.com | http://www.TheWorld.com Purveyors to the Trade | Voice: 800-THE-WRLD| Dial-Up: US, PR, Canada Software Tool Die| Public Access Internet | SINCE 1989 *oo*
Re: Verizon Public Policy on Netflix
Personally, I'm not being sarcastic at all. Right now, peering agreements are the wild west. But.. there's rulemaking going on at the FCC - driven by all the talk about network neutrality and Internet Fast Lanes -- that is likely to have real impacts on all of us. Most of what passes for discussion is posturing by various big players, interest groups, and pundits. (To an earlier comment - Verizon is not a small ISP; but neither is Netflix a small business.) These are real questions, that merit serious examination - not to mention serious input to the current FCC rulemaking from knowledgeable folks. Just one man's opinion, of course. Miles Fidelman deles...@gmail.com wrote: I've only been 1/2 paying attention, did I miss the sarcasm tag are are people really looking for those answers. -jim Sent from my BlackBerry 10 smartphone on the Rogers network. Original Message From: Miles Fidelman Sent: Saturday, July 12, 2014 6:11 PM Cc: NANOG Subject: Re: Verizon Public Policy on Netflix Joly MacFie wrote: Now we're so far off in the weeds, I can't even see where we started from. ^_^;; What I'd like to know is 1) when does a terminating network become a transit network, and.. 2 )are there, should there, be different peering standards for each, and 3) if so some kind of functional if not structural separation 4) by regulation? Ditto. These questions really get to the nub of the current issues! Miles Fidelman -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
On 7/11/2014 11:38 AM, Miles Fidelman wrote: Well... if you make a phone call to a rural area, or a 3rd world country, with a horrible system, is it your telco's responsibility to go out there and fix it? One might answer, of course not. It's a legitimate position, and by this argument, Netflix should be paying for bigger pipes. Then again, I've often argued that the universal service fund used to subsidize rural carriers - which the large telcos always scream about - is legitimate, because when we pick up the phone and dial, we're paying for the ability to reach people, not just empty dial-tone. This is also legitimate, and by this argument, Verizon should be paying to improve service out to Netflix. If you're a competitor to the monopoly then you don't get access to those funds. It sucks for you, but that's just how it works. The county/state government has determined that they need to pay someone to make their network better in that region. They chose to pay the monopoly (whoever that is) and it wasn't you. It's the monopolies job to ensure good connectivity to Netflix. Oh, the monopoly is Comcast and they have a Netflix caching box but you don't? That is the cost of doing business in a rural market. You've got a few choices. Build out a fiber backbone to larger or more diverse markets, buy more transit, or go out of business. I service customers in small markets. Frequently they've got underpowered circuits because the incumbent won't sell MetroE or charges astronomical amounts for everything. If those were my only customers I'm not sure what I would do because I don't like their networks. I want to upgrade them but I'm being held back by various things. I've had situations where Monopoly entered a building at my expense to provide me fiber service so I could upgrade the users speed, then use that new fiber to undercut me on prices and take all the customers. People say the exclusive agreements for multi-dwelling units were bad for the little guy, but the truth is that the little guy could use exclusive agreements to allow the community to collective bargain for better internet. Now that those are gone, the competition is who can bribe the property manager more in pay-per-home connect fees. Either way, if one is a customer of both, one will end up paying for the infrastructure - it's more about gorillas fighting, which bill it shows up on, who ends up pocketing more of the profits, and how many negative side-effects result. No, it isn't. It's about monopolies telling a large company that isn't a monopoly that they need to pay them money to stay in business. Methinks all of the arguments and finger-pointing need to be recognized as being mostly posturing for position. Miles Fidelman
Re: Verizon Public Policy on Netflix
What is generally claimed is that I was the first to put the general public on the internet. Unix shell account, $20, connected machine, have at it. I got enough crap at the time for doing this that it must have been significant! ``Wot??? You can't put the GENERAL PUBLIC on the internet? What are you CRAZY??? You're illegally reselling federal property!!! (etc)'' The leap was that it was around $20 to ANYONE with a modem and a terminal (yes we had customers who actually used VT100s) or PC rather than thousands per month for a 9.6KB or 56KB leased line, router, etc. http://en.wikipedia.org/wiki/World.std.com On July 12, 2014 at 12:18 ra...@psg.com (Randy Bush) wrote: And, for the record, it's pretty widely acknowledge that The World (Barry Shein) was the world's first commercial ISP - offering shell access in 1989, and at some point started offering PPP dial-up services. As I recall, they were a UUnet POP. yep. and uunet and psi were hallucinations. can we please not rewrite well-known history? or are you equating shell access with isp? that would be novel. unix shell != internet. btw, not do denigrate what barry did. a commercial unix bbs connected to the real internet was significant. the left coasties were doing free stuff, the well, community memory, ... and barry created a viable bbs commercial service which still survives (i presume). a significant achievement. randy -- -Barry Shein The World | b...@theworld.com | http://www.TheWorld.com Purveyors to the Trade | Voice: 800-THE-WRLD| Dial-Up: US, PR, Canada Software Tool Die| Public Access Internet | SINCE 1989 *oo*
Re: Verizon Public Policy on Netflix
On July 11, 2014 at 22:31 o...@delong.com (Owen DeLong) wrote: Not to take away from Barry, but around that same time, some of us left coasts were also helping to build Netcom as a viable commercial entity providing shell and later PPP and dedicated line access (DS0, T1). That was several months later, Rieger et al were well aware of The World, and Panix for that matter which came after World but before Netcom. They were springing up, yes, but first is first, vague handwaves of around that same time is irrelevant. -- -Barry Shein The World | b...@theworld.com | http://www.TheWorld.com Purveyors to the Trade | Voice: 800-THE-WRLD| Dial-Up: US, PR, Canada Software Tool Die| Public Access Internet | SINCE 1989 *oo*
Re: Verizon Public Policy on Netflix
On July 12, 2014 at 07:16 mfidel...@meetinghouse.net (Miles Fidelman) wrote: umm what history am I re-writing? http://www.zakon.org/robert/internet/timeline/ - is as good a source as any for Internet history, which says this under 1990 The World comes on-line (world.std.com), becoming the first commercial provider of Internet dial-up access says ok - one can quibble 1989 (what Barry states on World's home page) PSInet was very late 1989, so there was that, I believe UUnet was 1990 I have ads and price schedules from October 1989 for public access internet. I could probably even dig up billing data from October or November. We actually started by offering shell and uucp access in August 1989 and then became a UUNET POP which put us directly on the internet in October. There was a T1 in our offices which back then was a pretty big deal! It was shared with other UUNET customers. We already had hundreds of customers using email etc when we became 192.74.137.*. UUNET and PSI internet wholesale were nearly simultaneous, I don't know the exact dates but early summer 1989 for internet sales. UUNET was already in the uucp biz for a year or two before that, we were a UUNET uucp customer when we started (and some other nodes like Encore, BU, etc.) Another reference is RFC2235 (I don't know why they used 1990 but it was written in 1997 and by then it didn't seem worth correcting) but there are a bunch of articles, I have most of them linked on my home page, http://www.TheWorld.com/~bzs What I did forget was NEARnet - which embarrasses me, since I was at BBN at the time. But, at first, NEARnet limited access to the NSFnet backbone to it's non-commercial customers (at least that was the policy - I'm not sure that filtering was ever really turned on in the gateways). I don't recall whether CSnet had any commercial members. Apple was a CSNET 56k customer. or are you equating shell access with isp? that would be novel. unix shell != internet. well now we get to rehash to very old definitional distinction between Internet Access Provider and Internet Service Provider and yes, if a service provider takes money, to provide access to the Internet in some way, shape, manner, or form, yes - that's providing Internet access or service - and as soon as dial-up included PPP, then that's a non-issue btw, not do denigrate what barry did. a commercial unix bbs connected to the real internet was significant. the left coasties were doing free stuff, the well, community memory, ... and barry created a viable bbs commercial service which still survives (i presume). a significant achievement. The other service Barry provided was pushing the whole issue of commercial access to the backbone. That was kind of epic. I agree, that's the real point. As I said, what I did caused a furor. And yes, they're still going strong. I still maintain an account - it's my backup for the rare case that I need a separate site for diagnosing issues with our cluster. Cheers, Miles -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra -- -Barry Shein The World | b...@theworld.com | http://www.TheWorld.com Purveyors to the Trade | Voice: 800-THE-WRLD| Dial-Up: US, PR, Canada Software Tool Die| Public Access Internet | SINCE 1989 *oo*
Re: Verizon Public Policy on Netflix
On 7/12/2014 5:19 PM, Barry Shein wrote: On July 12, 2014 at 12:08 ra...@psg.com (Randy Bush) wrote: And, for the record, it's pretty widely acknowledge that The World (Barry Shein) was the world's first commercial ISP - offering shell access in 1989, and at some point started offering PPP dial-up services. As I recall, they were a UUnet POP. yep. and uunet and psi were hallucinations. can we please not rewrite well-known history? or are you equating shell access with isp? that would be novel. unix shell != internet. You mean when you sat at a unix shell using a dumb terminal on a machine attached to the internet in, say, 1986 you didn't think you were on the internet? The shell machines were connected to the internet. You could FTP, email, telnet, etc etc etc. Back in 1989 that was on the internet. Heck, in 2014 it means on the internet. Right this minute I'm in a shell on a Linux machine connected to the internet and I'm pretty sure I have access to the internet. Consider the difference if you unplug that shell machine from the internet. Internet Service Provider. You got internet services. What hair are you trying to split? That you were using a shared address? Are people behind a NAT wall not on the internet? This must be the silliest recurring thread-topic on NANOG since the Spam is NOT an Operational issue (or DDOSes are not [ditto]) days. For the Subject: line -- when my provider stops providing what I want at a price I want to pay, I'll start looking for another one and as an end user I am not remotely interested in the nasties they have to through to GET what I want delivered. For the current thread position -- At this precise moment I am using Thunderbird (a messaging client with shell aspirations) under Windows XP (a shell with OS pretensions) talking to the network I developed, installed, pay for and maintain (could be called my ISP and separately my wife's ISP, and the ISP for invited and uninvited guests--could be but won't be because it conveys no useful information to anybody). That network is connected to a company's cable, which company is my ISP, my POTSP, and my TVP. Who and what they connect to to get th4e stuff I want delivered is only of academic interest. -- Requiescas in pace o email Two identifying characteristics of System Administrators: Ex turpi causa non oritur actio Infallibility, and the ability to learn from their mistakes. (Adapted from Stephen Pinker)
Re: Verizon Public Policy on Netflix
Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. what you are seeing, and creating massive noise around, is a business war between the last mile cartel and the content they envy and want to supplant or at least bleed. transit, peering, caching, etc. are just business and technical tools being used in that war. keep eye on doughnut, not the hole. randy
Re: Verizon Public Policy on Netflix
On 7/12/2014 3:43 PM, Barry Shein wrote: I don't recall whether CSnet had any commercial members. Apple was a CSNET 56k customer. As I recall, Schlumberger (http://www.slb.com/, a research site of theirs on the west coast) was one of the earliest CSNet member. So was HP. I put Schlumberger online circa 1981 or 1982. I believe they were among the first 5-10 sites I brought up. At that stage, it was only email relaying, of course. Packet services were later. Also, although CSNet started with NSF money, it was required to become self-funded within 5 years. Albeit on a non-profit financial model, I'd claim that that made it, essentially, a commercial access service. If one allows 'commercial' ISP to cover independent operations that happened not to have a profit-oriented motive, I suspect the first service to quality would be The Little Garden, operated as a direct consortium, rather than having third-party operations, as CSNet did. d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net
Re: Verizon Public Policy on Netflix
Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. to be clearer. by count, the vast majority of peering is done by small ops informally. this represents a small fraction of the traffic. you don't see the peering agreements because there are no formal ones. and i guess it looks chaotic from the outside. it looks pretty normal from the inside. e.g., i have a research rack connected to the six, tell most folk there that i have neither eyeballs nor eye candy, but peer informally with folk such as re networks where i need to move data. oh, and the rack peers informally with my $dayjob, which might be the only informal peering which $dayjob does. makes sense from the inside, looks strange from the outside. you have to know my business model for it to make sense. the big kids peer very formally, which represents the majority of the traffic, and often does not happen at exchanges. like many bi-lateral business to business deals in the commercial world, the details are confidential. to you, it may look like the wild west. to the players, it's just business. randy
Re: Verizon Public Policy on Netflix
On 2014-07-12 09:33, valdis.kletni...@vt.edu wrote: On Fri, 11 Jul 2014 19:22:52 -0700, Matthew Petach said: ISP until you're blue in the face, for all the good it does you; the incontrovertible point I'm making is that you don't exist as a recognizably separate entity from your upstream provider from the network perspective. If there's a problem, you're welcome to insist on calling his upstream's NOC and listen to them say that address is properly SWIP'ed to a customer until they're blue in the face because you claim the customer doesn't exist. Except when, as in the original example, it's not. Jima
Re: Verizon Public Policy on Netflix
Randy Bush wrote: Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. what you are seeing, and creating massive noise around, is a business war between the last mile cartel and the content they envy and want to supplant or at least bleed. transit, peering, caching, etc. are just business and technical tools being used in that war. keep eye on doughnut, not the hole. Sure looks like a wild west range war to me. And let's not forget that Netflix is not some tiny company anymore - 1/3 of Internet traffic or some such, 46million members, $1billion Q1 income. Yeah - big guys fighting, no established law or regulation (well, there was, but the Supreme Court overturned it) - looks like a range war to me. Miles -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
Randy Bush wrote: Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. to be clearer. by count, the vast majority of peering is done by small ops informally. this represents a small fraction of the traffic. you don't see the peering agreements because there are no formal ones. and i guess it looks chaotic from the outside. it looks pretty normal from the inside. e.g., i have a research rack connected to the six, tell most folk there that i have neither eyeballs nor eye candy, but peer informally with folk such as re networks where i need to move data. oh, and the rack peers informally with my $dayjob, which might be the only informal peering which $dayjob does. makes sense from the inside, looks strange from the outside. you have to know my business model for it to make sense. the big kids peer very formally, which represents the majority of the traffic, and often does not happen at exchanges. like many bi-lateral business to business deals in the commercial world, the details are confidential. to you, it may look like the wild west. to the players, it's just business. Exactly - all of this is informal, unregulated, done on a deal-by-deal basis - and the big guys fight things out with big guns. How is this not the wild west? Miles Fidelman -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Re: Verizon Public Policy on Netflix
Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. what you are seeing, and creating massive noise around, is a business war between the last mile cartel and the content they envy and want to supplant or at least bleed. transit, peering, caching, etc. are just business and technical tools being used in that war. keep eye on doughnut, not the hole. Sure looks like a wild west range war to me. And let's not forget that Netflix is not some tiny company anymore - 1/3 of Internet traffic or some such, 46million members, $1billion Q1 income. Yeah - big guys fighting, no established law or regulation (well, there was, but the Supreme Court overturned it) - looks like a range war to me. ahhh. so not government regulated == wild west got it randy
Re: Verizon Public Policy on Netflix
Randy Bush wrote: Right now, peering agreements are the wild west. no. those days passed in the last century. you just don't know them. but then, you are not an operator so no surprise. what you are seeing, and creating massive noise around, is a business war between the last mile cartel and the content they envy and want to supplant or at least bleed. transit, peering, caching, etc. are just business and technical tools being used in that war. keep eye on doughnut, not the hole. Sure looks like a wild west range war to me. And let's not forget that Netflix is not some tiny company anymore - 1/3 of Internet traffic or some such, 46million members, $1billion Q1 income. Yeah - big guys fighting, no established law or regulation (well, there was, but the Supreme Court overturned it) - looks like a range war to me. ahhh. so not government regulated == wild west got it randy lawless, big guys fighting with little guys in the middle == wild west -- In theory, there is no difference between theory and practice. In practice, there is. Yogi Berra
Verizon Public Policy on Netflix
This is Brett Glass; I have been alerted to some of the responses to my message (which was cross-posted by a third party) and have temporarily joined the list to chime in. The following is my response to his message, edited slightly to include some new information. Dave Temkin wrote: First and foremost, we built our CDN, Open Connect, Open Connect is not, in fact, a CDN. Nor is it peering. It is merely a set of policies for direct connection to ISPs, and for placing servers in ISPs' facilities, that is as favorable as possible in every way to Netflix. It costs Netflix as little as possible and the ISP as much as possible. with the intention to deploy it as widely as possible in order to save ISPs who are delivering our traffic money It does not, in fact, appear to save ISPs money. Note that Comcast asked for, and was given, additional payments even after it did all of the things that are part of Netflix' Open Connect program. Netflix, exercising inappropriate market power, has not offered smaller ISPs such as my own the same amount per customer. In fact, it has offered us no money at all -- even though our costs per Netflix customer are higher. Netflix thus discriminates against and threatens smaller ISPs, and by doing so, harms broadband competition. and improve our mutual customer experience. This goes for ISPs large and small, domestic and international, big endian and little endian. We've never demanded payment from an ISP nor have we ever charged for an Open Connect Appliance. The power and bandwidth consumed by an Open Connect Appliance (which is really just a hosted Netflix server) are a substantial expense for any ISP. Especially because the server is not a cache; it is stocked with content whether it is used or not and therefore wastes bandwidth on content and formats that will never be used even once. When we first launched almost three years ago, we set a lower boundary for receiving a Netflix Open Connect Appliance (which are always free) at 5Gbps. Since then we've softened that limit to 3.5Gbps due to efficiencies of how we pre-load our appliances (more on that below). Most small ISPs (the average in the US, in fact) have 1,000 to 2,000 accounts. If every one of those streams at 1 Mbps at the same time, which is highly unlikely, this still does not reach 3.5 Gbps. Therefore, most ISPs are excluded simply by this requirement if not by others (such as the requirement that the ISP alone pay for a dedicated connection to one of Netflix' relatively few peering points). We explicitly call our cache an Appliance because it's not a demand driven transparent or flow-through cache like the Akamai or Google caches. We do this because we know what's going to be popular the next day or even week and push a manifest to the Appliance to tell it what to download (usually in the middle of the night, but this is configurable by the ISP). What Netflix does not say here is that (a) it can only somewhat predict what will be in demand or go viral; (b) it wastes bandwidth by sending multiple copies of each video to its server in different formats, rather than transcoding locally or saving bandwidth on lesser used formats via caching; and (c) its server consumes large amounts of energy and bandwidth. A cache can be much more efficient and can be owned and managed by the ISP. The benefit of this architecture is that a single Appliance can get 70+% offload on a network, and three appliances clustered together can get 90+% offload, while consuming approximately 500 watts of power, using 4U of rack space, and serving 14Gbps per appliance. To put this in perspective: LARIAT builds its own caches which consume as little as 20 watts and can saturate a 10 Gbps Ethernet port. The Netflix servers are large, bloated power hogs compared to a well designed cache. The downside of this architecture is that it requires significant bandwidth to fill; in some ISPs cases significantly more than they consume at peak viewing time. This is why our solution may not work well for some small ISPs and we instead suggest peering, which has 100% offload. Because it requires expensive bandwidth that's dedicated solely to Netflix, peering (as Netflix calls it; it's really just a dedicated link) has 0%, not 100%, offload. The ISP is paying for all of the bandwidth, and it cannot be used for anything else. We've put a lot of effort into localizing our peering infrastructure worldwide. As you can see from this map (sorry for the image), we're in 49 locations around the world with the significant bulk of them in the US (blue pins = 1 location, red pins = 1 location in a metro) - more detailed version at http://goo.gl/eDHpHUhttp://goo.gl/eDHpHU and in our PeeringDB record ( http://as2906.peeringdb.comhttp://as2906.peeringdb.com) : Our ISP connects to the Internet in Cheyenne, Wyoming (a major Internet crossroads; it's where I-80 meets I-25) and Denver, Colorado (which is, if anyplace can make the claim, the center of
Re: Verizon Public Policy on Netflix
ahhh. so not government regulated == wild west lawless, big guys fighting with little guys in the middle == wild west at this point, maybe john curran, who you may remember from nearnet, usually steps in with a good screed on industry self-regulation. and, if we are really lucky, maybe geoff will use his deep knowledge of history beyond the internet to tell us the real story of the so called wild west, which was likely more organized than one would think. randy
Re: Verizon Public Policy on Netflix
On 7/12/2014 9:42 PM, Randy Bush wrote: ahhh. so not government regulated == wild west More like not civilized == wild west. Although as a native Westerner, I thing that is still an unfair slur. -- Requiescas in pace o email Two identifying characteristics of System Administrators: Ex turpi causa non oritur actio Infallibility, and the ability to learn from their mistakes. (Adapted from Stephen Pinker)
Re: Verizon Public Policy on Netflix
Brett, You've previously stated: https://www.grc.com/sn/sn-457.htm $20 per mbps per month 1.25 gigabits of bandwidth coming in Math: $20/mb/month * 1250mb = $25,000month If netflix is 1/3 of bandwidth... saving 1/3 of $25,000 -= $8,000/month. (OK, Keep 100mbps for Netflix to pre-populate, 100mbps is 30TB/month) (Now I'm curious how many GB/month Netflix pre-populates, hmmm) How would 4U of rent and 500W($50) electricity *not* save money? If it's a money thing, then you have a price... How much would Netflix have to pay you for you to consider? Can you elaborate on: It costs Netflix as little as possible and the ISP as much as possible. If your ISP isn't tall enough for Netflix, Akamai has a lower barrier of entry. Have you let Akamai give you a local cache? why or why not? Steven Tardy, maybe I'm missing/overlooking something... On Sat, Jul 12, 2014 at 8:22 PM, na...@brettglass.com wrote: This is Brett Glass; I have been alerted to some of the responses to my message (which was cross-posted by a third party) and have temporarily joined the list to chime in. The following is my response to his message, edited slightly to include some new information. Dave Temkin wrote: First and foremost, we built our CDN, Open Connect, Open Connect is not, in fact, a CDN. Nor is it peering. It is merely a set of policies for direct connection to ISPs, and for placing servers in ISPs' facilities, that is as favorable as possible in every way to Netflix. It costs Netflix as little as possible and the ISP as much as possible. with the intention to deploy it as widely as possible in order to save ISPs who are delivering our traffic money It does not, in fact, appear to save ISPs money. Note that Comcast asked for, and was given, additional payments even after it did all of the things that are part of Netflix' Open Connect program. Netflix, exercising inappropriate market power, has not offered smaller ISPs such as my own the same amount per customer. In fact, it has offered us no money at all -- even though our costs per Netflix customer are higher. Netflix thus discriminates against and threatens smaller ISPs, and by doing so, harms broadband competition. and improve our mutual customer experience. This goes for ISPs large and small, domestic and international, big endian and little endian. We've never demanded payment from an ISP nor have we ever charged for an Open Connect Appliance. The power and bandwidth consumed by an Open Connect Appliance (which is really just a hosted Netflix server) are a substantial expense for any ISP. Especially because the server is not a cache; it is stocked with content whether it is used or not and therefore wastes bandwidth on content and formats that will never be used even once. When we first launched almost three years ago, we set a lower boundary for receiving a Netflix Open Connect Appliance (which are always free) at 5Gbps. Since then we've softened that limit to 3.5Gbps due to efficiencies of how we pre-load our appliances (more on that below). Most small ISPs (the average in the US, in fact) have 1,000 to 2,000 accounts. If every one of those streams at 1 Mbps at the same time, which is highly unlikely, this still does not reach 3.5 Gbps. Therefore, most ISPs are excluded simply by this requirement if not by others (such as the requirement that the ISP alone pay for a dedicated connection to one of Netflix' relatively few peering points). We explicitly call our cache an Appliance because it's not a demand driven transparent or flow-through cache like the Akamai or Google caches. We do this because we know what's going to be popular the next day or even week and push a manifest to the Appliance to tell it what to download (usually in the middle of the night, but this is configurable by the ISP). What Netflix does not say here is that (a) it can only somewhat predict what will be in demand or go viral; (b) it wastes bandwidth by sending multiple copies of each video to its server in different formats, rather than transcoding locally or saving bandwidth on lesser used formats via caching; and (c) its server consumes large amounts of energy and bandwidth. A cache can be much more efficient and can be owned and managed by the ISP. The benefit of this architecture is that a single Appliance can get 70+% offload on a network, and three appliances clustered together can get 90+% offload, while consuming approximately 500 watts of power, using 4U of rack space, and serving 14Gbps per appliance. To put this in perspective: LARIAT builds its own caches which consume as little as 20 watts and can saturate a 10 Gbps Ethernet port. The Netflix servers are large, bloated power hogs compared to a well designed cache. The downside of this architecture is that it requires significant bandwidth to fill; in some ISPs cases significantly more than they consume at peak viewing time. This is why our
