NANOG 37

June 4 - 7, 2006, San Jose, CA

Attendee Survey Results

Thank you for taking a moment to give us your comments about the meeting!

How did you attend NANOG?
Via the web (6) In person in San Jose (114)

Overall, was this NANOG useful to you?
Very useful (39) Useful (75) No opinion (2) Not very useful (5) Useless ()

Is this your first time attending NANOG?
Yes (26) No (95)

If you have attended a previous NANOG, how does this NANOG compare?
Better (35) About the same (48) Worse (10) N/A (15)

Did you find the General Session and Tutorial/BoF schedule acceptable?
Very useful (45) Useful (59) No opinion (12)

Comments on the General Session:

"Information Collection on DDoS Attacks" Stated involving Japan or China made it difficult to track down the attacker which made me wonder why. I'm originally from Japan and don't see any difference.
liked the parts dealing w/secure routing
I was only able to attend the first half of Tuesday morning's session, but highly enjoyed the spammer behavior talk. Also liked DDoS mitigation.
Liked having the longer talks. Things were more in-depth
People (some) need to learn to talk in the microphones.
Research forum: all three useless. Durand: decent exposure, too long. Net Neutrality: so-so, not a lot of debate. Feamster: enough already. Claiborne: showing midriff doesn't overcome the talk aimed at the wrong audience. ENUM: yawn. IDC Power/Cooling: this was excellent and lively! SNDS: decent preso Fuller/Schiller: for gosh sakes, make it fit the time alloted.
this was a good format. Room was adequate and provided good access. Screens for presentations are too small to be seen from back row, without effort.
I enjoyed the Net Neutraility but again the audience seemed a little flat. I think that was due to too many facets of netneutrality coming to light ( or perhaps a general feeling of "non-issue" amongst the population) For the Cooling IDC, while it was good and got good response/involvement, I think inviting Leibert or others actually involved in cooling IDCs would have been interesting. Perhaps a good talk in the future would be a report or status from teh 24/7 conference dealing with these issues specifically.
I really enjoyed most of the morning talks... I thought the Net neutrality could of used more discussion and debating a bit....
Managing IP Address, A Durand was VERY good. They sky is falling, and the sky is comcast. Lightning talks great!
Nicely balanced agenda with variety of subjects. Great job from the PC.
Found lightning talks very useful (Pilosov, Hannigan and Kapela). Would like to have more talks about the practical side of operating a network provider (including WDM, datacenter panels, etc).
No. Too spaced out Net Neutrality panel missed half the real issues DOS logging, presenter was ill-informed
Lightning talks rock! Keep them coming. Themes help in the morning. Request an Enum update for October please.
All good
- post talks and speakers for lightning talks-if you missed intro, there's no way to catch up - good talks-Alain Durand, Nick Feamster (good topics, good/organized presos) - the two prelunch panels needed stronger moderations or more time.
Found 1) A simple coordination mechanism for interdomain routing/issues-Rotul Mahajan-very intersting 2) Managing root million IP addresses-Alain Durand which will lead us towards IPv^
General Sesson and Tutorial/BoF schedule not very useful Most talks were good The first 2/3rds of "Info collection on DDoS" was too much like a tutorial. The good part of the talk was the last 1/3
General sessions were very good overall. Exception was ddos and enum talks which I did not feel were relevant.
Generally very good, topical agenda, mostly with well-thought out talks. Some things seemed to be squeezed on time for Q&A/discussion. Maybe this is a factor of having interesting topics, but that should be factored in when doing future agenda planning. Great lightning talks again! Brilliant stuff. Keep them!
I like the 3 day format of this meeting Lightning talks & research talks were good addition
General sessions were very well organized and varied in topics enough to keep them interesting.
2 day format is better
Good discussions. May be usefull to seperate topics with similar subject matter. Keeps the pace changing.
The talk on response to DDoS attacks did not seem to provide any new data. If there were operators who needed to know this, it might have been a tutorial. P.S. Like the notes idea - sometimes they've caught points I missed. (More true of panel discussions than individual talks)
I like the short sessions, good control, no wasted time
Generally good
Lightning talks worked well Comcast talk was good - real deployment and ops issues Prolexic speaker did not seem well informed
It was overall pretty good. Hats off to the Program Committee.
Generally everything was topical. I would have liked to see more on advanced services over IP (voip, iptv, whatever's next) and operators' struggles/success with these.
This was the first NANOG I've attended with the "new" format (Sunday - Wednesday). I think it will take a few more NANOGs before I get used to the new format. The tough thing for me is that we have essentially extended the entire conference by 1 business day. If I want to fully participate from start to finish, it means one more day on the road. On the plus side, the pace is a little more relaxed, and I don't feel like I'm trying to cram everything into a short period. I really liked the Lightning talk session. I left NANOG on Tuesday afternoon and watched the rest from my desk on Wednesday morning. Great job on the streaming video. For the lightning talks it was a little frustrating not the have the slides posted in advance. I can appreciate the difficulty of the video guys having to cut between camera shots of a live speaker and a projected presentation. Some of the presentation material was extremely difficult to read via the webcast. Other than going to something like a Webex solution, I'm not sure how to caputre a live person and static presentation matrial for a remote audience.
Another great NANOG
I liked that there a number of topics covered briefly. Found the Ratul and Anna's topics pretty interesting.
Good presentations Liked the lightning talks Anycast talks on Wed. were particularly interesting
Liked the lightning talks this time around.
enjoyed Pilosov/Rothschild/Patrick presentation on CWDM; would like to see more talks of this nature, and other topics of interest to smaller networks doing creative things. also enjoyed Kapela lightning talk and would like to see this evolve into a proper tutorial or presentation.
I was expecting a more professional conference. Having all of the presentations on a single PC/Mac could help this. Why people that are on the stage, who are not speaking, looking at their laptops is very odd.
It seems there have been a couple of presentations where too much time was spent on background (for example, the IPv6 talk which was mostly review but was supposed to be about open issues) which should already be known to operators who are interested in the area. I'd like to see the "coach" idea that was suggested in the community meeting include tuning the presentations to exclude information that is old hat or has been under discussion/available for some time already.
Generally good, balanced programme. Tuesday morning's security related and IDC panel were particularly interesting. Wed morning was of less interest with the exception of the lightning talks which were good
I only have one complaint -- the talk Vince Fuller and Jason Schiller did could have used up 2 or 3 hours. They were presenting an issue that hits a lot of engineers' hot buttons (Randy Bush, Steve Bellovin, Tony Hain, Rob Seastrom, Jared Mauch, etc. all went running to mics) and needs more discussion and action by the NANOG community. 30 minutes was too short a window (and yeah, Vince was longwinded, but whaddya gonna do).
I prefer the half day morning sessions. it is easier to clear a half day in my schedule, than a full day.
Many uninteresting subjects.
Suprisingly even the research session was engaging.
In general they were good. Unfortunately there wasn't enough time to discussion on the IPv6 Routing/Multihoming presentation. Maybe it should be discussed again in the next meeting.
I enjoyed the talks
Good tutorial talks Auth for PCP-base protocols was very good & technical. Active measurement of AS pre-paid was interesting &pertanent to conventions of routing policy to prefer customers over peers
Overall, a better than average agenda. I was quite pleased.
I think the agenda was good, nice mix of different topics.
Overall I would say that the content was better in Los Angeles however there were bits of the content at this meeting that was excellent, i.e. Anna Claiborne.
I thought there was a good coverage on a wide array of topics. Unfortunately I couldn't always pay as close attention to some of them as I was working off and on during much of the conference, but the online proceedings and notes by Matthew Petach should help fill in the blanks. There seemed to be a continuous problem with sessions running late and I felt at least one time a moderator was too strict to the point of bordering on rude to one of the people at the mic during the Q&A time. Q&A time should be limited, but I don't think it was necessary to cut that portion of the program just because we were running behind. Although pointless jabs at good people like those at ISC by the likes of Sam Weiler may have been appropriate to sanction.
Liked the interaction to work out issues real time.
Vince Fuller's IPv6 talk is brilliant & the most sensible IPv6 talk I've heard in a decade. Comcast's IPv6 talk was too specific to Comcast and not useful.
Most talks useful, interesting, however, during Research Forum the disadvantage of accepting talk from non-English speaker (Hong Kong), was they did not understand the points, questions etc that NANOG participants were trying to communicate. Not sure how interesting their results were as well.
I found authentication for TCP-based Routing & Pretty good BGP talks quite interesting. Managing 100+ million IP addresses also informative. I enjoyed the Network Neutrality panel although I expected it to be more heated :)
Good.
The talks were for the most part good and well paced. Didn't have problems falling asleep.
The General session was pretty good and informative. Tighter controls should be kept on the scheduling though so things follow the time format.
General Sessions as always are good.
Requesting slides upfront clearly limits the # of submissions. I would suggest accepting proposals, w. a deadline for draft slides & receiving folks off the program of late or not up to par.
4 day schedule (sun - wed) seems to drag a bit.
Less useful Not that interesting overall. A few good ones.
More time for discussion, specially the panels.

Comments on Tutorials:

Didn't attend the tutorials, as on Monday they were at the same time as the BOFs. Too bad there's not more time, so they don't have to be parallel, but my time was very limited anyway so I probably would have had to miss one or the other even if they were in series.
I didn't attend any. they did not meet my interests.
I didn't attend any Tutorials...=( I found the BOFs subject more intresting...
NA
Did not attend
NA
Ask P. Smith to do an Advanced BGP tutorial!
I attended only passive monitoring access tutorial. It was ok, but seemed kind of weak-I can't quite pinpoint why though
1) BGP Techniques for service provider - Cisco Philip Smith very useful
Did not attend
mpls was good. passive monitoring was more like a vendor sales pitch.
Didn't attend any tutorials this meeting.
I was particularly pleased with the tutorial presented by Philip Smith on BGP. Much of it I already had a firm grasp of but it reaffirmed my understanding of BGP and additionally provided insight into proper scaling techniques. We are in the process of building our network out from our current east coast POPs to the west coast so this was very useful. I understand the time limitations in any presentation, but more configuration examples would have been helpful. I know it's vendor neutral, but being a Cisco representative was presenting, they could at least provide some things such as how to setup communities using route-maps for example. Overall I was very pleased though.
Philip Smith's BGP Techniques for Service Providers was fantastic. I will refer back to the video stream in the future.
Only attended part 1 of the BGP course. I found it useful.
Did not attend any tutorial
More topics on enterprise specifics even though some of them may not have anything related to Network operators. Such as enterprise VOIP implemenation, SOX, etc.
N/A
okay.... just okay.
Both tutorials I attended were good. Philip's BGP 1 & 2 as well as the MPLS tutorial by Pete.
didn't see
did not attend
don't attend tutorials. I didn't like the "new" format as tutorials effectively waste two half days and require an extra travel day (Wed). Please move them back to Sunday - use "old" format
Didnt go.
n/a
They are ok.
Did not attend tutorial
Didn't attend any.
Tutorials should have tables-chairs only are hard to take notes
The MPLS TE tutorial was quite basic. I would have appreciated mention of new technologies.
Philip Smith is always a winner
Did not attend any tutorials.
Enjoyed the tutorials it created opportunities to hear about new items
The Net Optics talk about wire taps was a glorified ad and a total waste of time
Did not attend a tutorial
Phil Smith, given his level of subject talk-practice :) is very good.
MPLS tutorial was quite good, and the advanced section of the Passive Monitoring tutorial was quite useful.
Not applicable.
BGP/Phil Smith-he intermixed real and unreal/historical stuff with too little flagging between them. Make the flagging bold--change slide backgrounds, etc
no specific comments.
Didn't attend the tutorials, since the BOF's were on at the same time. The multitrack system at a conference works if there's a group of people coming from the same organization splitting up to cover everything (USENIX Annual Tech Meeting deja vu). If you're alone this is not possible to do, unless both the tutorials and BOFs are recorded for later viewing - and I think it's a pity that the BOFs at least aren't kept around because there's a lot of good stuff coming out of them for the community to benefit from. Somehow it has to get out.
Passive monitoring was good, but a bit too introductory. However, not very original-clearly a preso. Requesting slides in the submission proposal basically ensures no origianl material will be presented, because folks want to put in the work to build a tutorial unless there is interest
BGP tutorial could be at a more advanced level. Most folks attending these sessions have a pretty good understanding of that routing protocol.
Didn't attend
All East coast NANOG attendees now have the opportunity to participate in Tutorials. BGP for Service providers by Philip Smith was excellent. However, Phil ran out of time. The allocated time was not enough.

Comments on BoFs:

great BGP tools & bgp for netops; useful to take BGP concepts and help various network operators implement useful policies and monitor them
Add an evening session after security BoF for further sole division (hacking, DoS, Spam, Application) to get birds together
Peering BOF was very useful. BGP tools BOF was less useful, I think because it was in the big general session room that was not conducive to interaction. It might be better to have it in a smaller room.
IXP BoF was great.
Why did you took two bofs - Peering and IX ? at the IX bof not that may people came, because the Peering-Bof was taking too much time and Beer and Gear was calling.
BOF's were interesting, rooms were too small, Big room was not us use, should have used it as oppsed to small conference room. Peering, EP BOF, Security BOF. need White boards available as well.
I was a little disappointed to see attendance at the Exchange Point operators BOF be so thin. Especially given the panel on heating and cooling in the IDC that was so lively. On the other hand the loose atmosphere of the Peering BoF that has worked so well in the past seemd too dis-jointed and made the BoF a little flat. I understand there it requires more work, but the peering debates brought out more issues and genreally engaged the audience more deeply.
Peering BOF fantastic, as always
Peering BOF excellent as usual. Really like the seating arrangement and the parliamentary rules.
The peering BoF always draws a large crowd. :)
Bill Nortons peering BoF always runs out of time. Can there be a part 2 after the break? Maybe IX Bof & Power BoF the second afternoon next time?
All good
Good stuff
The BoF timings were very tight, heavily constrained. Consider having 2hr BoF slots available where necessary, maybe? The IXP Operations BoF was very welcome, however it needed to be more focused on IXP Operations - real operational issues, tools, etc. I realise that needs the IXPs to participate and submit talks. Please keep the IXP Operations BoFs running, we've definitely got things to talk about, it just needs a bit of tuning, and getting people to step up to the plate.
Did not attend any. Was at the tutorials...
Seemed like small group discussions for others to watch and learn. Difficult to get discussion going when people are forced to get to the mic.
BoFs were good. Needed More seating space. The peering BoF seems to have resurrected iteself as the place to be. The individual talks in Security BoF should be shorter - the BoF lacks the fun and usefullness of the Rob Thomas Barry Greene show.
Security BoF ok-but some topics (like attacks) should have been a talk on themselves
Peering BOF is cool I liked the interactions and pace of the peering bof
Great ones (atleast the ones that I attended). I attended Peering and Exchange Point Operators BOF
The OPSEC WG BoF was a bit weak. I don't think BoF's are proper forums for comments on papers; rather, BoF's should discuss the issues and provide the seeds for future papers.
Peering BOF - Bill Norton did great as always. IX BOF - was not well attended I think due to lack of large IX participation. Need to change that.
The oval setup for the peering BOF was a great idea. I don't think it scaled well to the number of people participating.
better
Only attended the Security BOF. I liked that session. Good lessons learned from various problems that have occured recently.
Enjoyed the security bof's, though those were the only ones I attended.
WBN's Peering BOF was, as always, a treat. Would like to see this at future NOGs, though that is the plan I'm sure.
I felt that much of the discussion in the community meeting on the difficulties around venue selection and pricing was better left to a small group. It was good to hear a summary of the issues as part of the budget overview, but I would have rather heard discussion on the agenda items more related to network operations. A brief update on Business Issues is good, but I'd prefer to leave the details to an out of band conversation with those folks experineced in Business matters and those operators who care about that side of NANOG. It's certainly important and does require attention (and praise for those who take it on). But like nanog-futures I'd like to be able to filter it out from the more operationally related topics.
Peering BOF was good, but seating plan was very poor for this well attended BOF. The floor was not comfortable! IX Operator's BOF did not work well unfortunately. As with other IX events it turned out to be IX operators talking to other IX operators about topics they probably knew already.
Feamsters BGP tool BOF was very disappointing. A rehash of topics at NANOG Chicago. If I wanted to read about the router checker thing, I would have looked in the archives. Peering BOF was the worst I've been to. The YouTube presentation seemed pointless. Sure, I learned that Lundy isnt the best colo around and YouTube is struggling with IPv6 on their gear (huh). Topics strayed away from peering except for talks on net neutrality, which proposed potential scenarios and not options for content providers and SP's to work together.
Peering BOF was interesting, room was packed over capacity. Security BOF was somewhat less compelling than opsec WG BOF.
The security BoF was specially interesting. Regarding the BGP tools BoF, I'd suggest to use a smaller room and not the main plenary room. Smaller rooms are better for BoF's.
I learned alot in the opssec BoF
BOF are always vry good and useful. Should be longer.
Good quality. All were interesting.
I liked the discussions at the IX and peering BOFs.
Peering BoF good but not as good as in recent times.
I thought the ISP security BoF could have used some more audience participation and discussion like you find in the peering BoF. Perhaps bring some of the techniques Bill uses in the peering BoF to the security BoF? Otherwise it and the BGP Tools BoF were useful overall.
The peering BOF was useful.
Have the Peering BOF in a larger room. Keep BOFs (like the Peering BOF) on topic.
I missed all the BoFs because the other titles sounded interesting
Found them useful. Peering BoF, Exchange Operators BoF and OPSEC BoF
Thought the security BoF's were excellent.
Please make the IX Bof a standard, and take the vendor OFF the Peering BoF. It's a major conflict.
no specific comments
Peering and Security BOFs were very useful.
BOF for Exchange Point started late due to the room being used by the Peering BOF, which ran late. Consequently, we lost a lot of people towards 5:20 because of the desire to go to Beer & Gear. We ran really late as well because the presenters (myself included) had a lot more material than the time permitted for. We more or less ended around 6:20 I think.
Opsec was very useful
The peering BoF was good. though if people identified themselves before speaking, it would help.

Did you find the public laptops near the General Session room satisfactory?
Worked fine (14) Did not use (96) Bring back the Terminal Room! (4)

What did you like/not like about the meeting venue?

I did and did not like it because it is close to my office. ;-p
hotel workers were on strike
I liked the "wide" format of the main conf. rm
It was fine, no complaints.
I live nearby, not crowded, there was a Starbucks in the building. Plenty of room to spread out in the sessions.
Organized lunch would help keep people together
Open the main room at 8:00 please, not 8:40. Spacious hallways were nice.
Good venue
Representin' South YAY Area! = )
LIKED: Easy to get to, not far of a walk from the hotel. A number of the crowd and the 'inner circle' found their way to the hotel bar, which allowed me to easily approach people and talk to them.
Expensive parking
I liked that it was close to home. I didn't like that it was close to home. (Had to go home each night instead of staying out 'til 2 am with folks. :)
Hotel food horrible
Too bright. The cozy feel form Dallas was missing. The hallway tv of the webcast should return
Everything is just fine!
- close to home; meeting room was too cold - did not like that talks ran over into breaks, esp for lunch-moderators should better monitor time of program was not put together well-to allow for enough time - the breakfast/snack food was good (granola!) - it was often very hard to hear-either sound getting lost in room, mikes not up enough, people not talking int omikes, doors remained open, people talking at back of room
Network access fees - bad deal Convention location - decent shops & restaurants nearby Close to airport
it was as expected
Hotel seemed very good. Complimentary bottled water in the room each day was very welcome. Downtown San Jose wasn't as dull as I was expecting :). What an enormous cavern to meet in! The main session room was also very cold, the air conditioning in the room actually felt draughty. Well done to Rodney and everyone involved for making it happen.
No wired ethernet in main meeting room little to no power in common areas and open seating room
Being my first NANOG, I don't have much to compare to so I'd just say in general it was positive. No dislikes come to mind.
Main meeting hall was a little warehouse like.
Not like: hote Internet access was VERY slow. Even at 3am! Small Nit (this is not unusual, and I've learned to deal, but...) I drink tea, not coffee. Tea selection & quality was poor and the water had a strong off flavor. I heated bottled water in my room to make tea.
I had no issue with the venue. Easy parking and I hit no real traffic
too big for this group - but the curtains helped I liked the small peering room - keep it intimate
I did not even know there were public laptops available. Definitely no use for terminal room.
San Jose is great. Cavernous and empty as if we were having it in the ruins of an ancient castle...
San Jose is not exactly an exciting town. SFO would have been better. Great Weather!
The screen in J2, J3 cannot be seen clearly. Words of some presentations are too small.
great location. easy access to local freeways and eateries.
Since this is my first I don't have much of a comparison point. Thought the location was good as there is plenty of food within walking distance of the hotel.
Tables in the BOF/Tutorial Rooms(like in Exhibit 3) would be nice
nice area, easy to get from the airport to hotel. good food accomodations around the area (walking distance).
hotel was too costly; charged unreasonable fees for merit cabling
I appreciate all of the effort & sacrifice that people put into this conference.
OK, didn't use conference hotel as NANOG rate had already expired on the final day. Found a nicer hotel anyway.
San Jose is kind of a wasteland, esp at night
it was fine.
the lighting on one mike left the speakers in dark shadows. for speakers without slides on the web, please focus more time on their slides than their face. the webcast access is generally great, but I had a few times were the session dropped, so I missed a few minutes of presentations. Also, I found the window media stream to have better sound and be more reliable for me on day 1, so I stuck with for day 2 and day 3.
San Jose is played out, however, downtown San Jose offered everything you could want. Was not hard to find a place to eat or anything, bravo.
Was oddly spacious - very unusual for a nanog.
Good access to food.
Convenient to fly into and out of. Mild weather, nice hotel.
BOF room too small
I'm satisfied, liked the interaction in the hallways.
Much like other NANOG venues - big hotel. I liked it that it was central and you did not have to travel far to get into town.
Location was relatively convenient.
Great place for a meeting. Should come back here in the future
The lighting in the General Meeting hall was horrible. I did not like having the General Meeting in a convention hall. There was lots of echos and noise.
San Jose is kind of "meh", but the location was convient.
I like being on my home turf. Great local restaurants. Lots of space
Liked the water in the rooms, like the location in downtown San Jose (enough restaurants, attractions, etc). Enough locals around to tell us which places were good, etc. Would move the Peering BoF and Exchange BoF to be on different days or at least further apart, both seemed to need more than the 1-1/2 hours allotted. Beer and Gear/reception space was a little too open.
Parking info/costs. I parked in the SJCC parking, but I won't do that again given the $18/day fees.
Rooms were a little small for the BoF's.
This was good.
Easy to get to, very close to freeway for easy access. Rooms were big enough for the meeting.
Icky lighting. Excellent reception food. Poor hotel service.
The venue was fine.
Good meeting venue.
Wireless did not reach the hotel rooms. Hotel charges for in-room internet.
Good central location in Silicon Valley.
it was local for us to attend.
BoF/tutorial layout (U shape) was not useful
The starbuck coffee was bad.
1)Variety of restaurants are within walking distance. 2) Close to San Jose airport.

What worked well and what should be improved for the next NANOG?

Security BoF was excellent
I liked how similar topics seemed to be grouped together into blocks. I, for one, was interested in Monday afternoon, the Beer & Gear, and Tuesday morning's topics, which was fortuitous as I could only stay for that small amount of time anyway. Good scheduling, from my point of view.
Liked the long, in-depth talks. The best as usual is the people networking. Would have more lightning talk sessions or maybe a session that is in between long talks and lightning talks instead of a tutorial
more tme for Peering-Bof, as there were a lot of people attending - and include the IX-issues together in one bof.
Bring a networked printer
Pad the agenda, and/or hold speakers to their time slot, and/or use longer time slots.
Good weather is always desired.
3 day format
Worked well: connectivity! thanks to all involved given the challenges.
1. It would be great to have a physical bulletin board in the lobby that everyone can use to facilitate communication. 2. On the registration form, you might want to add an organization category for IX & colospace provider. Or you may want to expand the category to be Internet infrastructure provider. It seems to me that participation by these guys is growing and becoming significant.
Rodney is a gem and the tshirts were funny. Fewer people on the panels and more time for questions would help.
Lightning talks need just a little more time
Liked it better when main session was morning & afternoon & BoFs were in evening
Good to see that the community meetings are still being well attended and the community remain interested in the workings of NANOG. It's reassuring to see that challenges of the last couple of years weren't in vain. Don't squirrel open seating off in a side room - it needs to be in a foyer/circulation area. Glad to see that this was sorted by the second day here. Try and get power to tables in the open seating area.
I think to incorporate a broader sense of community into the community meetings, some form of VoIP teleconferencing or even video conferencing should be implimented for valued input from members of the community that were unable to attend in person. This can be patched into the A/V system. This seems somewhat trivial to setup initially and can be refined over community meetings to come. We are network engineers after all. It was just surprising to me the lack of interactivity from members watching the live streaming video with all of the technology we have. P.S. Don't put the question "How did you attend NANOG?" on the paper version of the survey form handed out at the event! :-P You might as well put "Are you filling this out on a computer?" on the web survey.
Community meeting: Presenters need fixed times - point of oder!! Maybe the finances don't need to be read line by line unless there is an item for discussions. We need to understand where did we loose and a small blurb why, and where did we succeed, and why.
Tools BOF was good - please continue
I switched to EUDO after an hour or so. Wifi was slow for me.
More time for the Peering BOF - should be 3 hours! More discussion and interactions Have a NANOG Social Event
Add more mics on the table when you have panels. One mic on each side was inadequate. Otherwise, I thought it was pretty good. The lighting for the speaker I thought was not very well lit. But if I were a speaker, I would not like to have light in my face.
Something must be done about the propensity to run over...who approved 40+ slides for a 30-minute presentation (IPv6 routing issues)? But that's just one example. Build in question time, force people to stop with a warning (5 minutes left!) system, etc.
No big hosted socials...those were great at past venues and should always be pursued.
Don't use yellow for the T-shirts! It's OK to get experiemental, but yellow? At least the orange shirts at NANOG 28 had cool embrodery.
I'm not sure why most of the people attended as very few seem to pay attention to the speakers & presentations. I know we are all busy, but if this important to thme or their employers, stop the tapping away on the laptops.
the separation of the general session track from the tutorial/bof track, and how it was scheduled, i thought was fantastic.
I felt there was a better theme feeling to each days sessions this time than many previous times. In general, all the talks were of excellent quality. Also, the researcher talks were more relevant to real-world problems than in the past. The panels weren't as good this time, but the data center one was pretty good by dividing the two camps on each side of the podium.
Venue did work out well. Randy Bush needs to *STOP HOGGING THE MICROPHONE*. He holds up the entire agenda by being one of those people who always has to comment on something, and the comments have no substance at all typically. Seriously, just get him a muzzle, or just please ask him to tone it down. A lot of people are getting tired of that crowd that hogs the mic's in the general sessions and the BOF's (namely peering bof).
many sessions ran long - perhaps time estimates for the duration of a presentation and q&a need to be examined closer. perhaps this implies we have more content than we think we do.
Perhaps build overrun time into the schedule. Each day's plenary session ran long by some amount of time.
I'd like to suggest an effort to make all presentations available onthe website.
More power plugs Do a joint ARIN/NANOG meeting in Spring and Fall
Operational Experience with TCP & Anycast was focused on content providers. It was nice to see. Should be set time for Q&A after each 10 min lightning talk (ie- 3 to 5 min)
The hotel internet connection didn't work well, and the NANOG WiFI didn't reach my room, so it would be good if there was plans to provide fully working internet in the rooms as well.
Try and keep up the community. More social events with timing like Beer and Gear but more RIPE style, so no gear, just sponsors. Its a good opportunity to get everyone in one place (free beer) and to get around and meet everyone.
Find a way to prevent welching hosts.
As usual, I like the tool talks.
Would limit the exchange updates in Exchange Operators Bof to no more than 3 and 5 slides each to save time for more meaty issues. Need to allow time for peering coordinators to exchange cards -- schedule went over and led to many people leaving for break before cards could be excahnged. Perhaps have a space like for PGP key signing that is for peering coordinators to exchange cards?
Meeting schedule up sooner...
The format works well and I don't think anything needs to be improved.
Lightening talks were good!
I think we should get "real" keynote speakers. Famous people. People to come and make visionary speeches. I think this would help us get back some prestige.
I didn't use the laptops, only the printer, but put the printer on the net.
While peering was encouraged, I dont feel that I picked up the level of peering I was looking for. I did make the contacts that I wanted though.
Not having the NANOG meetings in a long time, I found the scheduling to be a bit jarring - but I'm a dinosaur. On the one hand, I like the spreading out of the tutorials and BOF's over the successive days, but on the other hand, I miss having the Sunday, Monday, Tuesday single-track marathon. I'd have to attend another NANOG to decide which way I'll swing my preference for. :-) Anyways, I think Bill Norton's Peering BOF really needs a separate venue where it can continue to run on with contraints or running into another BOF scheduled for the same room. Yes, I know, another room is an additional cost. Speaking of room and cost, question: While the tutorials and BOFs were in the smaller rooms, the big room for the general session held in the mornings went unused in the afternoons...was NANOG on the hook for renting that? If so, a late-running BOF could have been warning to the organizers (both NANOG and forthcoming BOF moderators) to move the next BOF into the big room. Otherwise, the big room in the afternoons sitting empty seems a waste. Just an observation.
Stick to the schedule. Sessions were consistently late. The general session was very weak. Much fewer topics compared to previous meetings. Very weak Beer N Gear reflects the fact that folks (including vendors) lost interest in this conference
the original 3 day schedule would be nice (though with the Arin meeting being held too, i head that it would be)
More operational/engineering presentations would be useful. Good that BoFs didn't run so late or overlap
one more social event ?

Do you have suggestions for future NANOG presentations? (Topics and/or speakers)

more on IPv6 identity vs locator
More how-to's
Using solar and/or DC power in datacenters (see AISO.net for example).
This is part of a larger issue of NANOG's identity crisis that has been lurking for the past couple years.
Keep up IPv6/Global scaling work
Part I: What I want from an IXP. Part II: Responses from various IXPs. Panel? I might be interested in doing this presentation if no one else will. (It might be seen that I'm biased. *shrug*) The first part would be an adaptation of Bijal's work at RIPE: http://www.ripe.net/ripe/meetings/ripe-48/presentations/ripe48-eix-isp-ixp.pdf
More "practical" topics.
Internet 2/Transit RAIL/NLR/Startap. A summary of what is up with so many research networks. Dave McGaugh, Darrell Newcombe, Yvonne Hines & maybe Startap lead for a panel?
Invite back Microsoft/Google/AOL & Yahoo to talk about Spam/Operational issues
1) Maybe a tutorial for IPv6 as it will help us to test before implementing if we will move towards it in near future
IPv6 including continued address of the multihoming issue.
We really need to harness some of these researchers from academia to do useful things for us, rather than lots of similar/repetitive work. I wonder if there's some other way of doing a research forum - maybe a research BoF session, where ops can talk with researchers and get them hooked on our problems, rather than them repeating all the same things, and coming out with the same answers or different versions of the same tool.
I'd like to see more on network scaling and capacity planning...
Yeah. DDOS attack tracking, IPv6 Deployment, Botnets Disabling
P2P technologies and characteristics
IPTV; maybe a panel with vendors, exchanges and service providers...lots of possibilities.
Bring back the power panel each NANOG and show how this issue is progressing. get some presentations on future cooling technology. Dan Golding did great!
Would like to see some topics related to metro ethernet. Such as initiatives in the MEF: http://www.metroethernetforum.org/
A suggestion was raised about the legal implications and requirements for security incident handling.
would like to see more presentations on the datacenter/facilities/physical plant side of the house. glad this was touched on at this conference.
I always like to hear from Randy Bush, Steve Bellovin, and Paul Vixie. More talks by Vince Fuller would be welcome. How about bringing back the "great debates" idea from the early Interop days? get two people to argue a different side of a controversial topic, like end-to-end networking? for presentation my topics of interest: net neutrality, data center power/cooling, ipv6 multi-homing, WAN acceleration, latency intolerant applications, and end-to-end performance monitoring.
Yeah I'll get to that...
Hey, these guys seem interesting: http://www.firstmilesolutions.com Also: http://www.sctdv.net/ http://news.bbc.co.uk/2/hi/technology/3489932.stm
IPTV and video distribution, voip/session border controllers/real world issues
Content provider BOF Content provider at the peering points.
Not really. I think the community drive folk should keep up the good work and should continue to engage the community. Go Bill, Randy etc. I think Merit staff on reception do a good job. They are very efficient. (sorry no other place to put this comment)
Generating BGP configs using IRR Toolset, etc.
Would like to see some discussion of CALEA and other legislation that might significantly impact operation of networks and burden on staff etc. I like the discussion on COLO/vendor and would like to see perhaps a talk for newcomers (similar to Bill Woodcock's talk about Internet Exchange in a Nutshell) that goes through process of picking space (of course this should be given by someone who is NOT a colo vendor).
Physical capacity & mapping (i.e. fark fiber routes and providers) as viewed by NOT The Usual Suspects. In short, what unusual capacity is out there and at what cost.
iPv6 (performances, transition mechanisms,...) deployment
Better qualify the speakers, especially on panels. Also, drop political statements. The Net Neutrality thing was a little over the top in terms of personal opinion of the moderator - and the political comments were un-called for.
Exchange Point Operators Tools BOF. Or, at least have more IXP BOF's to include tools and sharing of tools to foster collaborations on the permanent agenda.
A bit more on routing operations. Non-academic, real world ops stuffs.

How many times per year should NANOG meetings be scheduled?
Existing Schedule (45) Fall, Winter, Spring (11) Spring/Fall (47)

Schedule comments:

3 times a year seems to pull an appropriate amount of papers, research, and presentations
I like the sun-wed sched, there should be about enough info to fill that time 2-3 times/yr
I randomly picked the existing schedule, since I can't really attend that often anyway because I have childcare issues. Now, if you provided facilities for onsite childcare (as discussed on the net-grrls list), I might be able to attend more often if I could bring said child out of town with me.
Three meetings, of which only one is joint with ARIN, means everybody has to make four trips a year, instead of tow. And there's not enough good content for three
As a non-north-american I would like to reduce it to two
January-March sucks, December is out. With eight months left over, fitting three in doesn't work well. Do April and October, and move on. Oh, and I'm sick and tired of hearing Merit squash every attempt to keep it at three - suck up and listen to the community.
3 is too many to attend per year
Two times a year meand too long between meetings. I cannot make every meeting so if you assume i miss every other one it becomes less useful
tough questions. there were great points presented on each side. honestly, I believe most people will not care and will still attend. there seemed to be a belief that attendance/revenue would fall if only held twice a year. this may be true. however, it would be interesting to see if the same 400 people are showing up to the three meetings a year. I suspect that attendance would actually *increase* because the value of NANOG would be greater. I do not make it to every meeting ( although I would like to) I suspect there are some poeple like me would would be more likely to hit NANOG both meetings. but again, I think whatever is decided ppl will still show up.
While financially it probably makes more sense to have meetings twice a year instead of three times, I think the relationships people build, and future relationships that could be made could possibly suffer.
about right. I usually only attend one.
Regularity. Please schedule away from RIPE and Apricot.
Two We're not coordinating the net anymore!
Budgets are easier for si-annual, one with (ARIN & NANOG) and a stand alone at a fixed location
-Easier to justify travel expenses -Try NOT to schedule NANOG near other events ie Fall NANOG very near VB Montreal
Existing and Fall, Winter, Spring are the same I'm neutral. most important is to know dates as far in advance as possible
3 mtgs/year is the right frequency
I think the current schedule works. Dropping to two meetings per year slows things down (activity often "clusters" around the meetings), and if you miss one meeting, it can be a whole year until you get the same people together again. Look for other meetings to "pair up" with (beside ARIN).
I'm honestly not that familiar with the existing schedule vs. Fall, Winter, Spring other than they both occur three times a year. I would think keeping it at three times a year would be beneficial and keep the community tighter and offer the ability to speak on recent significant events before people forget about them. Raising the cost to attend would probably be the most logical thing to do. My company sees a value in sending me to these events so unless the cost is astronomical, I don't see why they wouldn't be willing to spend additional money to help offset Merit costs. You get what you pay for. I find NANOG to have much value.
The largest cost is the travel. Staying an additional day is insignificant. Two meetings a year enable more freedom to attend other conferences. The joint meeting with ARIN is an excellent success. Fall meeting should be earlier than later as large corporations tend to lock the expense budgets around Oct if their targets are slipping. Late September or first week Oct would be much better than last week of Oct. A couple of weeks can make a big difference in attendnce for my company.
New Internet behaviors show up too often to go to twice a year meetings. I don't see much difference between "Existing" and "Fall, Winter & Spring"
I like the new schedule. I am better able to attend BoF etc. In past there were evening conflicts.
Don't see too much new topics between previous NANOG and this one :)
Whatever it is keep it the same for predictability of budgeting
If it ain't broke...
There are too many meetings as it is. Apricot, RIPE, Global Peering Forum all need to be attended for different reasons than NANOG. Spring and Fall would increase attendance and focus. DO NOT EVER DO BACK TO BACK WITH RIPE! This schedule will suck this fall.
I generally try to attend all the NANOGs, however I seem to consistently skip the winter session. February just feels to close to the October. I think part of it has to do with so many of the US holidays happening at the end of November, December, and early January. Little "work" is actually getting done at many companies, and I don't feel that I have anything new to share by February. Also - location is a big deciding factor. I skipped NANOG 33 mainly because I can't stand Vegas.
Don't have strong feelings either way. For me it's easier to make two meetings rather than three meetings.
Less meetings with more content. Less schedule machinations for each set of conferences. (arin/ripe/nanog/work...)
two per year is sufficient, easy to budget and plan travel.
I've attended NANOG twice and have been impressed by both the breadth and depth of content that is timeless alongside the talks on current issues. It seems like there is enough content to support three meetings. Even if I can't make all of them its nice to attend via the web for things I care about.
I don't think there are many of us that travel round the world going to conferences and it is important we do not changee NANOG to suit ourselves rather than the majority of attendees who do not go to overseas conferences. Since I'm not a North American Network Operator, however, I don't think my opinion should carry any weight on this topic!
three times a year was always overkill (ok ok - except during the boom). twice a year (once with arin, once on its own) is perfect. will raise the signal:noise ratio and guarantee higher attendance (which == better social networking opportunities)
The existing schedule works well for me. but I don't really care if you change it.
Two is better, three is just too much.
2 Meetings per year would make MERIT lose more money while making the attendees more interested in attending.
Travle budgerts are tight as is available staff time. Often I attend not just on my own dime but on my own TIME (vacation, etc).
That's just fine
I'm poor prefer less meetings to attend
NANOG should only be twice a year! This will increase attendance and the amount of fresh topics
I'm finding the 3-day schedule and 3-times a year layout pretty useful. SInce a 3 conference a year schedule spaces things out quite a bit anyway, I don't see where dropping down to only twice a year would offer much advantage. Sure there you would some overall travel cost savings for folks, but not significant enough to make it worth a change.
Winter is likely to cause travel related problems. With 2 meetings a year, face-2-face communication will be less.
Well three meetings. I travel from UK but I think to lessen the no. of meetings will mean even less of a community at the fewer meetings.
Seems to work for me thus far.
bring back the old Mon/Tue schedule
I think staying w/current meeting schedule is important
I would prefer to have the meetings for NANOG and ARIN take place on the same week in the Spring and Fall. Plus, by going spring/fall you can hopefully avoid weather extremes (Dallas in Winter, Phoenix in Summer)
I'm likely to only attend one mtg per year, so more than two is superfluous for me.
Fewer=Better. It's difficult to keep getting approval. Stuff goes across mailing list anyways.
Two meetings a year would be acceptable, but by having three there is the possibly of reaching more communities. I strongly feel that the meetings do need to move around (and would also suggest a meeting in Puerto Rico or St. Thomas if possible). It is possible to do multipe year agreements for alternating years. Also, we might want to investigate whether particular hotel chains would be amenable to creating a master agreement that would allow us some flexibility with networking, etc.
Perhaps Fall/Winter in person and Spring via online attendance?
I spoke at community meeting re: this.
Travel is a pain.
I liked the diversity of the talks. I fear that if we were to have less meetings a year there would be a lot of good topics thrown out that we would not be able to see. This would then give nanog less value.
Twice a year is probably enough.
Need to keep short frequent meetings to have relevancy to current issues.
Bi-annual will cut NANOG costs, but this is really more of an operations forum, tutorials and the time alloted to them, notwithstanding. Just because of that, I could make the argument that NANOG should be quarterly - but it may be too much. Three seems the perfect number and frequency. Also, is there an economy of scale if more NANOG's are held jointly with ISOC or IETF meetings whenever possible?
Don't care as long as it is not too close to the IETF dates.
Too many meetings means lower attendance. Fewer meetings but longer may work better.
Three meetings are too much. Not enough occurs between meetings, fewer submissions won't hurt and companies may allow people to attend all meetings if there are fewer
Three times a year makes it a lot of meetings to co-ordinate, as there are other meeting in Europe and Asia which needs to be attended.

Preferred meeting locale:

SF Bay area
the current schedule (most general sessions on a half-day schedule) works well near silicon valley and east coast corridor. people can convince co-workers and manager types to leave the office for one half day of general sessions, BOFs, tutorials etc
Weather & cost are primary considerations :)
Well, I am partial to the Bay Area (Oakland was a good choice from my point of view--too bad I was getting married about 1.5 miles away on the very Sunday that NANOG started!). Obviously the Bay Area is not going to work for everyone, though. Other west coast cities or Arizona, Colorado, or even Tulsa, OK would suit me personally. :)
West coast
More West Coast
at least one per year at the east cost.
Europe *d+r*
Location needs to easily accessable to most people by non-stop air service. Locations without this will suffer loss of guests. It would be good to rotate east/west alternating for meetings and you should pick up more people LAX, SFO, CHI, NYC, DCA, DEN, DFW are all good choices. OVERALL price is a big considderation not just cost of room or cost of admission
I actually enjoy going to different cities around the country, but I have to admit I tend to go to the West Coast ones more.
West Coast LA$ VEGA$
At least on/yr in SF bay area.
One specific "home" location per year helps budget planning.
I would prefer west coast, but at the same time, I would prefer for the meetings to be held if geographically diverse locations to facilitate attendence.
Northeast (NYC area)
East & West
Fixed in California or Virginia based on attendees zip codes from pat two years
Having a regular "anchor" site would be great, with rotations for the others provides significant value
1 East Coast 1 West Coast
You should skip Dallas!
West Coast
west-coast and mid-west. Canada in winter is a dumb idea.
I don't think there's a problem with having one "regular" location each year, one which is pre-wired with connectivity, and in a popular and easy to get to location (e.g. an airline hub with good selection of International flights) - especially if this eases meeting logisitics. For example, I thought San Francisco (NANOG 31) worked really well. Good facilities, didn't seem to be hideous connectivity problems, if people didn't want to stay in the meeting hotel there were plenty of other options, and there's plenty to see and do in the evenings. Seattle Westin is a similarly good option.
East coast is easier for me but the world doesn't revolve around me. Makes no difference as my company foots the travel expenses.
Very nice to rotate when possible. East, West, CANADA (The exchange rate can be sometimes prohibitive to come into the US but works well for folks to go North) - We have been a little West Coast heavy lately driving the cost of travel for those out East.
Don't care
East Coast
keep it moving around
No comments.
Hawai'i?
East Coast and West Coast.
I really appreciate the attempts made to create a rotating location around the US. While I'm in San Jose, I actually like getting out to other locations. I think we are about due for a NANOG in Canada.
West Coast
3 meetings/yr gives a good variety of possible locations Wireless network was fine for me.
no preference.
NYC area: not done before, would be great location
Mid-West
West Coast (I'm located in the Bay Area). East Coast or Mid-West is fine as long as there are sufficient airline options (both carriers and flights).
East & West Coast, otherwise tier-1 cities, not places like St. Louis...
Bay area and East Coast (not Reston) would be good.
pacific northwest
No preference, but try to repeat at least once a year in cities that people showed the most favorable responses to (Chicago, Reston, SF/Bay area).
West Coast
No location preferences.
West Coast in Winter, East Coast in fall
It may be handy to permanently situate one NANOG in the DC area, and one in the Silicon Valley area. If the decision is made to have three NANOGs per year, then allow the remaining session to be variably located.
No preferences at all. I'm from Brazil and it always take too long to get here :)
West Coast!
Hiltons
East Coast/West Coast
All parts of the country are reasonable. I might suggest partnering with a given chain of hotels (e.g. starwood, hilton brand, etc) to perhaps get discount/price breaks but still allow locations to be different each meeting.
Since I travel from europe, I prefer it to be where there are good international connectivity.
East coast please please please
No preference.
east coast
I believe one West Coast (e.g. Phoenix) and one East Coast per year would work well.
West Coast
West Coast. We have better weather than anybody else
Move it around, include Canada (Mexico?, Puerto Rico?, St Thomas? Belize? These may be hard to get sponsors but would be reaching to group of operators that have trouble getting travel funding to attend meetings in US.....
Schedule in areas with high volume (lost-cost, highest-choice) air transport.
West Coast, but that's because I live on the West Coast! ;-)
South-West
East Coast.
San Francisco bay area
No preference, but give some thought to airline availability. I'm very willing to go to Canada--Vancouver is lovely
I'm ok with nanog being hosted anywhere in the US, provided there is sufficient notice. Is it possible for Nanog to partnet with 1 specific chain for multiple events in order to get cheaper rates?
NANOG in San Diego, and Seattle would be nice.
Please keep it mixed.
This was being done with USENIX (I was a long-time attendee) at some point between 1998-2002 - not strictly, but pretty closely. They would be the best folks to realte their experiences. Why learn by ourselves when others have gone before us? Mind you, they get more attendees (2000+?) so reserving venues into the 3+ year range in the same spots ensures they get a proper venue. I was going to say predictable but costs rise every year regardless, so the budgets can't be fixed.
West coast
no preference
Keep rotating venues
heh!! in the Bay area..

Suggestions for future NANOG hosts:

Hmm...I guess Williams/WilTel is out of the question now...what about Level3?
I will be glad to provide venue consulting/engineering asssistance if needed.
We are very much interested in hosting NANOG at Las Vegas. We are a colocation provider in Las Vegas, and have numerous relationships with a number of the hotel/casino's. We have spoken with Betty, and will contact her within the next week or two to continue the discussion and getting a list of the requirements.
Sylvie.LaPerriere@vsnlinternational.com +1 514 868 8539 nanog 39
As me again in a few years! :) (I would happy to help on a personal basis as requested.)
Comment on question # 4: How is this question useful to you? You don't know whether I liked or disliked previous NANOG mtg or which one I attended Misc Comment: Can Steve & Rodney post opening & closing remarks. I'd like to include demographics in my meeting report
Any thoughts to raising the price of registration to offset sponsorship particpation? At $350 per session, NANOG is a real bargain. Maybe introduce a 3 tiered pricing structure: Student/Teacher, independent consultants, big corporations. Just an idea.
no
I am interested in sponsoring and would prefer to do free beer in the evening than a break session etc. Vanessa Evans vanessa@linx.net
Aliant in Canada should host one
Yes! PureGig would be more than happy to host a future meeting in Phoenix AZ. Please contact me (Neil Davidson, neil@puregig.net) or Shawn Freeman, shawn@puregig.net.
Possibly-Stafford A. Rau Integra Telecom, Portland, Or stafford.rau@integratelecom.com
Microsoft might do it in Seattle again (talked to some people last night) possibly in conjunction with the higher ed community there. Also overheard a couple of people saying they wouldn't mind Seattle as a permanent location.
Contact me with information on what it takes to host NANOG. Siggi Bjarnason, Microsoft Windows Live Network Operations, siggib@microsoft.com
We in Ottawa were at one time very interested in doing it, but we got spooked.

Return to NANOG home page.