February PersonalTelco Meeting Notes
- Feel free to add and edit these notes!
Next Meeting: MeetingMarch2002
Adam Shand's notes on the February Meeting are posted under this summary. Adam discusses how meetings are structured. The opinion of some is that breaking it up into two sessions, a less techical (general) session and a technical (geek) session might address the needs of a larger group. Other issues are also discussed by Adam.
NOTES BY SamChurchill
February's PTP meeting began at 6 pm and raved until 11:30pm (for hard core geeks and drinkers). Due to a booking conflict it was held Monday, not the usual last Wednesday of the month.
Several new faces were seen at this meeting, perhaps due to the February 19th Oregonian story by Jeffrey Kosseff. The cozy space at the Lucky Lab is ever popular and features some of the best microbeer but the question of space was mentioned again. A helpful solution was offered - trade rooms; we'll move to the main hall and the rest of the tavern can have our space.
== February PersonalTelco Meeting ==
- 6:00PM - Special Interest Group Reports
6:30PM - JerrittCollord - IPv6 Practical Crash
7:00PM - Jim Binkley from PSU - Wscan demo on FreeBsd
7:30PM - CourseWirex Presentation
- 8:00PM - Official Meeting (based on running agenda)
Reports from InterestGroups Adam questioned whether we really need a UserNodeSig, a StickerPage, a PublicRelations, a NonTechnical, a ManSig, InfoBits, CategorySig and AccessPointSig. Lucas (of the ManSig) opined that keeping the SIG intact is a good idea since it may be useful in the future. That seemed to be the consensus. While there was not much news to report from the SIGs, per se, there was news on the periphery.
New Nodes: Adam estimated that PersonalTelco has 25 Nodes operational with about 50 individuals having APs. Most are in NE and SE Portland. NigelBallard reported that there was interest on lighting up Water Front Park and that a proper 501(c)3 suitor was being sought for a second story DSL job. The node in the Lucky Lab is not running (yet). Your reporter must have been getting another cool one when the topic came up.
Pioneer Square Node: The success of the Pioneer Square Node had people stoked. About a dozen hands were raised as Pioneer Square users. Adam was in the Oregonian AND Red Herring.
Personal Telco Operational Standard: There was discussion as to what a Node should and shouldn't have to qualify as a PersonalTelco Node. The consensus:
Run NoCat (Rob's groundbreaking AP software)
- No MAC filtering (everyone's invited)
- No WEP (It's supposed to be OPEN)
ESSID would be "PersonalTelco.net" (or variation)
- No Firewall (within reason)
- No Port Filtering (within reason)
XM Satellite Radio - Possible Interference? An interesting point was mentioned by an enginner who works in the KOIN building in downtown Portland where a satellite radio repeater is located. He says there could be a problem with Satellite radio which uses the 2.3 Ghz band. XM uses geosynch satellites about 45 degrees so radio repeaters are used to boost the signal in canyons or other pockets. But the 2000 watt transmitter is overloading this engineer's RF front end on some gear and he suggested that they may cause a problem for the 2.4 Ghz band. Thanks to this kind sole (whoever he is) for point out this potential problem.
6:30PM - JerrittCollord - IPv6 Practical Crash Course
JerrittCollord promised to explain IPV6 in 5 minutes. We gave him 15 but I'm still not up to speed. Jerritt explained that IPv4 uses 32 bits for IP address space while IPv6 allows 128 bits, allowing 340282366920938463463374607431768211456 (three hundred forty undecillion) nodes to be uniquely identified on the Internet. That's thousands of nodes for every square inch on earth.
Compelling for Borgs. And anyone who has ever tried to send H.323 videoconferencing through NAT. NAT looses track of the IP. IPV6 promises true device-to-device connectivity. Everyone (and everything) gets a number. The Phone Company must be pleased. As Rob (and Adam) say, NAT makes the internet TV. Cellphones may be the driver for V6. Jettitt drew block diagrams on the board and explained how it might work in a few years. I can't wait. Neither can AT&T Broadband.
Jerritt explained that Freenet6 delivers IPv6 connectivity for end stations using IPv6 over IPv4 tunnels. Computers connected to Internet can use this free service to get connected on the 6Bone. Microsoft XP, Cisco and Linux do it. As my mother told me, "would you jump off a cliff just because your friend does?" Well, no, but IPV6 may be the shortest route between two points.
7:00PM - James Binkley - Wscan demo on FreeBsd
Dr. James Binkley described an innovative proposal that, if I understand correctly, could deliver location-based information to mobile users. The technical aspects of delivering this service was over my head but it sounds promising. I asked Binkley if an organization was to "unwire" 5th and 6th Avenues in downtown Portland would that enable handheld users to get the latest Tri-Met schedules while receiving the latest specials from Starbucks as you walked past the coffee shop? He said it would. Would that enable "free" internet access with "sponsors"? Would it be a good thing? Will James Binkley become the next Bill Gates or Ted Turner? Hey. Stranger things have happened. GPS Wayfinders lookout. Qsent has arrived.
7:30PM - Steve Beatie and Crispin Cowan- A Proposal for NTIA Grant
Steve and Crispin from Wirex and its subsidiary Immunex, presented a grant idea to Personal Telco. They proposed to help write a grant to expand wireless nodes around the community in exchange for using some of the grant money to fund their Open Source software and server appliances.
The NTIA (Tops) Grant it was claimed, is made for this sort of thing. Crispin said that he is a qualified grant writer and that the March 21st deadline was not a showstopper. As an ex OGIer, he wrote grants and his current employer, Wirex, a spin-off of their OGI telecommunications work. The grant proposal prompted Adam Shand to stand on a chair. Adam proclaimed that PersonalTelco should always be "free" and shouldn't be just an ISP with antennas. Here's more on http://www.personaltelco.net/index.cgi/CommercialUsage Commercial Usage Policies].
After several excited exchanges a motion was made to take a vote on it. About 2/3rds voted to go for the grant idea. The number of nodes, locations or types of services were not specified. Your humble reporter (Sam) promoted My University Park wireless plan as a possible model. For more information or input to Steve <email@example.com> or Cripsin <firstname.lastname@example.org>
Adam's Notes on the February Meeting
Hey All. I'm hoping that Sam will be coming up with the actual meeting notes since Christian wasn't there last night, here are some issues that I wanted to make sure were publicly announced for people that may have missed the meeting.
For those of you who were first timers last night, thanks for sticking through it, that was a pretty full on meeting! We are going to try and restructure the meetings to be a little friendlier to new comers.
After the meeting I was approached by Brian, Nigel and Lonnie. They felt that we'd missed a prime opportunity to rope in new members that showed up because of recent press (portland business journal, oregonian, red herring) by launching pretty much straight into "geek talk" with the node standardization stuff and then the urld and ipsec talks. I've been feeling for a while that meeting desperately needs to be re-organized but I haven't been quite sure how to do it. As I see it our monthly meetings try to solve all these problems and it just isn't working very cleanly.
Newbie introduction to Personal Telco (introductions, goals, basic presentation and Q&A session etc)
- Our advanced topics presentations (mid to high level tech stuff)
- Meeting business (decisions, thrashing stuff out, politics etc)
- Technical brainstorming (how do we ...)
- Drinking beer and hanging out.
So we need to figure out what we want our meetings to be, and I honestly don't think we can do it all in one meeting. This means we either axe some of the things or we have a second meeting. So if there are people on the list for whom this was their first meeting (or still remember their first meeting!) we'd love to hear from you. What did you like, what didn't you like. For the regulars, you're important as well, what do you get from the meetings that you like, what do you want to see more of?
We continued the "What makes a Node a Node" discussion. Here was the closest we came to consensus. A node:
- has an 802.11b access point
- has a standard ESSID (see below)
- broadcasts their ESSID
- has a DHCP server
- has an internet connection (may be via a connection to another node)
should run a captive/active portal (ie. NoCatAuth)
- provides open access to the internet (no WEP, no overly restrictive firewall etc)
We briefly touched on the ESSID standardization issue which has gone around the list a few times. We need to make a decision on what the standard we want to adopt is, I think it should be something recognizable as a URL so that people can put it into a browser and find out what it's all about. The options as I see it are:
- pro: facilitates roaming
- pro: is simple
- con: causes problems with dense node population
- pro: doesn't assume that discrete nodes have a common backbone
- pro: allows node specific information in the ESSID
- con: breaks layer 2 roaming unless ESSID set to "Any" (which can cause problems if there are other non-personaltelco nodes around)
- use 1 for nodes which are part of the "wireless cloud" and 2 for nodes which are logically discrete from the rest of the network
- pro: solves all the problems!
- con: is more complicated
- con: what happens as networks merge (eg. there's a NE "wireless could" and a SE "wireless cloud" that don't meet)
It was also suggested (and I now agree) that there should be two new types of nodes on the map. An "end user" and a "repeater". End user nodes are people that have a permanent connection to a node but don't share out themselves, it is basically an extension of the "Interested" node type but implies that they already have some gear and clue and could be useful for RF debugging/troubleshooting purposes. A repeater node is one which doesn't share out access but connects two or more nodes together (either via adhoc links or with multiple client connections). I have written up a more comprehensive proposal here, if you are interested in such things please read it over and offer suggestions:
Austin Schutz has offered to take over maintenance of the map server code base and assist with fixes, patches improvements etc.
We discussed using the Personal Telco web site as a kick start for the Free Networks site. We have been approached by a couple members of the Free Networks groups asking if we would be willing to donate our web content to get the Free Networks website going and work on building a global site with technical content. Many people said that they would no longer contribute if it wasn't a local site and that they felt that it should stay a Portland based site.
We discussed letting Boingo list our nodes, and everyone agreed that unless Boingo shows a willingness to contribute back to the community (which they haven't so far) we aren't interested in letting them list our nodes. I will send them a message with this message.
There was a huge debate over the "no commercial usage" statement which devolved into a total mess. The problem with this is that as soon as we start making rules about what it can and can't be used for you begin to find much grey area. My inclination is to keep the rules as simple as possible, right now the only thing I would like to prohibit going over the network is commercial re-sale of the network itself. This means no charging for *access* to the network, everything else (unless it is illegal or spam) is legit until it causes a problem. Please add your thoughts on the Wiki.
Steve and Crispin from Wirex presented on the grant which they would like to help Personal Telco get. The basic idea is that they will help us write the grant, in exchange we will use some of the grant money to fund their hacking on Open Source software which can be used by the community and they can sell as a bundled part of their product. This also devolved into a mess but in the end there were 17 people who said that they were interested in pursuing it further and no one said that our pursuit of it would "irrevocably alienate them from the project". Several people did express concern over possible outcomes though. If you want to help out with this, no of people that might be willing to offer up cash or "in kind" donations please email email@example.com.
So that's a lot of stuff!