Japhy, Good for you in recognizing a need and being willing to go after it. I've been involved from time to time with non-profits, and there are a # of advantages, but also some disadvantages.
If you haven't done it already, here's some steps/questions that I'd recommend that you'd do: 1. You are in an area that does have Internet access available...what need are you looking to fill? Digital Divide? Public Internet space? Increased competition? All of the above? 2. Contact some of the community wireless projects that actually succeeded. CuWIN (http://www.cuwireless.net/) is a good one and not to far from you, there are others out there. Sascha Meinrath ([EMAIL PROTECTED]) is a name that I've seen frequently in that space and might be a good point of contact. 3. Nail down a business model. Run the numbers. This is true even if you're doing a non-profit--you still have to be sustainable and, ideally, "profitable" if you want to grow it at some point. A lot of non-profits forget that a non-profit business is, in the end, a business... For-profit has advantages 1. The legal/tax work involved with a non-profit is substantial, and much more so than with for-profit companies. I know of a couple of non-profits, though, that help with this. 2. Many WISPS, as mentioned by George, are legally for-profit companies even though they don't really generate any profit or likely ever will: but, a lot of people in this space are motivated by similar motivations as yours--community, etc... 3. It (somewhat) forces you to better build the business model. 4. It is easier to get investments, however, my sense is that there's really not much investment money for WISPs available, unless you do some really good selling around your community. Non-Profit also has its advantages 1. Use of volunteer labor. This is a _lot_ of work, though--attracting, training, and keeping volunteers requires a lot of effort and creating a good workflow to cover these areas. However, given the scale, non-profit may really be good to lighten the load in terms of upkeep / maintenance. 2. Easier access to grant money. 3. Donations. All of those are a lot of work, just so you know, and, in the end, require building a business process. As a general rule, I'd avoid building a non-profit that runs on donations unless you're really good at getting donations and/or you've got a cause that hits a very small sweet spot. In the end, donations are very unpredictable and drop rapidly during economic downturns. This is especially true if you're dealing with a smaller community, since you'd have to get the same people to re-donate year after year. So, while it may be worth getting donations to cover startup costs, it isn't really advisable to figure that in your operating costs... Technology wise, keep a couple of things in mind: 1. If you can get the funding to do it, fiber might be a much better option, especially in terms of economic development. It also can be more viable in the long run since it can be more easily monitized and has less frequent upgrade cycle. 2. Often the problem in coverage isn't from the access point, but the client speaking back to the access point. So, while your radio may be able to go 5 miles, a laptop's wifi card can't even go a small fraction of that \. Generally, most WISPs give/sell/lease some sort of customer premise equipment to the customer that has at least a somewhat engineered link back to your access points. This is more of a point to multipoint architecture (one access point speaks to multiple clients). If you want to provide the service of someone opening up a laptop and hitting an access point, you'll need to have an AP pretty much on each corner block or so; people in buildings will still likely need some sort of radio. 3. The term mesh can mean a couple of different things (utilizing the same concept). In a generic sense, it is the idea that each node can have paths through multiple nodes to redunancy purposes. - 1. Access point / backhaul mesh. So, if you setup 4 access points, and make sure that they can all talk to each other in addition to their clients, then that can give you additional redundancy on your backhaul links. This is also done on a block by block level, as mentioned above. Regardless, this is an "engineered approach" where each node is carefully setup to talk to particular other nodes and some link engineering is required to make sure they have good wireless links, etc... - 2. Some network architectures (Meraki pushes this approach) use an approach of "throw a lot of cheap hardware at the problem". This, in the end, is more of a solution for the _client_ portion, and not the access point portion. So, instead of a WISP putting an access point on premise that is carefully lined up to get a good signal back to the access points, you just give a lot of customer's these "mesh clients" that talk both to the access point and each other and (usually/hopefully) the software can work out a path back to the Internet. WDS? Don't do it. It is a way of doing mesh, but it doesn't work well at all--not scalable at all and horrible performance. Clint Ricker -Kentnis Technologies On Sat, Apr 5, 2008 at 1:33 PM, Japhy Bartlett <[EMAIL PROTECTED]> wrote: > I've gotten a couple private responses to this thread, so I'd like to > throw out a few more things: > > The overwhelming advice though, seems to be to avoid mesh. ( Also, > Tony Morella of demarctech pointed out that I misunderstood the olsrd > literature.. sheesh. ) > > So, I'm basically working in a pretty dense, suburban/urban area, > sitting in a valley. From where I'm typing, I've basically got clear > LoS to the major areas I'm trying to cover, maybe 20 square miles. My > logic was that to provide consistent coverage over all of it, I'd need > 3 or 4 overlapping APs.. which seems ideal for a mesh setup. > > If I'm using a dual radio backhaul/AP setup, am I going to get > interference between the units? > > What's the difference between WDS and mesh? > > A few people have mentioned that the standard for free access is to > have users pass through a portal with some sort of legal disclaimer. > Is there actually any legal precedent for suing a hotspot provider > over the actions of a user? I guess I'm just being naive, but that > seems.. silly. > > Regarding CALEA - I understand recording VoIP for the authorities, but > what are your responsibilities before you receive s request for that? > Just to have the tools in place? > > A lot of people mention using DSL instead of a T1. There is a very > large international corporation headquartered here, renting their > internet T1s for $20/mo (!!!). Most likely, as a service to the > community, they'll be able to negotiate a similar price for me. At a > rate that low, it just makes a lot of sense, but I am getting quotes > from the DSL providers. > > > On Thu, Apr 3, 2008 at 8:05 PM, Chuck McCown - 2 <[EMAIL PROTECTED]> wrote: > > My question for you is "why"? 100 users are enough to be a headache and > use > > up all your free time but not enough to even pay you for your time or > > investment. > > (Did I mention I'm writing a grant proposal for this? Cue the strings :) > > It's not really about money. As a state, Michigan is fighting to keep > people from moving away, and as a city Benton Harbor is struggling to > retain or attract the kind of talented people that can sustain some > sort of economy. In the meantime, most of our citizens are > poor/unemployed and uneducated. Providing internet access is a good > way (imho) to at least open some doors for them, and put another > bullet point in the list of reasons to come visit/live. > > > Anyhow, thanks a ton for putting up with stupid questions. The advice > from this list is invaluable. > > Japhy > > > > > > > > ----- Original Message ----- > > From: "Japhy Bartlett" <[EMAIL PROTECTED]> > > To: "WISPA General List" <[email protected]> > > Sent: Thursday, April 03, 2008 5:44 PM > > Subject: [WISPA] Feasibility of a "non-profit" WISP > > > > > > > Hi all- > > > > > > I'm finishing up a grant proposal to build a wireless network for a > > > smallish (2 sq. mile) neighborhood. My only real experience is having > > > read this mailing list for a year or so, and independently researching > > > via the internet, so I'm hoping some of you veterans can offer some > > > insight before I get in too deep! > > > > > > The basic, mile-high premise is to build a 2.4ghz mesh network, using > > > Linux, or at least Open Source Software, wherever possible. I'm > > > pretty comfy with Linux, and it seems like the route a non-profit > > > should take. > > > > > > Since the idea is to be providing access not only to locals, but also > > > to people visiting (it's a mixed commercial/residential area), > > > sticking to the 802.11b/g protocol seems like a good way to make sure > > > strangers can get on with whatever gear they've already got. > > > > > > Specifically, I'm looking at gear that would run olsrd > > > (http://olsr.org); more specifically, demarctech.com's RWR HPG units > > > (https://www.demarctech.com/products/reliawave-rwr/rwr-hpg-15a.htm). > > > > > > The "business" model is to offer capped speeds for free, and uncapped > > > speeds to subscribers. So, presumably I'm going to need to do some > > > traffic shaping. > > > > > > the RWR unit lists both "OLSRD Mesh" and "Bi-Directional Traffic > > > Shaping with QoS (VoIP) via IP or MAC" , so it would seem to be ideal! > > > But how do those features play together? Can I assume that the unit > > > is running some sort of *nix with a shell? More importantly, am I > > > going to be able to script something that will link the traffic shaper > > > with a database of MAC Addresses? (Or script something to assign IP > > > address subgroups based on MAC!) > > > > > > Anybody ever done this before, or have a better solution for a tiered > > > network? > > > > > > I'm estimating that we'll have 100 users tops. A while back I looked > > > up ratios and figured that a T1 would probably be enough bandwidth for > > > something that size, but what else do I need in my NOC? Can I get by > > > with a Linux box, or do I need a heavy-duty router? > > > > > > I suppose a lot of this is best answered by the vendor, but it's nice > > > to hear from someone who isn't simultaneously trying to sell you > > > something. Any criticisms/anecdotes/advice are greatly appreciated! > > > Who's going to sue me and for what?! > > > > > > Cheers, > > > > > > Japhy > > > > > > Benton Harbor, MI > > > > > > > > > > > > > > -------------------------------------------------------------------------------- > > > WISPA Wants You! Join today! > > > http://signup.wispa.org/ > > > > -------------------------------------------------------------------------------- > > > > > > WISPA Wireless List: [email protected] > > > > > > Subscribe/Unsubscribe: > > > http://lists.wispa.org/mailman/listinfo/wireless > > > > > > Archives: http://lists.wispa.org/pipermail/wireless/ > > > > > > > > > > > > -------------------------------------------------------------------------------- > > WISPA Wants You! Join today! > > http://signup.wispa.org/ > > > -------------------------------------------------------------------------------- > > > > WISPA Wireless List: [email protected] > > > > Subscribe/Unsubscribe: > > http://lists.wispa.org/mailman/listinfo/wireless > > > > Archives: http://lists.wispa.org/pipermail/wireless/ > > > > > > -------------------------------------------------------------------------------- > WISPA Wants You! Join today! > http://signup.wispa.org/ > > -------------------------------------------------------------------------------- > > WISPA Wireless List: [email protected] > > Subscribe/Unsubscribe: > http://lists.wispa.org/mailman/listinfo/wireless > > Archives: http://lists.wispa.org/pipermail/wireless/ > -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: [email protected] Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/
