You and I understand municipal Wi-Fi differently. I still want it to be ‘all things to all people,’ meaning that the goal should be the creation of a single infrastructure over which all uses, public and private, for a ‘community intranet’ would be supported.

True, it is easier said than done. But there are economies of scale with building a single infrastructure to deliver education, health, and civic information, public safety, electrical monitoring. I think both Larry Karisny, and Craig Settles, recently with columns here would agree with me on this — the municipal network is a local infrastructure of dumb pipes over which any type of service can run, with smart devices on the edge (that leverage cloud computing).

Mesh has its limitations. It cannot support multimedia — VOIP, Video — very well. A muni-mesh network should not have public traffic on it. But mesh is well suited, as you know, to public safety because it offers an inexpensive footprint, is redundant, self-healing.

I use a point-to-multipoint solution using the gear of a company called Altai Technologies. With the Altais I can serve out a lot of video to a lot of simultaneous users. This is what people want in their public spaces. At the same time, the network can also support the kinds of municipal services you mention, with the Altai supporting mesh segments, providing them with backhaul.

Since you have elected a mesh system, that means that you as you note be mounting on light poles. That can be extremely costly. In NYC,the light poles are leased to you for $500 a month in prime areas by one of six franchise holders. If you happen to select a light pole that is somehow defective — in backhaul or in terms of power — you have to assume the cost of repairing that lightpole, which after permits, the bucket truck, closing off the street for a time, can run you $70,000.

An extreme example, but my point is that if you can at all avoid having to install in multiple sites, do so. With mesh, unfortunately, you need to. Point to Multipoint — central wireless hubs on key community buildings feeding small repeaters sitting in a window a mile or more away. You can even bill by each of four SSIDs on the repeater.

As for charging for the service, even something nominal, the watch out here would be that as soon as you start charging for something, you must assume customer service and support costs. If you just charge $5 a month, you may lose money.

If the only choice is mesh, then we need to pare back how we define municipal wireless. If its not, then a single infrastructure can support all IP based services in a community.