It was only a couple of days back we heard that there were some problems with the coverage in St. Cloud, Florida. And now comes word that Google’s Mountain View Network might need more access points in order to get decent coverage. And that might mean delays from the proposed launch date of June 2006. (Google is being optimistic that it will meet the self-imposed deadline.)
Google’s begun testing the network and, in so doing, has discovered it might need to add more Wi-Fi transmitters than originally thought to deliver the coverage and service quality it promised, according to Ellis Burns, the city’s economic development manager.
Now the other day I was talking to Dewayne Hendricks who is helping build a county-wide wifi network, (not a town or city wide network) in New Mexico using off the shelf components and white label gear. He is having no problems whatsoever, and well, most of these networks are seeing re-configuration, to put it politely.
Maybe its just me, but maybe it has something to do with network planning and the gear which is resulting in dead spots, and spotty coverage. Esme thinks so too, and has a cheat sheet of what to do and what not to do when it comes to MuniFi.
I just read where Google has backed out of the Valley wifi bid. Wait until Wimax?
Of course it’s an issue of network planning. I’ve worked on many a wireless project (WiFi, WiMax, Cellular) and detailed engineering is often an afterthought that happens after a bunch of people who don’t really know the technology have done a bunch of back of the napkin work and come up with a bill of materials and budget. The other thing that happens if you are working inside an existing network operator is that the budget people have no understanding of technology at all and want everything to be cookie cutter to make their lives simpler. You can explain until you are blue in the face and such people will never accept why your real world situation deviates from their expectations and if by some miracle they do understand they still don’t give a shit.
I’d also look at the geography of the areas you’re talking about. I’ve only been in Mountain View once but it looked pretty hilly to me. As for St Cloud, that shouldn’t be an issue given FL’s billiard table flatness…
Tree and wet foliage attenuation and the differences in residental construction material [wet and dry] and screens are almost never accounted for in the fade margin…something cellular skipped in the past…did they care how well the phone works inside, you can move the phone around, lap tops are harder to find the window.
It is the single radio nodes that is the problem folks. When one needs to regenerate the bandwidth after every 3 hops one has a problem.
What about the clutter of all of the other WiFi networks out there? I have problems with the 10 other G networks, who knows how many cordless phones, and what not in range of my living room…had to switch to A.
At http://www.fon.com we already signed up 35,000 foneros, people who are happy to share excess bandwidth at home in exchange for free roaming around the world and in the process build a global wireless cooperative. We estimate that in 120 days these 35K foneros will be up and running. We may lose some along the way, but we are getting many more each day. I know these numbers seem small but 3 months ago before Google and eBay invested with us we were only 3000. And think about T Mobile, they have invested years of labor and over one hundred million dollars and only have around 20K hotspots according to their web site. So with all respect to my partners at Google and the muni wifi folks I think that Fon´s citizens based approach is a better way to reach massive coverage than going light pole after light pole. Replacing individual wifi routers for social wifi routers is better than adding more wifi on a crowded spectrum.
Could it be that Tropos over promised and under-delivered?
Single radio = big problems … It’s like a router with one interface (what we used to call a one arm router)…