...then look at open-mesh.com. $50 a pop, but no distributor in the UK. How about a review or group test, el Reg?

An interesting AI.Gor.IThm.ic problem - how to find your (packets') way OUT of a mesh network. Easy to see that packets looping and sniffing about for an exit strategy could eat bandwidth - a system which DDoS's itself in no time at all. But what could a deterministic mesh bring to the table in the current investment infrastructure regime? Little, one suspects.

So let's look thru the other end of the telescope. What could a mesh with non-deterministic routing achieve? Push and forget. Now we're talking. We're baying hue and cry. Talking the Twilight Barking (not yet the Twilight of Barking). Real wi-fi for real communities that doesn't need ISPs, Big Brother, or big bovver. Nor BT in bad form.

Watch the hand. Watch the hand of the politician.

Like here: http://www.emersonprocess.com/smartwireless/index.asp

Field instruments communicating with the DCS in non-critical control loops don't need to transmit a lot of data, but may find themselves in hard to reach (or even on moving) locations which would ordinarily be difficult to wire up. Even if mesh networks find no friends in the domestic market, they're proving quite interesting to Industry.

"No longer just for socialist dreamers" - but for Big Oil as well ;-)

>> But what could a deterministic mesh bring to the table in the current investment infrastructure regime? Little, one suspects.

Competition

Looks like amanfromMars has a split personality, only slightly more comprehensible. I said slightly.

So? What good has competition ever done the broadband market overall?

Yes residential broadband is now cheaper than it was five years ago, but the benefit that competition (as regulated by Ofcom) brought to the UK market is BTwholesale prices so high that they're almost insane, with the direct consequence that "fair usage" or even outright quotas or PAYG are the order of the day for any BTwholesale based ISP. And there aren't any other "wholesale" ISPs to speak of - there was Tiscali, and moving swiftly on, there was going to be Easynet's LLUstream, but Ofcon kindly let Murdoch buy them and so now there's no competition, no choice of connectivity provider, for ISPs who don't own their own national network (which is most of them).

In the wireless market, which hopeful wireless competitor to BT hasn't disappeared without trace? Where are Liberty/Tele2/Pipex... (even Pipex/Intel's WiMax is invisible)? Where is Sunshine Broadband? Where is the ISPA-award-winning wireless ISP, West Dorset Internet? Where, even, are the winners of Ofcon's UK fixed wireless broadband licence auctions, PCCW/Now/whatever they're called this week? Competition (the auction) worked well there, didn't it?

There's plenty of competition in the urban areas, where multiple DSL providers are competing with each other to see who can charge least, and consequently many exchanges actually don't have room for much more LLU kit. Several sets of kit, several sets of parallel infrastucture, all has to be paid for by someone, and it won't be the kit suppliers or the infrastructure providers.

And if you're not in an LLU area, you'll soon be finding out the true meaning of BTwholesale regional pricing under the much-over-hyped 21CN [1].

So, maybe, if the Wisps don't mind competing in areas where the LLU ISPs have decided they can't make any money, maybe there'll be another chance for competition to show its benefits.

Otherwise, it's back to BTw, and have to hope that Ofcon wake up and actually start regulating.

[1] http://telebusillis.blogspot.com/2007/08/21cn-bands-and-regional-pricing.html

This really isn't about broadband though is it? This is about mesh networks and their multiple applications. Yes, they can be used to provide broadband, quite cheaply and efficiently as it happens (think 1 broadband connection used by 1 person 25% of the time using 15% of its available throughput [actually a lot less generally] compared to 25 people using 5 connections (the costs of nodes are minimal about $50 or £30).

The benefits of a community wireless mesh should be obvious (OK, OK, broadband 'power-users' like most of El Reg's readership probably still require a dedicated connection but your average joe reading emails/browsing pRoN/reading BBC news has all he needs and for a fraction of the cost). It already works successfully in many parts of the country - a good example is SussexNetShare (SnS), it actually has meshes all across the country and is a major contributor to the open-mesh.org project.