Tom DeReggi wrote:

No I am not. I am asking you to tell me what you know, so I don't have to waste time replicating your research. Thats the purpose of this list, to exchange knowledge and data. Not just making claims, but disclosing why.

I recall in an earlier email you making claims without disclosing why. Are we going to debate the debate or mesh?

I never suggest that. And Neither am I.
But no, I do not own a city wide MESH network. I decided against MESH, which is why my opinion is biased against MESH. I can count the total ISPs on one hand that have completed that task as of today. Maybe two Muni's that had been legally allowed to proceed doing a large public network.
So my experience level does not lessen my point of view.

Why does the mesh have to be city wide? We haven't nor will we do a city wide mesh. We believe mesh is only appropriate in certain pockets of geography that allow the economics of mesh to exceed that of P2MP.

Of course you do. Just like most WISPs do. I can't count how many single unit Mesh system we had sent to evaluate. Its likely you did the same.

We don't do single unit mesh evaluations. What's the point? You can't really test a mesh with just a single unit.

As a router operating system manufacturer with protocol level coding experience, (but never went to market), we also have a great deal of talent on staff, to investigate the trade offs of various technologies. But you have to understand, that sooner or later a WISP needs to put the science projects aside, and start making sales, and they don't always have time to keep up to date on every latest and greatest daily enhancements to a technology, when they've decided on a different path to follow. I really don't have time to evaluate every manufacturers' Mesh product on a weekly basis to prove right or wrong their latest theories in their field.

I am not referring to some science project. We have active mesh deployments existing in the field now. Additionally, we are starting work next week on two more. One is a Tropos-based mesh, while the other is going to be built using some pre-release gear. The later is a mixed use development spanning hundreds of acres. That is the kind of scale we use to test mesh technology.

It doesn't appear you have done nearly the research we have
 and it doesn't appear you have any significant mesh deployments.

This is NOT a competition to prove who is the smarter technician and network designer. I do not claim to be Grand Master MESH. But I am more than qualified to carry on intelligent debate on the pros and cons of various routing and wireless technologies.

I'm not competing; just waiting for the intelligent debate to begin. You have made statements against mesh, but haven't made a single technical argument that was backed up by facts, research, or field experience.

I don't disagree with that definition. And technically in the dictionary, if it had to be defined, that could be it. But I feel MESH is more of a mindset than a definition. In practicality and real world, that MESH definition is two broad to cover all the many ways of implementing MESH. That definition does not define why someone would benefit from usingthe technology. I look at MESH as a concept of how to better gain coverage to a large number of people and/or geographical area, when Line-of-sight to a central or common sources are frequently obstructed, which typically requires more radios, installed closer togeather, and a method to manage their relationships, apposed to defining the way nodes communicate.

You are overloading the term mesh then. You can't take an accepted industry term and twist it to mean something else. Mesh is a network architecture that is more heavily used in wireline networks than wireless networks yet you want to suggest mesh has something to do with coverage. Each node in a mesh or each basestation in a P2MP system has an associated coverage area. Given a mesh node operating in the same spectrum and power output as a P2MP system you would expect a similar coverage because the coverage provided by the radios in question is a function of spectrum and power not network architecture.

Some people believed in John Henry, some believed in the Steam engine. I agree that the ultimately a computer (or technology) has more potential to be better equipped to make those decissions. However, today is not that day yet, and I have more confidence in the engineer (human). Not because the computer isn't capable, but because the engineer has not yet been proven capable to program the computer to be more capable.

Again I disagree. There are a number of large networks where a piece of software is used to configure network devices instead of humans configuring the devices directly. The same decisions are made by humans in either case, but in the former case the software can detect configuration issues before they are applied. Humans alone can't do that without a business process and additional humans to approve the changes. I'd much rather have a piece of software perform that task then to have more people and extra business processes. Software is cheap compared to human labor.

Wether you recognize it or not, technology has no value if not applied to a business case to solve. Technology's applications are well relivent to technolgy discussions. I'd argue that one of the big mistakes of technical people is they get trapped inside the technology, and design without adequately understanding the applications and ultimate goal of using the technology. For example, the task is not to reduce packet loss, its to be able to serve consumers more reliably. There is a big difference between the two. One approach is narrow and one is broad. What often happens, is technical people make these beautiful products from a technical point of view, but they are worthless because they don't solve the problems that need to be solved for its applications, which were the reasons for originally developing the technology. Just my 2 cents.

While I would agree with the above it seems irrelevant to the thread. Technology decisions without without the associate business are just as worthless as business decisions made without adequate understandings of the technology involved. Let us have a technical debate about mesh separate from a debate about its place in the market.

Most people appreciate that each technology has its place and we are often quick to point out when a technology doesn't fit, but how quick are we to look at the same technology and ask the more important question; where could it fit in my business. Our business makes use of a variety of networking technologies, mediums, architectures, and topologies. We don't have this level of complexity because it is fun; we have it because it gives a best-of-breed network that allows us to disrupt the market in a way that our competition isn't able to respond to.


WISPA Wireless List:



Reply via email to