No. of Recommendations: 55
I just reviewed a video by some leaders in the telecommunications industry. It is not public.

My take away.

The changes over the next 10 years will be the biggest in the telecommunications industry ever.

The curve, the exponential curve when you graph anything to an exponent is in front of us, not behind us. The curve that is a definition of change will begin to really steepen in about 2016 or 2017, this is when the circuit switched network will cease to exist. There will be pieces and maybe even a nation wide system still in effect as late as 2020, but many pieces of the circuit switched network will be turned down by 2016, with more and more migration from circuit switched to packet switched happening faster and faster until about 2020 when only some odd bits of circuit switched equipment will still be in the network.

While this seems really important, and it is, it is even more important than one might think. There are two things that I have seen from the inside looking out, the first was in the video, the latter I will explain in a bit.

In the video there is a fundamental shift in how the equipment for the network will be built and installed. Today, the equipment is still effectively black boxes. When a piece of equipment is purchased, you get the hardware, software, support and training from the vendor. This locks you into a vendor. There are really only three vendors for telecom packet switched equipment at this time. (Cisco, Juniper, Alcatel) They have gross margins of 60 percent.

The telecommunications industry has no intention of letting that continue. There is a path laid out to separate the hardware from the software and push the hardware into a commodity situation and take control of the software. The implications are very large, more than just bad news for vendors gross profits. It changes the telecommunications structure in ways similar to the idea of DOS being open to different hardware in 1980. Not only can the cost of the network be driven down, the creativity of many many people can be plugged into the network to create applications and services we have not even bothered to dream about yet. This is very big, and very fundamental.

The other thing is this.

In the past, the telecommunications system has had a hub and spoke structure with the Central Office, or Switch in the center. Even though the switch is doomed, most people in telecommunications assumed that there would still be geographically located hubs, even if they were not traditional switches. The reason for this is; in a circuit switched environment, bandwidth is priced mile/speed. In other words, for a circuit path you select the speed and multiply it by the mileage to get the tariff. (This is very simplified.) In packet switched, mileage doesn't matter. In the old days, when you made a long distance call, the rate depended on how for away you where calling and how long you talked. Today when you buy internet service, no one asks if you are going to be down loading your videos from China or Russia (Except the NSA of course) they don't care. They only want to know how big they are, and how fast you want them delivered.

I have been indirectly involved in a project to migrate some of our outlying communications sites from circuit switched paths to packet switched paths. What I realized, except for round trip delay, there was no constraint on distance.

For example, I have a legacy switch in North Dakota. It will be retired in a few years. Most likely the office will be closed. However, today we have to keep the service running. But, if our outlying sites were moved from circuit switched paths to packet switched paths, we could divide up the load on the existing switch and parcel it out among idle resources all over the county. Not only that, if we did the same thing for our new equipment, we could interleave the support sites with the remote sites so that the complete loss of a support site, (think massive earth quake, nuclear strike, invasion from Mars) half or more of the remote sites would continue to operate, maybe all of them as the remote sites themselves are gaining intelligence.

Finally, with the distance constraint loosened if not removed, the locations of support sites, (Think Server Farm replacing Central Office) can be chosen based on low electrical costs, and low cooling costs. In other words, Canada and/or Iceland. (I do not know who would have low electrical costs and low cooling costs in the Southern Hemisphere)

One last thing, with packet switched data, we can more easily spread the intelligence of the network across the network, and the network can self heal. So, today we harden our sites as the sites get closer to the core. More power backups, better buildings, tighter security. If the intelligence is spread across the network, hard points are not required. Rather the network units would be allowed to fail. A fail soft situation. In this way the network becomes truly Borg like and would cost less also.

Qazulight (You will be assimilated, hummm? Donuts!)
Print the post  


What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Community Home
Speak Your Mind, Start Your Blog, Rate Your Stocks

Community Team Fools - who are those TMF's?
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.