Not so fast, Henry, because Microsoft’s entry sets up a fascinating natural experiment in business models, as varying approaches compete with each other, and pre-judging the outcome is rash.

A mobile smartphone system has four parts: hardware; operating system; carrier; add-on applications. These can be furnished as a complete or partial package — Apple, for example, provides the hardware and the operating system, maintains tight control over the applications store, and integrates its offerings with the carrier, AT&T. Or they can be furnished as stand-alone items, as Google furnishes only the Android OS. Or something in between is possible, as handset makers load Android, but are then integrated with carriers which provide subsidies while they duel over who will control the applications.

Competitive advantage can be sought from any element of the system. Apple profits greatly from the elegance of its hardware and the devotion of its fans to its OS. RIM (Blackberry) has long been a favorite of corporate America because of its security and integration with corporate systems. Android gets an edge by being both free and open, two different attributes. Free appeals to the handset makers, who like the idea of paying nothing for an OS and at the same time having no responsibility for developing or maintaining it, and open appeals to everyone (makers; carriers; programmers) who might want to tweak it for their own particular purposes.

But competitive advantage is a tricky thing, and the characteristic for which the tech world most loves Android, which is that it creates a great playground for geeks, and encourages the development of thousands of applications, is probably the least important concern for the vast mob of consumers, who can barely program a video recording machine. For the mass market, the most important considerations are ease of use and transparency, lack of hassle in use, reliability, security, enough of the right apps. Above all, the time of the user is the most important resource.

Windows won in the PC not because anyone loves Microsoft or because it lacks many annoying idiosyncracies (for example, its help functions were written by brains that do not work like mine) but because Microsoft invested heavily to ensure that anything that is plugged in, works. Anyone who remembers the early days of the personal computer, when adding a printer was a day-long project, understands how important this is. And how quickly lost time wipes out any money savings – for any sensible person, the $15 that Microsoft plans to charge for Windows 7 will be a bargain if it saves him a single hour over the life of the phone, and this is without consideration of any possible ancillary costs, such as security breakdowns.

So what approach to a smartphone system will win in the market? The only thing we know for sure is TANSTAAFL – there ain’t no such thing as a free lunch. Every element of the system must be paid for in some fashion, at the outset and most probably by continuing investment in maintenance. Microsoft, and all other software companies know this, as is shown by the stream of updates, many security-related, that appear regularly on my PC. As Apple has grown more popular, it too has become more of a target for malicious hackers, and it too must be devoting more resources to countermeasures.

So, who is going to maintain Android? Will Google? But how, since a chief characteristic of Android is openness, which means people can tweak it in private ways? And how does this fit with Google’s business model, which is to collect money for advertising that is connected to Android, and to use its software to collect information that improves the effectiveness of that advertising; there is no logical or necessary relationship between this model and the costs of maintaining an OS on a continuing basis.

Will each handset maker that installs Android undertake the responsibility for maintenance, and is this economical? It appears quite duplicative on the surface; indeed, it would not be long before the handset makers got the idea that they should all contribute to a third party that would maintain Android for all of them, but oops! Wasn’t that that $15 per instrument savings that just went out the window (pun intended)?

Blodgett followed up his original post with one quoting an unnamed “person familiar with Microsoft’s mobile strategy” who contended that “Android comes with all sort of hidden costs that drive up the per-unit expenses such that Microsoft’s $15 per unit actually seemed like a better deal.” Said person notes the plug and play point, for example:

Android’s laissez faire hardware landscape is a fragmented mess for device drivers. (For background, just like PCs, mobile devices need drivers for their various components—screen, GPS, WiFi, Bluetooth, 3G radio, accelerometer, etc.)  Android OEMs have to put engineering resources into developing these drivers to get their devices working.  The Windows Phone 7 “chassis strategy” allows devices to be created faster, saving significant engineering cost.  It’s essentially plug and play, with device drivers authored by Microsoft.

Blodgett asks for continuing input on the issue, and Amen to that. In theory, my guess is that the the $15 per instrument charge for the OS will win out, but maybe not, or maybe Google or the handset makers will find that continuing services are not necessary or that the economies of scale are such that they can be provided at a cheaper price. The scale is certainly there, as hundreds of thousands of Android phones are being activated every week.

So let the games begin, and may the best business model prevail!

Wall of cell phones from Jason Dunn.

Let’s apply some high ping jitter to any gamer or Vonage VoIP user and see if they “obsess” or not.  Anyone who lives with a gamer or VoIP user knows never to mess with a man or woman’s ping.  We gamers or VoIP users will kick everyone in the house off YouTube or Netflix or BitTorrent whenever we need to use our Lingo VoIP service or do online gaming.  I personally will go as far as disabling the Wi-Fi if I need to.  It’s quite common to hear your VoIP or gamer peer on the other end say: hold on a minute as I kick the BitTorrent and Netflix users off the network because they cause so much jitter.  Is that obsessive?  You better believe it but that’s what happens when engineers are told by religious zealot non-engineers that they can’t do their jobs even though the only way to manage downstream jitter is on the network provider’s equipment.

The bigger problem with Pegoraro’s column is its insistence that wireless networks are the same as wired networks when they aren’t.  Pegoraro claims that the larger capacities of 4G eliminates the congestion issues, but that isn’t even remotely close to reality.  No matter how much capacity there is in a 4G cell, that cell has to support hundreds and possibly thousands of users on a network that has less total capacity than the guaranteed capacity of a single piece of fiber or short-haul copper on a Verizon FiOS or AT&T U-verse broadband connection.  No matter how much Rob Pegoraro or anyone else wishes it to be true, we will never turn mobile networks covering more than a few rooms into a high quality Video on Demand network because it’s technically impossible.

The other major technical difference is that wired broadband connections offer a level of immunity between the users in the sense that jitter generated by a single user doesn’t affect any of the other users in the same broadband network.  On a wireless network, a single BitTorrent user poisons the entire cell with massive jitter and makes the network unbearable for all the users on the same cell and even the entire cell tower because the backhaul is jammed up with jitter.  A wired broadband subscriber can only cause jitter for his or her own broadband connection which makes it painful for the people living in the same house but not the entire neighborhood.

The other differences are economic.  When was the last time you’ve seen wired broadband service sold on a per device basis rather than a per-home basis?  You can buy single-device wireless service for very specific devices like the Apple iPhone 4 or iPad at half the monthly cost of a any five device service plan like the MiFi portable hotspot service plans.  Would Pegoraro insist that these existing single-device service plans be outlawed?  That’s what Net Neutrality would do as it doesn’t allow a network provider to place contractual limits on the device.  When was the last time a content provider like Amazon paid for your wired broadband service but dictated what webpage, application, or device you could use?  It never happened on wired networks but it is happening on wireless networks and the Kindle.

When was the last time someone sold a device that came with very cheap but very limited wired broadband service?  It hasn’t happened, but it is happening in the wireless space with devices like the Peek Simply Email which only supports email but at a very low monthly cost.  Furthermore, would this specialized type of device and connectivity be a bad thing for the people who don’t own a computer or broadband today?  What if they could get an affordable and simple iPad-like device that had a very cheap but limited service offering such as email and non-video web?  Wouldn’t that allow computers and the Internet to reach populations that have never been reached before?  Should the market get to accept or reject these business models or should we have Soviet-style central planning for the future of the Internet?

As many are aware there is a coming IPv4 address shortage.  T-Mobile is making strides to move into IPv6 infrastructure in the coming months to sidestep any issues.  Recently T-Mobile posted to their Google groups page that they intend to begin a IPv6 beta program called T-Mobile USA IPv6 beta service.

T-Mobile believes that this head start will allow them to start offering IPv6 enabled devices to their customers, and the beta period will allow testing to verify that these IPv6 devices will be as “feature rich” as their current line up of IPv4 devices.  Additionally they believe this beta will provide an opportunity to move past the limitations of Network Address Translation (NAT) limitations to the eventually standardized Softwires Network Address Translation (SNAT) that will allow IPv4 and IPv6 networks to communicate with each other.

Participants for T-Mobiles IPv6 beta program were asked to meet 5 criteria:

1. Be a T-Mobile USA subscriber with an unlimited data plan
2. Have T-Mobile coverage, not roaming or WiFi
3. Have a Nokia 5230 Nuron or the Nokia E73 Mode phone.  The N900 also works, but it’s IPv6 support is much less mature.
4. Be willing to help T-Mobile improve the service, forgive us as we grow and refine the service, and accept that this beta service is not supported within any T-Mobile support channel, including Customer Care or any T-Mobile store or reseller.  This Google groups forum is the only channel for IPv6 support during the beta friendly user trial.
5. Accept that the service is still evolving and that many services like Visual Voice Mail, MyAccount, MMS (picture messages), and several other services do not yet work. Web and Email both work well, but many other data services are still coming online with IPv6.

Finally the group points to several YouTube videos showing IPv6 in action, which can be found here.

For more information on the T-Mobile beta program and the full article you can head here.

The concern for AT&T customers across the board here is whether or not this will begin to cause congestion on the network.  As the decision by AT&T to not allow FaceTime over 3G from the get go seems to suggest that the network was incapable of handling several million people potentially using FaceTime over 3G at the same time.  Bare in mind that Steve Jobs couldn’t place a single FaceTime call over WiFi at June’s WWDC until all WiFi users shut down their computers.

If the percentage of jail broken phones on AT&T is low, then it may not be a widespread problem on the networks, but could still cause some local turmoil for other customers using the same tower as a 3G FaceTime user.  It will be interesting to see how Apple and AT&T respond to this.

Bennett comments to the FCC on disclosure of network performance for wireless networks.  He believes that customers should receive clear information about wireless network performance when signing a contract with a provider.  He recommends a,

“…system of operator-supplied coverage maps, primarily verified by consumers through extensive refund rights when user experience does not conform to the claims implicit in the coverage map.”

Bennett covers 6 areas of interest in his comments:

1. The Internet is Intrinsically Unpredicatable
2. Mobile Networks are Intrinsically Unpredictable
3. There is a Risk in Over-Generalizing Survey Data
4. Applications, Users, and Handsets Strongly Influence Performance
5. Taking Meaningful Measurements
6. Advanced Measurements

In conclusion, Bennett believes that the current drive for data can often lead to a collection of data that only supports a specific policy goal and ultimately becomes “data driven by policy”.  He feels that the bottom line is customer satisfaction, but urges restraint on the FCC with its efforts in data-gathering, believing that in the interest of privacy, a reliance on operator-supplied coverage maps is the best option in the near future.  He states that these maps should be updated frequently giving consumers the most current information so that they can make the best purchase decisions for their situation.

You can find the comments in full here.

Levy says,

“This decision is a first step toward opening up wireless networks and releasing the stranglehold that companies have on consumers who, until now, were legally restricted from doing whatever they want to their phones.

When you buy an iPhone, Android or another smartphone, that phone should be considered yours to use as you see fit — just as it is with a desktop computer.”

I mean these statements are simply asinine.  No one has ever been realistically restricted from doing whatever they wanted to do with their iPhone or any other contract bound device.  What they have been restricted from doing is doing whatever they wanted with their phone and maintaining the warranty in their contract.

Levy and FreePress want everything both ways in everything they report on and this is no different.  The reason Apple doesn’t want you jailbreaking their phone is not because they are scared of you getting freedom, it’s because they know how things work inside the walled garden platform that they have created.  Sometimes that means certain apps may not get approved, or you may not be able to take advantage of the phone in a certain way.  But that garden protects the consumer because if something breaks within that garden you are protected by a warranty.  If you leave that garden Apple and AT&T cannot be responsible for the problems because they don’t control what’s outside of the garden.

But FreePress wants you to get all the potential benefits of jailbreaking without the consequences.  Furthermore, as I’ve already discussed before, if you want the phone without a contract, just buy the non-contract device.  We as consumers can’t complain when we don’t want to pony up the dough for a non-contract device and AT&T is subsidizing the cost of the phone in exchange for putting us in a contract.

And not to continue to burst Levy’s bubble, but we got to see a phone available for multiple carriers without contract available for purchase from Google.  It was called the Nexus One and it was a gigantic bust.

Karl Bode at DSLReports thinks that Siegel is insulting people’s intelligence when he calls this a perception problem, but Bode misunderstood what Siegel said.  Bode sees this as Siegel saying that there is no problem when Siegel is saying that there are problems but they’re not as bad as how the media perceives and reports this, and the Yankee study seems to be verifying what Siegel is saying.  Bode even admits that AT&T is working on their problems ($36 billion capex on wireless in 2009 and 2010), and he even admits that the problem seems to be concentrated in big cities like San Francisco and New York (not to mention the fact that Apple’s iPhone baseband shares a lot of the blame for dropped calls), but his anger is disproportionately aimed at AT&T.

The big issue with big cities like San Francisco and New York is that they make it darn near impossible to get a permit to install a new cell tower or even to expand an existing tower.  City officials in San Francisco even like to brag that they haven’t authorized a new tower in some places for nearly a decade, and this stems from the irrational fear of cell towers.  People want better cell tower reception but they don’t want a cell tower near them when the reality is that a near by cell tower allows you to be exposed to less radio energy.

People have a hard time believing this last statement because they lack an understanding of the physics.  The fact that the cell phone is about 10,000 to 100,000 times stronger than a cell tower and that the cell phone can emits many times less energy when it’s near a tower in a strong coverage area.  So the closer a person resides near a cell tower, the less radio energy they are exposed to.  And as for concerns of ascetics, cell towers are being disguised as palm trees so there shouldn’t be a concern for property value.

While the marriage may indeed be loveless, Apple is in no short supply of receiving the love now that they are pulling more profits than AT&T ($3.38 billion for Apple and $3 billion for AT&T).  While AT&T has much higher revenues, AT&T has to spend nearly 16 times more on capital expenditures (17.3 billion in 2009) versus Apple’s 1.1 billion in 2009 doing the upgrades to keep the iPhone afloat.  Furthermore, a significant portion of the monthly fee that consumers pay to AT&T actually goes to Apple which makes AT&T look like the bad guy.  AT&T has to do the dirty work of collecting the money and being accused of being “greedy” when Apple is the one getting the bigger chunk of the profits.

UPDATE 3:00 PM – Note that Vogelstein’s charts show a few billion less CAPEX spending for AT&T but I don’t know how he’s counting it.  I included links to AT&T financial statements as the source of my figures.  AT&T is actually increasing their CAPEX spending from $17B in 2009 to $19B in 2010 but MG Siegler at TechCrunch things it’s hard to feel sorry for AT&T pointing out that AT&T isn’t spending as much as they did in 2008 based on Vogelstein’s numbers.  I’m not sure it’s fair to expect a company to meet record CAPEX spending levels every year and Siegler admits that no amount of investment would be enough anyways.  This is one of AT&T’s biggest problem in that no matter how much money they spend to improve the network, they’re not going to get much if any credit for it.

Lastly, I find it interesting that Siegler still blames AT&T for the dropped calls even though he’s recommending Vogelstein’s article that explains that the dropped call problem is much to do with Apple’s bad baseband design.  But it’s clear from reading the rest of Siegler’s piece is that he carries a very irrational hatred of network operators and love of Apple and Google because he can invent the facts as he deems fit.  He credits Google for “opening the spectrum” when Google spends less then 1/20th the CAPEX of the network operators.

Another interesting nugget from Vogelstein’s piece is that much of the problem with iPhone disconnects were actually due to the iPhone’s baseband software yet the press and the public blamed them on AT&T network problems.  While this might have been hard for most people to believe a few months ago, they’ve recently been made aware of how important a role the phone plays when it comes to connection stability by the whole iPhone 4 “antennagate” saga which has apparently been turned into a feature animation (video is self explanatory even if you don’t understand Chinese).  People learned that Apple is very reluctant to accept any blame and they would go as far as blaming their own customers even when the problem is obviously Apple’s design.

One other interesting point was how difficult it would be to create a CDMA version of the iPhone as opposed to the current GSM version.  Vogelstein wrote: “The new chips were a different size, which would require Apple basically to rebuild the iPhone from scratch.”  I knew it wouldn’t be trivial but never imagined that it would be that bad.  Vogelstein also pointed out that Apple wasn’t sure if Verizon would be any better to work with.  I’m guessing that Apple isn’t sure how good the Verizon network would be once they’re forced to deal with iPhone loads.

Lastly, Vogelstein noted AT&T’s elimination of the unlimited $30 data plan resulted in more accusations of “greed” being thrown at AT&T.  However, I noted recently that AT&T stands to potentially lose up to half its revenue from 70% of its user base because those customers can change to the $15 plan without having to change their web usage habits.  The more rational explanation is that the smaller caps with lower prices are being implemented to slow the growth of data usage to a manageable pace that the $19 billion 2010 investment in infrastructure could keep up with.

The fundamental support for mesh networks stems from simple anti-establishment attitudes.  Mesh proponents believe that you could build a large or even nationwide network without any wires or cell tower base stations with just end-point devices and you could completely bypass and obsolete the phone/Internet company.  The mesh cult even comes with the obligatory conspiracy theories that explain why mesh networks aren’t widely deployed due to corporate suppression, but that completely ignore the laws of creative destruction.

Why mesh doesn’t scale

The fundamental problem with mesh networks is that it simply doesn’t scale.  If you wanted to build a network based solely on end-point devices like laptops or Smartphones, it would take at least 64 active and optimally placed end-points just to provide the equivalent coverage of a base station with 8 times the range as shown in figure 1 below.  Real world conditions would likely require many times more end-points and it would be even more difficult to substitute for a long range cellular base station.

Figure 1 – Mesh scales poorly compared to hub-spoke architecture

One might ask how is it fair to give the base station 8 times more range, but that’s the fundamental advantage of a base station not possible with end points.  Base stations usually pick out a fixed and strategically advantageous locations, come with an antennas that’s are many times longer, and have transmitters many times more powerful.  Cell tower base stations are far more powerful than 8 times the range.  End point devices by definition have to move around to serve their human owner and be small enough to fit in their human owner’s pocket which severely limits the size of the antenna and battery.  All these factors make end points horrible makeshift base stations.

History lesson of the OLPC XO:

The OLPC XO computer was a huge failure because the design was tragically weighed down by unrealistic and unnecessary design goals because of its Nicolas Negroponte’s insistence on mesh networks and other technologies.  Negroponte believed that schools and even teachers (which he characterizes as drunks) were unnecessary and that the solution was to “leverage the children themselves” for learning.  Based on this philosophy, his team designed the OLPC XO to function in daylight (spun off to Pixel Qi) so that it didn’t need classrooms, originally designed the XO with a hand crank generator assuming that there would be no electrical infrastructure, and implemented mesh networking to function in the absence of network infrastructure.

The latter two design goals were complete flops.  The outdoor screen seems to have spun off into a viable commercial technology though it remains to be seen how successful it will be.  The generator idea was too difficult to deploy in a sufficiently reliable form factor and the mesh feature simply inflated the cost of the OLPC since there were few off-the-shelf mesh capable radio subsystems and it reduced the XO’s battery life.  The XO had to constantly remain on to act as a mesh relay node which drained the batteries.  A simple $100 Wi-Fi base station would have been a superior solution with a much larger and more reliable coverage area while reducing the cost of every XO laptop.

The true believers of mesh networking even touted Africa’s ability to leap ahead and connect the entire continent of Africa through mesh networks.  The BBC even called this “the Great Leap Forward” which was truly ironic and fitting considering how eerily it matched the Chinese Great Leap Forward of the late 1950s.  They didn’t need any established methods of agriculture or industrialism for they would simply invent their own way and leap beyond the west.  Being born at the end of the Great Leap Forward, I am keenly aware of the tragedy it caused.

Why mesh networks are fundamentally slower

In the real world, mesh networks are overwhelmed by overhead traffic to the point that they melt down with any kind of real user load and the Mongolians learned this the hard way during deployment.  But even without overhead traffic, mesh networks are fundamentally flawed because mesh network are fundamentally many times slower than direct hub-spoke solutions that employ base stations.

We can imagine a cargo fleet of 30 airplanes transporting goods from San Francisco to New York, but having to stop in Las Vegas and Denver Airports. We would have to allocate 10 planes on each leg of the flight rather than all 30 airplanes for one continuous trip from coast to coast. Even ignoring layover delays and the extra time it takes to take off and land, we’re still going to lose 2/3rds of our direct shipping throughput.

Figure 2 below illustrates how each node cuts down capacity because the relay node has to either split the frequency or split the time it can use to forward traffic.  A relay node that receives and transmits at the same time would be forced to use half the radio frequency for receiving and the other half for transmitting.  If the relay node alternates between sending and receiving, it could use all of the radio frequency but it only gets half or 1/3 of the time to transmit.  The bottom line is that a lot of capacity is lost whenever we deploy relay nodes.  The relay design works great when we can’t go direct point-to-point due to structural obstacles like hills, trees, or even just the curvature of the earth, but to deliberately forgo the direct route by eliminating base station infrastructure is just silly.

Figure 2 – Mesh networks lose capacity with each relay node

Mesh might work for military applications

Many proponents of mesh networking often point out how the military is very interested in mesh technology, but they fail to see that military requirements are largely irrelevant for civilian applications.  The military needs end points that can survive attacks on base station infrastructure but this isn’t a concern for civilian networks.  A soldier could carry 40 lbs of communications gear, but civilians balk at anything more than 5 ounces.