T-Mobile's G1

Net neutrality is more than just one of the current industry buzzwords, and according to T-Mobile, Android is smack dab in the middle of it all.  After reading a great article over at Fierce Wireless, I got to thinking about just how the outcome of the whole net neutrality issue can and will affect Android as a platform.  Hit the break to see my thoughts on the whole issue -- as well as an interesting tale about how some badly coded app nearly took down T-Mo's network. [FierceWireless]

Many of you know my feelings about openness and free open source software, so you can probably guess my position on net neutrality.  For the rest, I'm a firm believer that people who stand to profit from something should have no say in it's use or development.  Cellular carriers should not be allowed to police network traffic by throttling bandwidth under any circumstances, and if having things wide open isn't working, let someone elected (or appointed by those we elected) determine the policy.  And yes, I'm used to the hippie communist pokes I get for it, but those are my beliefs. 

On the other hand, you have the carrier point of view (we're sticking to cellular talk here, I'll fight with Comcast and Verizon FIOS in another venue).  I've always just written off their side of the argument as a by-product of corporate greed (cue those communist hippie jokes right about now).  But an article by Mike over at Fierce Wireless really caught my attention.  You should take the time to read it, so be sure to follow the source link when you're done here.  In T-Mobile's filing with the FCC in January of this year, they have a very interesting story about an improperly coded instant messaging application that damn near crashed their network in "certain densely populated network nodes".  Here's the quote, direct from the FCC filing :

"...T-Mobile network service was temporarily degraded recently when an independent application developer released an Android-based instant messaging application that was designed to refresh its network connection with substantial frequency. The frequent refresh feature did not create problems during the testing the developer did via the WiFi to wireline broadband environment, but in the wireless environment, it caused severe overload in certain densely populated network nodes, because it massively increased signaling—especially once it became more popular and more T-Mobile users began downloading it to their smartphones. One study showed that network utilization of one device increased by 1,200% from this one application alone. These signaling problems not only caused network overload problems that affected all T-Mobile broadband users in the area; it also ended up forcing T-Mobile’s UMTS radio vendors to reevaluate the architecture of their Radio Network Controllers to address this never-before-seen signaling issue. Ultimately, this was solved in the short term by reaching out to the developer directly to work out a means of better coding the application."

There is no mention of which application (we all have our guesses) or which locations, but it's a pretty scary scenario if you're a wireless carrier -- one "improperly" coded application could bring your network to its knees if it gets popular enough.  This could happen to the big boys (Verizon, AT&T) just as easily, considering the sheer number of subscribers.  T-Mobile goes on to say what we've already figured out by enjoying our Android phones to the fullest:

"T-Mobile has first-hand experience with ballooning demand for wireless broadband access, and in our experience, demand increases whenever new devices and new capabilities emerge. Customers with devices that lack full web browsing capabilities consume minimal amounts of data and require lower throughput. But that has changed as T-Mobile has introduced new devices with capabilities to make use of T-Mobile’s 3G network for all sorts of web-based applications.  Eighty percent of T-Mobile’s myTouch users now browse the web at least once per day, and two-thirds do so several times per day. Thirty percent of T-Mobile’s data traffic today consists of bandwidth-intensive video streaming—most of which is done by Android users."

And even more telling, in the full version of the FCC filing, including comments, we read this:

"The issue is not just that more wireless users are online. To compound matters, the type of wireless broadband usage has changed over time, as well. As speeds and handsets improve, wireless consumers use their devices for longer periods of time and for more bandwidth-heavy applications. For example, as we have reported, G1 handset users consume over 300 megabytes per month—more than 50 times the data of the average T-Mobile customer. More than 40 percent of T-Mobile myTouch users access social-networking sites multiple times per day. Over thirty percent of T-Mobile data traffic already consists of video streaming—a majority of which is attributable to Android users."

I'm done with the quotes.  You should really read both the filing by Grant Castle, Director of National Planning & Performance Engineering at T-Mobile USA and the detailed filing at the FCC's Electronic Comment Filing System to get a handle on exactly why and how T-Mobile claims this affects their network, and how they feel they can fix it.

Back to the matter at hand -- T-Mobile has a point.  Our Android phones can do just about anything, and eat some serious network bandwidth while doing it.  Other smartphones can, too -- but I'll let Dieter, Kevin, Rene, and Mal Daniel handle that.  I don't want data caps, but their coming.  I didn't want to pay Sprint an extra $10.00 a month because their new Android phones can eat five times the bandwidth as last years models, but I did.  I certainly don't like paying the money to tether and get all I can eat, but I do.  I have brought myself to grips with all these, as there really isn't another solution that works for both sides, at least not one I can see.  And since I'm streaming a NetFlix movie on my home internet, and using Sprint's tethering package to download and install Gentoo Linux on a netbook while I write this, I'll pay the fees and not complain.

But apparently this isn't enough.  If T-Mobile is correct, and they will be unable to provide more important services like voice unless they have the control to "manage the network load and fairly serve the different needs of different users, devices and applications" (OK, one last quote), then we all will suffer.  My idea of letting the people willing to spend the most money and offer the best service come out the winner doesn't sound so practical anymore.  Letting T-Mobile (or Verizon, or AT&T, or even AlaskaTel) prioritize traffic to and from certain places (Bing! anyone?) doesn't sit well with me either.  And to top it all off, I really don't have faith that the FCC will make the right decision.  It's tough being a geek in the new millenium. One thing I do know -- if we can't give in to the carriers a little bit, they're not going to want to carry the next bandwidth-hogging Android wünderphone, or offer the full internet experience we've all come to enjoy. 

What I think is really important here, is that we all need to become involved in the whole net neutrality debate, because if we sit back and let someone else make the decisions we're not going to like the outcome. 


Reader comments

Android plays big in T-Mobile's net neutrality arguments


Just like battery technology is slowing smartphone users down, so is this bandwith capping.

Sprint went the right way... but AT&T and others... blah.

I have to say that the idea of data throttling sounds better to me than just having a cap, that if you go over your data package usage for the month you have to pay more money to add more data to your package for the remainder of your billing cycle. Having it so that your data speed is just throttled back a bit when you hit your package limit allows you to still be able to access data, just a little bit slower, and not have to fork out more money for the few days you probably have left in your billing cycle. Having no kind of caps or throttling would be ideal, but we all know that is not going to be around for too much longer on the carriers that still offer it.

Here is my issue, T-Mobile makes a point about an Android application reeking havoc on their network in certain areas (if they are not designed and tested properly), but we (the developer/super user Android community) have no way to test a device (think dev PHONE) if we are on at&t or Verizon or SPRINT.

For a few short months were able to purchase a Nexus Ones with North American 3G frequencies on board, but with Google back to selling AWS frequency dev phones only how are DEVs who are not with T-Mobile/WIND able to determine the effect of their apps on a UMTS/LTE/WiMAX network?

I do understand the needs of the mobile operators (although I do like it) and while I do not like spending dinero (does anybody really like shelling out cash?) I do so because I understand that resources are "somewhat" scarce (the tragedy of the commons anybody????) and I do not want to in the end really paying more for what I enjoy today paying only a little more.


If you don't like Sprint's mobile data pricing plan, then you shouldn't CHOOSE it. You should use your dollars and spending choices as an informed consumer to affect corporate policy instead of complaining about it and expecting the government to fix it. Government rarely solves anything, quit complaining and do your part to solve the issue as a consumer.

If more consumers voted with their pocketbook, then corporations would have to respond in order to continue making a profit.

Nobody likes having to pay more. That wasn't his point, though. It was the fact that the additional charge was an acceptable method for handling the increased traffic.

Not to sound like a Google/Verizon shill, but that was one of the benefits of their proposal; it allowed carriers to make those sorts of adjustments, but required complete transparency about what those decisions were. That way consumers and government agencies can see whether they are simply prioritizing voice (or emergency!) data, or whether they are favoring Bing, or choking websites with political messages they feel don't serve their interests.

I know it was portrayed as some sort of cloak-and-dagger attempt to screw people, but the Google/Verizon proposal was eminently reasonable as a baseline for how to deal with this issue IMO. Too bad it was vilified by both ends of the political spectrum before it got a proper hearing...

300MB is a lot of use? That's lightweight. I did that much when I had an enV Touch that I could just use for web browsing and email. If that's really 50 times what the average T-Mobile user uses, I hope those users aren't paying $30 a month for data. Or maybe I hope they are to subsidize those that actually use data.

Ok, so an application did something that t-mobile did not expect would be done and that caused problems. The problems were identified, the issue corrected, and (per t-mobile's quoted text above) the network is stronger for the experience.

Sounds like everything is working as it should.

This is a bunch of crap, the carriers are using the classic "the internet is a series of tubes" analogy to justify their greed and people are just eating it right up

Maybe I should move to Europe or Asia where consumers don't put up with and apologize for their corporate overlords' shenanigans

lol, so instead of worrying about powerful companys that easily bend when people stop buying their products(free market), you would trade that for a severly over reaching nanny government in the EU and most european governments on top of that... that DONT bend when people stop buying their products. The grass isnt always greener on the other side buddy.

The problem that your arguement forgets is that in the last 3 or so years the bandwith usage of phones has gone up exponentially faster than the tower build-outs. The explosion of the Apple-Android fight has changed the landscape faster than anyone could have predicted. There is only so much bandwidth we can use. Its a lot like a 8 lane super highway. It seems like enough road for everyone, but too many cars on it brings everything to a halt. Ever seen LA or Atlanta highways during rush hour?

Also, I have found that people that use the term "Corperate greed" really have no idea what they are talking about. Its a dead give away for someone that listens to liberal talking points without knowing their meanings. Corperations only exist to make money and they have people that they have to explain themselves to (the shareholders) or people start losing their jobs. Its not greed, its survival.

That's really smooth, you've replaced "tube" with "highway". Well played

Personally, I've found that people who label other people as "liberal" or "conservative" and then dismiss everything someone has said because of it generally don't know what they're talking about

as shitay as this may be to some i have to say that the t-mobile throttling is fair enough at 5gb. i do alot of tethering since the airforce moves me alot and i still only use about 3gb's. im sure this is a way better solution. unless people really want to be running on cricket's data speed. point in this all is that we have to learn to give and take, and yes i know t-mobile can suck it or whatever but i'd pick throttling over data capping or waking up to cricket data speeds anyday. we gotta be considerate about how many people uses the same web u do

These people do not get it I am the costumer not Microsoft (bing) or Yahoo, ESPN ect when I pay for the phone that phone is mine when I pay that service I should be able to use how ever I want no one has ever used so much data. These carriers need to put there money were their moth is and upgrade the back-hall and crate and buy more bandwidth. I do not think that people understand that bandwidth is a man made thing and it can be easily be made in infinite amounts.

The network is like a road on a road you polices the polices are part of the government not the road or road maker ,and that is how should be.

bandwith is quite finite actually. there are only certain parts of the spectrum that are useful for what u want. And no, u should be able to use the service for what you agreed to (data cap, throttling after Xamount, ect).

What I think would be interesting to see is a surcharge for true unlimited with no data cap. Like tmobile has a 200mb/ $10 plan comming out for light use, the regular $30/unl with data throttling at 5gigs, and maybe like $50-60 for unl with no throttling (smartphone and aircard). That way everyone can get what they need at a fair price. With the extra money they get from the true unlimited they can boost their network capacity faster.

So how are the "problems" on wireless networks any different than on broadband just a few years back? All these guys are looking for are excuses to keep charging exorbitant rates for services on an exponentially rising userbase and traffic patterns, yet build their infrastructure at a barely noticeable linear rate.

T-Mobile's example happens to landline ISP all the time. So what!

Net neutrality must apply to ALL networks. Wired or wireless doesn't make any difference.

The thing is, it is different. With a wired network, you can always increase capacity by adding more physical infrastructure or upgrading your infrastructure.

With wireless, you have a certain set of radio frequencies that physics dictates can only handle so much. Once a cell site is at capacity, you can't add more capacity without increasing your radio spectrum. You also can't increase cell tower density past a certain limit without causing interference, which will lower speeds you can get. Backhaul is certainly an issue, but that can only get you so far.

This is why I think that throttling is a good idea as long as it's done in a neutral manner. I'd be against cell providers allowing full speed for certain services because of an agreement at the detriment of a competitor. As long as the transfer limit is clearly labeled and communicated, there shouldn't be a huge problem. If a customer doesn't like that limit, they can switch providers, not choose that provider to begin with, or pay a bit more to get a higher limit.

Why aren't they slicing into those fat profit margins and creating more proficient networks that can handle tomorrows bandwidth needs instead of upgrading to handle yesterdays? That's the real answer. I'll tell you why. This is another revenue stream to add to those fat profit margins.

If a carrier throttles your bandwidth then you should be given a % discount for how much they have throttled. Your paying for 3g you should get 3g speeds always if your paying for it

This is how I always felt. But I'm starting to think that it might be better to charge $x.xx more/month for the opportunity to NOT get throttled for things that normally are.

The carriers are gonna win this, one way or the other. We just have to recognize what the best compromise will be for US as users, and accept it.