It was a nondescript bar in the American Midwest, the sort of place where working men drop in at the end of the day to unwind before they head home. You wouldn't expect to find two senior business executives there, and as I sat in the empty bar at midday I wondered if maybe my contact had given me a bad lead. But then the door opened and a general manager from one of the leading web companies walked in, followed by a senior VP from one of the US's biggest mobile network operators. I hunched down in the shadows of a corner booth and typed notes quietly as they settled in at the bar.
Bartender: What'll you have?
Telecom executive: Michelob Light.
Web executive: I'll have a Sierra Nevada Kellerweis.
Bartender: Keller-what?
Web executive: Um, Michelob Light.
Telecom executive: Thanks for coming. Did you have any trouble finding the place?
Web executive: All I can say is thank God for GPS. I've never even been on the ground before between Denver and New York.
Telecom executive: I wanted to find someplace nondescript, so we wouldn't be seen together. The pressure from the FCC is bad enough already, without someone accusing us of colluding.
Web executive: No worries, my staff thinks I'm paragliding in Mexico this weekend. What's your cover story?
Telecom executive: Sailboat off Montauk.
Web executive: Sweet. So, you wanted to talk about this data capacity problem you have on your network...
Telecom executive: No, it's a data capacity problem we all have. Your websites are flooding our network with trivia. The world's wireless infrastructure is on the verge of collapse because your users have nothing better to do all day than watch videos of a drunk guy buying beer.
Web executive: Welcome to the Internet. The people rule. If you didn't want to play, you shouldn't have run the ads. Remember the promises you made? "Instantly download files. Browse the Web just like at home. Stream HD videos. Laugh at an online video or movie trailer while travelling in the family car."
Telecom executive: That was our marketing guys. They don't always talk to the capacity planners. Besides, who could have known that the marketing campaign would actually work?
Web executive: Don't look at me. I've never done a marketing campaign in my life. I think you should just blame it on A--
Telecom executive: You promised, no using the A-word.
Web executive: Sorry. But I still don't see why this is a problem. Just add some more towers and servers and stuff.
Telecom executive: It's not that simple. The network isn't designed to handle this sort of data, and especially not at these volumes. Right now our biggest problem is backhaul capacity -- the traffic coming from the cell towers to our central servers. But when we fix that, the cell towers themselves will get saturated. Fix the towers and the servers will fall over somewhere. It's like squeezing a balloon. We have to rebuild the whole network. It's incredibly expensive.
Web executive: So? That's what your users pay you for.
Telecom executive: But most of them are on fixed-rate data plans. So when we add capacity, we don't necessarily get additional revenue. It's all expense and no profit. At some point in the not-too-distant future, we'll end up losing money on mobile data.
Web executive: Bummer.
Telecom executive: More like mortal threat. Fortunately, we've figured out how to solve the problem. The top five percent of our users produce about 50% of the network's total traffic. So we're just going to cap their accounts and charge more when they go over.
Web executive: Woah! Hold on, those are our most important customers you're talking about. You can't just shut them down.
Telecom executive: The hell we can't. They're leeches using up the network capacity that everyone else needs.
Web executive: Consumers will never let you impose caps. You told them they had unlimited data plans, that's the expectation you set. You can't go back now and tell them that their plans are limited. They won't understand -- and they won't forgive you.
Telecom executive: First of all, the plans were never really unlimited in the first place. There's always been fine print.
Web executive: Which no one read.
Telecom executive: Off the record, you may have a point. On the record, the fact is that you can retrain users. Look, you grew up in California, right?
Web executive: What does that have to do with anything?
Telecom executive: Once upon a time, there weren't any water meters in California. Now most of the major cities have them, and they'll be required everywhere in a couple of years. Something that was once unlimited became limited, and people learned to conserve.
Web executive: The difference is, I can read my water meter. You make a ton of money when people exceed their minutes or message limits, and you don't warn them before they do it. If you play the same game with Internet traffic, it'll scare people away from using the mobile web -- or worse yet you'll invite in the government. Look what happened with roaming charges in Europe.
Telecom executive: Jeez, don't even think about that. Okay, so we'll need to add some sort of traffic meter so people will know how much data they're using when they load a page.
Web executive: Great, that'll discourage people from using Yahoo.
Telecom executive: Huh?
Web executive: Oops, did I say that out loud?
Telecom executive: Then there's the issue of dealing with websites and apps that misuse the network.
Web executive: Not this again.
Telecom executive: I'm not talking about completely blocking anything, just prioritizing the traffic a little. Surely you agree that 911 calls should get top priority on the network, right?
Web executive: Of course.
Telecom executive: And that voice calls should take priority over data?
Web executive: I don't know about that.
Telecom executive: Oh come on, what good is a telecom network if you can't make calls on it?
Web executive: (sighs) Yeah, okay.
Telecom executive: So then what's wrong with us prioritizing, say, e-mail delivery over video?
Web executive: Because when you start arbitrarily throttling traffic, I can't manage the user experience. My website will work great on Vodafone's network but not on yours, or my site will work fine on some days and not on others. How do you think the customers will feel about that?
Telecom executive: Not as angry as they will be if the entire network falls over. Listen, we're already installing the software to prioritize different sorts of data packets. We could be throttling traffic today and you wouldn't even know it.
Web executive: But people will eventually figure it out. They'll compare notes on which networks work best and they'll migrate to the ones that don't mess with their applications. Heck, we'll help them figure it out. And if that's not enough, there's always the regulatory option. The Republicans are out of office. They can't protect you on net neutrality any more.
Telecom executive: You think you're better at lobbying the government than we are? We've been doing it for 100 years, pal. Besides, we have a right to protect our network.
Web executive: You mean to protect your own services from competition!
Telecom executive: Parasite!
Web executive: Monopolist!
Telecom executive: That's it! It's go time!
They both stood. The telecom guy grabbed a beer bottle and broke it against the bar, while the web guy raised a bar stool over his head. Then the bartender pulled out a shotgun and pointed it at both of them.
Bartender: Enough! I'm sick of listening to you two. Telecom guy, you're crazy if you think people will put up with someone telling them what they can and can't do on the Internet. The Chinese government can't make that stick, and unlike them you have competitors.
Web executive: See? I told you!
Bartender: Shut up, web guy! You keep pretending that the wireless network is infinite when you know it isn't. If you really think user experience is important, you need to start taking the capabilities of the network into account when you design your apps.
Web executive: Hey, he started it.
Telecom executive: I did not!
Bartender: I don't care who started it! Telecom guy, you need to expose some APIs that will let a website know how much capacity is available at a particular moment, so they can adjust their products. And web guy, you need to participate in those standards and use them. Plus you both need to agree on ways to communicate to a user how much bandwidth they're using, so they can make their own decisions on which apps they want to use. That plus tiered pricing will solve your whole problem.
Telecom executive: Signaling capacity too. Don't forget signaling.
Bartender: That's exactly the sort of detail you shouldn't confuse users with. Work it out between yourselves and figure out a simple way to communicate it to users. Okay?
Web executive: I guess.
Telecom executive: Yeah, okay.
Bartender. Good. Now sit down and start over by talking about something you can cooperate on.
Telecom executive: All right. Hey, what's that guy doing in the corner? Is that a netbook?
Web executive: He's a blogger!
Bartender: There's no blogging allowed in here!
Telecom executive and web executive: Get him!
I ran. Fortunately, the bar had a back door. Even more fortunately, the web guy and the telecom guy got into an argument over who would go through the door first, and I was able to make my escape.
So I don't know how the conversation ended. But I do know that I wish that bartender was running the FCC.
Showing posts with label operators. Show all posts
Showing posts with label operators. Show all posts
Thursday, 29 October 2009
Friday, 30 November 2007
Another example of why the tech industry and mobile operators don't get along
When you work in consumer electronics, one of the rules that gets drilled into you very early on is that you never do anything to disrupt the holiday selling season. In the US, the month between Thanksgiving and the end of the year can account for three months' worth of sales, if not more. During that time, you don't change prices, you don't alter your ad campaigns, and most of all you don't ever say anything about future products, because that might cause customers to hesitate before making a holiday purchase.
So the CEO of ATT, giving a speech this week in Silicon Valley of all places, says that a 3G version of the iPhone is in the works:
Not only does he spill the beans, but he acknowledges that Apple hasn't announced it and then talks about it anyway (link). The next day the story is carried by the AP, MSNBC, the Times of London, Wall Street Journal, Bloomberg, the SJ Mercury News (which had the version of the quote above), and 318 other publications according to Google.
In the US, the assessment from a lot of commentators is that this won't have much impact because the iPhone is so popular anyway. Maybe, I guess, although the iPhone isn't sold out, so any loss in sales is still a loss. But in Europe, I think it could be a big problem. iPhone sales there are not going great to begin with, and folks in Europe are generally much more conscious of 3G vs. 2G issues. The acknowledgment that a 3G iPhone is coming could cause a lot of people to hesitate before buying.
If AT&T competed directly with Orange, O2, and T-Mobile Germany, I'd be tempted to speculate that they made the announcement on purpose to hurt the competition. But they don't, so I suspect this is just a case of a CEO who wanted to show that he's not controlled by Steve Jobs but instead demonstrated that he doesn't understand consumer electronics.
So the CEO of ATT, giving a speech this week in Silicon Valley of all places, says that a 3G version of the iPhone is in the works:
"Has Jobs announced that? I don't think he's announced that, but you'll have it next year."
Not only does he spill the beans, but he acknowledges that Apple hasn't announced it and then talks about it anyway (link). The next day the story is carried by the AP, MSNBC, the Times of London, Wall Street Journal, Bloomberg, the SJ Mercury News (which had the version of the quote above), and 318 other publications according to Google.
In the US, the assessment from a lot of commentators is that this won't have much impact because the iPhone is so popular anyway. Maybe, I guess, although the iPhone isn't sold out, so any loss in sales is still a loss. But in Europe, I think it could be a big problem. iPhone sales there are not going great to begin with, and folks in Europe are generally much more conscious of 3G vs. 2G issues. The acknowledgment that a 3G iPhone is coming could cause a lot of people to hesitate before buying.
If AT&T competed directly with Orange, O2, and T-Mobile Germany, I'd be tempted to speculate that they made the announcement on purpose to hurt the competition. But they don't, so I suspect this is just a case of a CEO who wanted to show that he's not controlled by Steve Jobs but instead demonstrated that he doesn't understand consumer electronics.
Tuesday, 27 November 2007
Questions about Verizon's new "open" attitude
More than half of the traffic to this weblog comes from outside the US, so there are times when I feel obligated (and a little embarrassed) to explain how the mobile market works here. This is one of those moments.
Verizon, the largest US mobile carrier, made headlines in the US today by announcing that by the end of 2008 it's going to make its network available to any device and any application that the user chooses to install (link).
This will seem remarkable to people living in GSM countries where it's normal to choose any device you want. But in the US, it's an unusual idea. Here mobile usage is split between GSM and CDMA. GSM phones have SIM cards, which technically allow you to switch your account to any phone you want. But in practice, almost no users are willing to give up the several hundred dollar subsidy for buying a phone and service plan together, so they only choose phones that come through the operator.
Things are even more restrictive in the CDMA space, where there are no SIM cards. If you buy a Verizon phone, it can only be used with a Verizon account. Same thing for Sprint.
So Verizon's announcement is a nice change, on the face of it. It's also something of a pleasant shock, since Verizon has the reputation of being the most conservative and controlling US operator. But the announcement's actual impact on the market is going to depend on several questions that Verizon hasn't answered yet:
--How will open access be implemented? Verizon says it's going to define a process by which phones can be certified to work on its network. That could be routine or it could turn into a huge barrier to entry. We also don't know how a user's account will be switched between phones. Is Verizon planning to start installing SIM cards in its phones (something that has been done with CDMA in China)? If not, will you have to take the phone to a Verizon store to get it activated? How much will that cost?
Verizon apparently said something about doing activation through a toll-free number, which could be cool.
--How will the service be priced? Verizon's service plans include recovery of the several hundred dollar subsidy for hardware. You pay for the subsidy as part of your monthly bill. Since Verizon doesn't have to recover a subsidy cost on its open access phones, there's about $10 or more a month that it could pass along to consumers in the form of lower bills.
If Verizon doesn't price the open service lower, what happens to the extra money? Does Verizon pocket it? Or will they offer some sort of rebate on purchase of open access phones?
The answer to this one is critical. The US GSM carriers are technically open, but the subsidy prevents significant sales of alternate phones. If Verizon pockets the subsidy money, very few people will take advantage of the open service. The whole thing could turn out to be a PR gesture rather than a genuine change.
But in the hope that Verizon wants it to be more, here's what they ought to do:
--Make the monthly cost of the open plan lower than a traditional service plan, reflecting the absence of a subsidy.
--Make the handset certification process simple and low cost.
--Make it easy for users to switch their account to a new phone (preferably via a SIM card or website or that 800 number, so they don't have to come to a store).
That's an announcement I'd stand up and cheer for.
Impact on the industry
Until we hear the answers to the questions above, it's impossible to guess how impactful this announcement will be. The most important factor may be how the other US operators react. The best result would be if they start competing with each other to see who can make their network more open. If that dynamic takes hold, competitive forces might drive them to really open up even if they don't intend to.
Verizon, the largest US mobile carrier, made headlines in the US today by announcing that by the end of 2008 it's going to make its network available to any device and any application that the user chooses to install (link).
This will seem remarkable to people living in GSM countries where it's normal to choose any device you want. But in the US, it's an unusual idea. Here mobile usage is split between GSM and CDMA. GSM phones have SIM cards, which technically allow you to switch your account to any phone you want. But in practice, almost no users are willing to give up the several hundred dollar subsidy for buying a phone and service plan together, so they only choose phones that come through the operator.
Things are even more restrictive in the CDMA space, where there are no SIM cards. If you buy a Verizon phone, it can only be used with a Verizon account. Same thing for Sprint.
So Verizon's announcement is a nice change, on the face of it. It's also something of a pleasant shock, since Verizon has the reputation of being the most conservative and controlling US operator. But the announcement's actual impact on the market is going to depend on several questions that Verizon hasn't answered yet:
--How will open access be implemented? Verizon says it's going to define a process by which phones can be certified to work on its network. That could be routine or it could turn into a huge barrier to entry. We also don't know how a user's account will be switched between phones. Is Verizon planning to start installing SIM cards in its phones (something that has been done with CDMA in China)? If not, will you have to take the phone to a Verizon store to get it activated? How much will that cost?
Verizon apparently said something about doing activation through a toll-free number, which could be cool.
--How will the service be priced? Verizon's service plans include recovery of the several hundred dollar subsidy for hardware. You pay for the subsidy as part of your monthly bill. Since Verizon doesn't have to recover a subsidy cost on its open access phones, there's about $10 or more a month that it could pass along to consumers in the form of lower bills.
If Verizon doesn't price the open service lower, what happens to the extra money? Does Verizon pocket it? Or will they offer some sort of rebate on purchase of open access phones?
The answer to this one is critical. The US GSM carriers are technically open, but the subsidy prevents significant sales of alternate phones. If Verizon pockets the subsidy money, very few people will take advantage of the open service. The whole thing could turn out to be a PR gesture rather than a genuine change.
But in the hope that Verizon wants it to be more, here's what they ought to do:
--Make the monthly cost of the open plan lower than a traditional service plan, reflecting the absence of a subsidy.
--Make the handset certification process simple and low cost.
--Make it easy for users to switch their account to a new phone (preferably via a SIM card or website or that 800 number, so they don't have to come to a store).
That's an announcement I'd stand up and cheer for.
Impact on the industry
Until we hear the answers to the questions above, it's impossible to guess how impactful this announcement will be. The most important factor may be how the other US operators react. The best result would be if they start competing with each other to see who can make their network more open. If that dynamic takes hold, competitive forces might drive them to really open up even if they don't intend to.
Subscribe to:
Posts (Atom)