Can An Open OS Ever Really Be Mainstream?

Nexus One via TwitpicThe recent announcement that Google plans to deliver an unlocked mobile phone into the market sometime next year has been an encouraging sign for fans of the open operating system that finally wireless carriers won't be able to control what phones their service customers can use. Many feel as the Wall Street Journal technology columnist, Walt Mossberg does that carriers have been acting like "soviet ministries" as they intermediate between the consumer and the providers of the handsets they use to connect to the carrier networks.

Having launched the T-Mobile G1 as an executive with the company, I have a great affinity for the open Android platform. I appreciate that the Android marketplace enables garage developers to create magic as moonlighting inventors, and brings innovation to the masses through the power of the open programming interfaces and developer tools Google provides online.  But I also saw first hand the customers who, after downloading 10 random apps, wondered why their battery life halved or the screen seemed no longer responsive.

The open developer model has given anyone who can code access to consumers without an accompanying process to ensure they put quality product on the shelves, and as a result more developers step in and create solutions like Astro, an Android task manager to help manage processes, tasks and files that may impact your Android device's performance. Much like on my Windows PC, I find I am delighted to have such a tool and aggravated when I have to use it. It seems I rarely find myself on my iMac, iPod or iPhone worrying about multi-threaded processes or unresponsive programs. And for most consumers, that's one more thing to love about the Apple OS. Sure, it comes with the cost that I can't have apps running in the background on my iPhone, but my iPhone rarely hangs, crashes or has a radical change in the battery life with each new app I might download to it.

Ratings and reviews of apps in the open market are meant to help consumers, but I often wonder which reviewers to trust and whether one app offers the complete solution I need or a more usable interaction model for my tastes. In the case of Astro, several apps purport to do some or all of the capabilities. Some charge. I then wonder, will the quality be the same for the developer who isn't getting paid?Courtesy of Gizmodo Will they maintain the app? Will they support me if I have trouble? Will they care if the application doesn't work well with other applications I may download? And how will I know if they conflict until I download them. A reviewer of the application may not have the same things on their phone that I do, or want to use their phone as I do.

In a world where there are infinite ways to configure a phone with settings and application combos that meet any user's specific needs, the best solution a service rep can offer when a customer complains about their device's performance is to wipe it clean and start over. But facing that experience when you need to place a call and your phone is frozen is daunting. As an example, last night, my home screen theme application was corrupted and the home screen displayed a message compelling me to force it to close. After five times of doing that and not being able to break the cycle, I removed the battery and I removed the SIM. Neither action, both typically offered as the first cure by carrier care reps who don't know what apps I may have downloaded and configured, repaired the problem. The device seemed completely inaccessible and unusable. After a trip to the T-Mobile Forums and a hard reset, which removed all settings and personalizations,  I was able to make a call more than twenty minutes later. But now, which apps to re-load? How do I know what was the offending piece of code?

As geeky as I am, I still want things to just work, and I get frustrated when I use applications that allow me to do things I really shouldn't or require me to understand arcane technical jargon. And I don't have the time to fuss with bad design to engage and interact with a solution. The challenge with open is that everyone can play, but maybe for consumers that isn't always going to be a simple way to have compelling experiences.

Update on Monday, January 11, 2010 at 12:05PM by Gearhead Gal

My Nexus One arrived while I was at CES. Having launched the first Android device, the T-Mobile G1, as a VP of Product Development at T-Mobile, I have a particular insight into what Google values and doesn’t value about the existing business model for mobile phones. So I was particularly intrigued to see their unlocked phone experience. Under the banner of “do no evil” they represent that the company – and the Android team – know what’s best for consumers. But if they did, they would know that sending an auto-generated email response to a customer care issue that suggests that a service response will come within 72 hours is not a good consumer experience. While they do not have a recurring subscription relationship with the customer that the carriers do, their brand is on the device, the consumer bought the product from their web site, and they have a responsibility to provide the service level a $500+ product should deliver. The fact that a GREAT customer experience may have been triaged for launch is no excuse not to have at least a good one.

I also believe that the consumer does not know how to use or trust the “community” in the apps marketplace. Reviews and ratings mean nothing if they come from the gang of software hackers and pirates in collusion with the developer who published the app. Enabling apps to enter your phone that can steal personal data also doesn’t feel like it falls under the banner of “do no evil.”  For an open OS to be mainstream, the consumer MUST know where to place their trust.  Apparently GigaOM agrees...

Comments

[Chris Karr]

I think that there are a couple of factors being conflated here. I understand the problems that you have identified, but I wonder how many of the problems are due to growing pains and how many of the problems are due to an inherent property of open platforms?

I also think that you romanticize too much the traditional model of the carrier serving as a quality-assuring gatekeeper that protects the consumer from lazy or malicious developers. As I recall, prior to the iPhone, the quality of devices pushed out by the carriers were hardly of the highest quality. If the carriers were doing their job quality-wise, I doubt that the iPhone would have been the hit that it was. I distinctly recall the terrible software that Motorola shipped on their devices, all the half-assed J2ME implementations, and carriers intentionally crippling devices in order to swindle their customers.

I'll take the Wild West of Android over 2005-era phones any day of the week. As a former iPhone user and recent Droid convert, I'm also quite happy being on a platform that isn't being artificially constrained by a "father knows best" attitude from its vendor.

[Gearhead Gal]

I appreciate the feedback, Chris. And thanks for the comment and thoughtful remarks.

I don't think I mean to romanticize the carrier role. And I agree the carriers haven't got a great track record as providers of great experiences. I believe, though, after watching customers return devices to the T-Mobile store because of sluggish performance and badly developed apps that drain battery by leaving the GPS or Wifi going or pinging the network when they aren't in use, that all app developers in an open system are not always going to look out for a customer's well being. Since the carrier is the one that the customer pays today, the carrier has an obligation to the consumer to make a better experience, even if they haven't shown their commitment to always do so.

With an unlocked device, the open model gets even riskier for a consumer. Now I have trouble and I don't know who to call - the carrier, the OEM, or the software maker. While the model exists in computing to separately purchase hardware, software and service, each one of the contributors in the value chain understands their role, and their business models enable a consumer to be supported. Today, with the unlocked phone you buy from Google, not HTC, and the way the handset is recognized on the network is through the carrier. How is it mainstream for a consumer to figure out where to go for service? Who is accountable to the consumer in making the diagnosis when a handset isn't performing because an app behaved badly or a component isn't working in the chipset? What we see now is the finger pointing, bad service levels (Google's email say someone will get back to you in 3 days!) , and no responsibility.

Google could have easily set up a outsourced model to ensure great customer service beyond web email. It could also simply have provided a (3rd party?) certification process for apps to ensure developer citizens play more nicely in the sandbox. Alternatively, it could provide users a toolkit (like utilities in the OS) that would help them verify, in simple English words not geek-speak, what the implications are for the device services the applications on the market use. Instead they list the technological components, but what consumer knows the implications? Especially if the app presents risk that their personal information may be co-opted, their phone may not place calls, or their device will have no resources available to load a website page or play a video?

That's all I am saying. Wild west is great for software innovation. It isn't always great to give consumers so much freedom that they render the product useless and then can't figure out how to service it.

[Chris Karr]

First of all, I agree with you that Google could have handled the Nexus One rollout a bit better. On the other hand, I'm willing to cut them a little slack as this is the first phone that they have launched. If I recall correctly, Apple's iPhone rollouts weren't so hot to begin with. Let's give Google some time to demonstrate that they have learned something before calling it quits on the open phone model.

Now, to address your other points:

1. With respect to risk, I disagree that the unlocked model is riskier. Looking at it in financial terms, the cost of a locked device is much more than the cost of the device alone. I might be sympathetic to this point if the carriers were not so greedy with their 2 year contracts. (Why doesn't the cost of my service go down after 2 years and the carrier's recouped their expenses?) If I buy an unlocked phone and it doesn't work out, I can turn around and sell it on the open market. If I'm locked into a contract, I'm stuck with the lemon for the next 2 years. From my perspective as a consumer, I would feel much freer under a system where I buy the phone for what it costs and can choose to go month to month.

2. Now with respect to the question of support, I think we would be better off if each member of the value chain supported the parts that it was responsible for building or providing. If I have a phone problem - I go to Google or HTC. If my AT&T service is spotty, I complain to them, and not Apple. Unfortunately, due to American carriers' propensity for lock-in, this scenario cannot happen because I cannot move my device and service to another provider that provides better customer support. If I have a bum phone, I have to deal with the carrier and not those who built the device.

Despite Google's initial missteps, I think that they should be thanked for what they are doing, and that's trying to break up what has traditionally been a very vertical & siloed set of "value chains". I think that in a more fragmented and commoditized marketplace, customer support will actually improve as other players can step in to fill the void left by negligent carriers or handset manufacturers.

I would point to the PC market as an example where support is not an issue with a more complex open platform. Sure people have problems, but the platform has developed a complimentary ecosystem of troubleshooters and service providers to help those having issues. Why wouldn't this happen in a free and consumer-friendly open marketplace for devices and wireless service is beyond me. I also don't understand why if these issues you cite are problems for open computing platforms, how the personal computer has become such a critical part of our lives. We can't trust Windows or Mac software from just any source, yet the platforms have become indispensable in modern life. I guess you haven't shown me how a phone is something other than a smaller computer with a wireless Internet connection.

3. Now with respect to a 3rd party verification service - why does Google have to be the only one responsible for this? Where's T-Mobile spending their resources to do this themselves? Rather than look at certification as an obligation for Google, can't we be a bit creative and recognize a carrier-sponsored 3rd party app vetting service as a selling point for one carrier or another? Think of the advertising: "Are you sure you're safe on those other carriers? Only T-Mobile's protects your phone with our 'TruSafe' Android Marketplace." I guess my suggestion is to let a thousand app stores flourish - the trustworthy ones will float to the top and solve the vetting problem. Or to make another analogy - I won't by Photoshop from some guy selling it in the subway, but I trust that I'm getting a good copy from Dell or Amazon.

4. I do agree that Google should use plainer language in describing the app risks. On the other hand, I'm willing to spot them a few iterations to get this figured out.

[Chris Karr]

Let me pose another question for you. If during your tenure as a representative of T-Mobile, you caught wind of a small regional customer support outfit that was in the business of offering paid support for proprietary T-Mobile devices, what would the carrier's reaction have been? Would this small business be allowed to continue supporting your users completely independent of the T-Mobile mothership, or would it be hit with a cease-and-desist letter the next day?

[Gearhead Gal]

Assuming I understand your definition of "customer support", and assuming you realize I am not speaking on behalf of any carrier, I see no reason a carrier shouldn't support that paid service. First, it would offload calls and costs. Second, that model exists today with Value Added Resellers for business accounts.

Am I missing something in the question?

[Chris Karr]

First of all, apologies for any overly strident tone in my question - that was not my intent. I just get fired up about this topic. :-)

To clarify, I was wondering if mobile carriers would support mom & pop (or similar) operations providing paid support to end users without the carrier's permission or paying the carrier any money to help local users solve problems with their phones. I can set up a small business troubleshooting and optimizing personal computers without seeking Microsoft, Intel, or Apple's permission. (Although I do recognize that there are big advantages to being certified through those organizations.)

This is my way of getting around to the idea the big players don't have to do anything (e.g. Google supporting each and every user). Where there's customer confusion and problems with devices, I see a business opportunity for independent agents to fill that void and become part of the open ecosystem. This could take the form of dedicated mobile device support companies that are paid by incident or existing support organizations adding phone support to their portfolio of services. What my question is asking indirectly is whether the market is sufficiently open and free that such services can exist or whether carriers would strangle any such efforts in the crib under the excuse of preventing interference with their subscribers.

Does that make sense?

[Chris Karr]

Correction:

"... big players don't have to do anything..." should be "... big players don't have to do everything..."

[Gearhead Gal]

I appreciate the passion and thoughtfulness in your comments. And believe me, you are wrong if you think I am married to the carrier model or don't believe in open innovation. My perspective, and the one I write about, is the average user. I feel the need and obligation to pressure Google and its ecosystem partners to realize that consumers care less about software and more about experience. My husband, my neighbor, my mom...they all just want their phone to place calls, receive a text message and occasionally display maps or web pages. They don't care to see HREF: appear, they hate using the Task Manager in Windows to kill a frozen app, and they don't like being asked if they are ok with the consequences of acting when they are doing something they have never done before.

Pardon me, but you are a software expert. You are not the normative consumer who picks up a Nexus One and is presented with call settings options like Enable FDN and says, WTF? Nor are you the person who might have trouble understanding that full bars doesn't mean you can make a call.

Sure, carriers can certify apps, but Google did not approve of signing apps initially. And carrier certification may come to mean "Carrier preferred" in the hands of a marketer, instead of device friendly. Microsoft certifies (people, apps, and devices that make their ecosystem), Apple certifies (accessories and software to ensure a good experience), so it should follow Google certifies..

My last comment on big players...I completely agree they don't have to do anything. That is precisely my point. Google could have established a network of paid support vendors while building the device. They could have cared about training and certifying support partners that consumers could reach, instead of servicing them solely with an auto-email. But they did NONE of that for launch. It showed a total disregard for the consumer paying $500 not to have ANY plan in place beside their normal impersonal service model, which works fine for most relationships I have with Google, but not when I buy a $500 device from them that is supposed to send and receive calls and it isn't working.

It's an open model when Google doesn't want to do something, and leave the clean up to the "market". My passion is that the whole customer experience is one they should have taken a more accountable role in and haven't.

[Chris Karr]

Thanks for for the final comments. I think we're both interested in reaching the same final endpoint (usable & powerful devices). I just doubt that the Apple approach is the best model long term for getting there. On the other hand, according to recent reports, Apple's addressing some of the major issues with the App Store approval process, so it's clear that they've been learning as well.

The thing that excites me most about this industry and this particular point in time is that I think we'll see how these open vs. closed platform debated play out in short order. I'm clearly a fan of something more like the Windows ecosystem than the current iPhone ecosystem, but it will be interesting to see if the market follows suit. There's quite a bit of upheaval at the moment, but I'm excited to witness the next major stage in computing evolution.

Thanks for the spirited discussion. I apologize if my passion for this debate got the most of me, but I'm quite happy that I discovered your blog in the process. :-)

[Gearhead Gal]

I'm all for a new ecosystem model...but I don't like the way that open oligarchical model of Android has been used as an excuse for enabling bad consumer experiences.

I'm delighted to have the dialogue. I have been delighted to see that anyone reads what I write, let alone take the time to comment! Thank YOU! Please keep reading and challenging me.