.

7/9/09

Google's Chrome OS, Cloud-oriented Computing, and User Freedom; or The Emperor has No Clothes!

Google has recently announced their plan to release a new open source operating system called Chrome OS, which they hope to make freely available by mid 2010. A quick review of write-ups about Google's announcement (like this piece by Miguel Helft and Ashlee Vance in the New York Times) reveals that many view the proposed Chrome OS to be an innovative approach to desktop computing, since the idea is to make Google's Chrome browser the focus of the operating system. With little or no native applications to bog it down, Chrome OS would thus provide the user with quick access to the Internet, and nearly all computing tasks would take place through web applications such as Google Docs and Picasa.

It might first seem counter-intuitive, but by giving away Chrome OS and their web services, Google will ultimately grow their business. Google makes its money from targeted advertising. Widespread usage of Chrome OS would mean that more people would be doing their computing in the clouds , which would in turn provide Google greater opportunities to offer more precisely targeted advertisements at a greater quantity, and hence greater opportunity to increase revenue.

If Google's cloud-centric operating system catches on, it will most certainly pose a challenge to businesses that currently sell traditional desktop operating systems, such as Microsoft and Apple. With an orientation on native rather than web-based applications (e.g. Word, Windows Media Player/Quicktime, iPhoto, etc.), Windows and OS X stand in the way of Google making more money, since the more time one spends on those native applications, the less time one spends online using Google's web-based services. By offering Chrome OS for free, Google undercuts Windows, OS X, and their expensive software, driving consumers to use Google's services instead.

All this might sound like a big win for the consumer, since he or she will get to use quality applications on a purportedly more reliable and efficient operating system at no cost. Under any ordinary circumstance, who could argue with that? But herein lies the problem. This is no ordinary circumstance. We're now dealing with cloud-oriented computing, and with this, free comes at a high price.

Cloud computing can be quite useful. However, to make an operating system completely dependent upon web services for its most basic functions poses certain dangers to the user. First, all one's computing becomes dependent on having an Internet connection, which means one must have an Internet service provider in order to utilize the system to its fullest potential. Google will likely further develop Google Gears, which currently allows the user to work with certain web-based applications offline, but it will probably never be able to provide the same functionality to web-based applications as one has with native applications. So for those who don't want or can't afford to pay for an Internet connection (yes, I know, this is a small demographic in many nations), or for those who have no access to the internet for an extended period of time, Chrome OS would appear to be practically useless. Even for those who have an Internet connection, why would they want to have that cost become an inherent part of their ability to use their computer?

Second, it isn't clear whether one will have the ability to write and run non-web-based applications on one's computer. Google may allow for such a feature, but it will probably be disabled by default, seriously restricted, or come at a price. These applications will compete for the user's time, which he or she would otherwise spend using the web-based applications that bring Google revenue. So the freedom of the user to write his or her own program and run it on his or her computer the way he or she sees fit will likely be restricted or taken away.

Finally, cloud-oriented computing means that one's private data will not (only?) be on your personal hard drive, but it will (also?) be sitting on the hard drive of some third party server, meaning God-knows-who could have access to your private information, doing God-knows-what with it. This raises a series of important related questions: is data in the clouds ultimately public? We're talking about a centralized storage location that contains very intimate details about the lives and dealings of billions of people the world over. This places unthinkable power in the hands of who or whatever owns that centralized storage location. If the information is there for another to be accessed, in what sense is it still private information? Do the storage location owners only own the storage, or can they lay any claim to the contents of the storage, as well? What can they do with the information, and who or what is to stop them if they try to do something with it that they shouldn't? Certainly the private/public distinction begins to get blurred; at the very least, this would involve having limited control over any personal information stored in the cloud. And when all or nearly all one's computing takes place in the cloud, one would have limited control over a large bulk of that information.

When all is said and done, the Chrome OS platform may end up being cheaper, more efficient, and more innovative than Windows or OS X. It may be built around the Linux kernel. It may also be touted as a free/libre open source project, but it cannot help but result in an arrangement that is at least as equally unethical as the arrangement between proprietary software companies and their end users, since users will end up surrendering to their service provider a modest amount of control over system functionality, as well as security of personal data. Chrome OS may be free/libre open source in name and in practice, but the very nature of the relationship between user and provider that cloud-centric computing fosters entails that it cannot be fee/libre open source in spirit.

Creative Commons License
Google's Chrome OS, Cloud-oriented Computing, and User Freedom; or The Emperor has No Clothes! by Nathan M. Blackerby is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License.

3 comments:

  1. First, let me state I am playing devil's advocate to some extent because we do not have all of the details about ChromeOS. This ChromeOS press release has created such a buzz and it genuinely surprises me. I have read 50 some blog entries and seen tens of new articles, all over a single blog entry and FAQ on Google's site.

    Before I get to my thoughts on the ethics of Google and cloud, I would like to state my doubts about Google's success and the technical & business challenges this project poses. I have seen so many bad reviews of the ChromeOS idea. Most criticize the fact that Google is loosing it's focus and to some extent I must agree, though not completely. First, Android has had a slow start, some might say very slow. Other's might say, it has been basically a flop. Sales of Android have been minimal.

    At best, I think Google does seem a bit brash in throwing out the idea of another operating system, which mind you, competes in some of the same space as Android. Overlapping products, is strange to begin with. Perhaps Google feels attacked by Bing, so they came back with an attack on MS Windows. In the end, I am very skeptical that Google has a chance of even making a dent in the OS Market with ChromeOS. Firefox has done well, but Linux usage lags far behind. Google, may have more work cut out for them than they can handle and Microsoft is probably thinking to them selves, "Wonderful, spend some of your war chest on this, Hua, Hua, Hua."

    From the usability and ethical perspective, there are some challenges. From the network perspective, Google gears basically solves the problem of needing an Internet connection. It basically performs a hidden sync of your data, so you can work off line indefinitely. Besides, if you never had an Internet connection, then you never downloaded ChromeOS, and you definitely never had time to upload/sync all of your documents, so I think it is a fairly easily solved chicken/egg problem.

    I think the biggest challenge is with regard to where your data sits. There are safe (safety is subjective and relative) ways to store data in the cloud, but the lawyers have more to say about that than the tech guys like me. Obviously, asymmetric (one way) encryption could be used, but the entire suite of Google apps would have to adhere to a strict legal standard that restricts them from storing/using keys/data. That is unlikely at best.

    More likely all of your privacy protection will rely on some kind of EULA which can change daily or hourly. This would make many feel uncomfortable. On the other hand, there are already bot nets out there that have millions of users in them, and those people never signed a EULA. People who's computers have been compromised by bot nets are sharing their data to people who are probably much worse than Google and probably providing the framework for all kinds of identity theft scams.

    Personally, I think I still like the technical control called "My Computer" in place where my data is mostly on my hard drive, but I may just be getting old. Really, I am not sure if it is safer or free-er, I just feel better. In the end Google's physical/network security may be better than my computer, but the legal security in either case is much harder to put my finger on. Remember, these line have been getting blurrier and there are many cases of government/law enforcement snatching up computers and network traffic under suspicion.

    ReplyDelete
  2. Thanks for the comment, Scott.

    I agree with much of what you say regarding the challenge Google faces in making Chrome OS a successful product. Why would they fund two products that they will give away for free and essentially perform the same function, revenue-wise? It's a major overlap (but perhaps they will put an end to the less successful project). And they really are going to be fighting an uphill battle. Some people are already talking about Chrome OS being a "Windows Killer." But I doubt it. Microsoft and Apple are clearly well positioned enough in the market that they have the luxury to observe the kind of success Chrome OS might have and in what areas. They can then adopt their own software to consumer demand and essentially marginalize the impact of Chrome OS. Based on my non-rigorous observation, I believe that a major factor - perhaps the most major factor - in getting consumers to switch to a new operating system is providing them with a familiar desktop environment, regardless of whether the operating system comes at a cheaper cost, or even no cost at all. Linux already has troubles in this regard, and it's both low/no cost and, with certain distributions, offers a mostly familiar desktop environment. Who knows what kind of user experience Chrome OS will provide? Consumers might be familiar with a browser, but they most likely won't know what it would be for "the browser to be the operating system." And many of them will not know what cloud computing is. Google will have to present these things (or, possibly better, not present them, since it may cause confusion) in a manner that consumers are already familiar with. Of course, consumers are already familiar with Microsoft and Apple software, so they have a decided advantage in this regard; all they'll have to do is add cloud features to their already existing applications.

    I also think the security and ethical issues are tied to the success of Chrome OS. You might rank the issues I raised as follows, based on what I believe consumers will be worried about:

    1. Will the system be able to fully function without an internet connection?

    2. Will personal data be secure?

    3. Will non-web-based applications be able to be developed and run on the system. (and will web-based applications be able to be modified to be non-web-based)?

    I'm not counting out the possibility that Google has an answer to each of these questions. In fact, I think that (1) is the easiest to answer, then (2), and so on. Most consumers will probably be happy if Google can successfully answer (1) and (2). But (3) is the real crux of the issue for me. As I mentioned in the blog entry, Google has no real incentive to allow development of non-web-based applications, so my suspicion is that they won't, or they will restrict it in some way by closing parts of the code, or they will treat is as a fee-based privilege . If they do allow non-web-based-application development, then I guess the emperor will end up having clothes after all.

    ReplyDelete
  3. Great information! I enjoyed this one, and many of the other posts you've been putting out lately.

    ReplyDelete