Microsoft Windows: more than operating system

(revised May 3, 2008)

Recent reports (such as this one) tell the drooping of Microsoft Windows in the market of operating system (OS). How should Microsoft react to this slide? There may be several options. For example, (1) to accelerate improving Windows Vista, (2) to speed up the migration towards Windows 7, (3) to buy Yahoo, or (4), as I am going to suggest in this post, to explore the potential of Windows that is more than operating system. In particular, I would rather devote this post to the Microsoft Live Search team. Live Search might become a central piece of the new Windows as I would suggest in the following.

Windows, less of momentum to keep on going?

The progress of Windows seems have been slowed down in recent years. I had once asked myself, should I upgrade the installed Windows XP in my laptop to Windows Vista? At last, I decided to wait for a while because I could not see any emergency of the upgrade. Now it seems that this decision might be smart. Many users do not like Vista and they would rather stay with the old Windows XP even for their new computers.

Besides this Vista story, another thread to Windows is from a competitor of Microsoft---Apple. The increasing influence of the Mac operating system has been gradually severe to Microsoft. When the technology of operating system on PCs is becoming more and more mature now, adopting Mac instead of Windows becomes a more and more persuasive excuse especially when the UI design of Mac is generally thought prettier than that of Windows.

Do Microsoft engineers suddenly lose their mind to create this unsatisfactory Vista product? Or are Apple engineers becoming much smarter to compete to Microsoft? Actually, none of the answers are true. The truth is that any further improvement of OS on PC has become so difficult and there are so few rooms to grow on advanced OS technologies on PC that the version-upgrading strategy executed by Microsoft for years may come to its end if no major change of action would been made. We are actually not complaining of how bad Vista is (in fact, Vista is not a bad product). The real issue is that Vista is not innovative enough to be a new version of Windows. Hence if someone really looks for experiencing a new operating system, switch to Mac is probably a better option than upgrade to Vista. This is thus the problem.

Windows, more than an operating system!

To save Windows from collapsing, buying Yahoo is not the only option as some experts have suggested. Although it becomes harder and harder for Windows to upgrade just as an operating system, Windows is actually more than an operating system.

The term "Windows" represents operating system when we talk about personal computers (PCs). But there is a question: will Windows mean something more when a PC is linked to World Wide Web? This question is crucial.

About Windows and World Wide Web, here are several statements.

1. Windows was born when the Web was infant.
2. Windows is not designed for the Web.
3. Windows is an operating system for PCs but it is not a Web operating system.
4. There are no Web operating systems so far and probably there should never been a Web operating system ever.

But all these statements still do not touch the question we just asked: what does Windows to users mean when a Windows-installed PC is connected to the Web?

To answer this question, we should first look at what a PC becomes when linked to the Web. By linking a PC to the Web, the PC automatically becomes part of the Web. Or more precisely, the storage space on the PC becomes a small portion of the gigantic space of the Web. When Windows manages the space on the PC, the Windows manages part of the Web. Hence in this local environment, Windows plays the role to be a web-resource operating mechanism (WebROM). This is a critical recognition of a new function of PC operating systems.

There is another effect caused by linking PCs to the Web. When PC users explore the Web through their PCs, the user activities on the Web result in a local topology of the Web on the PC. If we connect all the Web nodes navigated by the users of a PC, we can obtain a topology of a typical portion of the Web that reflects the interest of the respective PC users. Therefore, linking a PC to the Web causes not only PCs on the Web but also the Web mapped into PCs.

Based on the discussed we have done in the previous two paragraphs, a PC operating system (such as Windows) is indeed a WebROM of not only a particular Web space (the particular local PC) but also a typical topology of Web resources that are physically stored in other places. This is why Windows is more than an operating system by the traditional mean.

I must emphasize that the last statement just made does not, however, suggests Windows to be a Web OS. As I discussed in an earlier post, a WebROM is fundamentally different from a WebOS. What each Windows manages is a topology of a very small portion of the entire Web. Such a topology is consistently a closed world by contrast to that the entire Web is always an open world. I do not believe in generic Web OS.

Live Search, bring a new life for Windows

One Microsoft product will be crucial if Microsoft decides to explore the new Windows by expanding its ability of Web resource management. The product is Live Search.

According to Web resource operating, a central task is search. Unlike local resources stored in personal computers, PC users do not have the full control over Web resources in general. Many traditional OS issues on PCs such as I/O device management are not critical to the role of WebROM. By contrast, PC users do have the right to SEARCH all Web resources even though they cannot control most of them. The resource search operation thus becomes primary.

Up to the date, Live Search at Microsoft is following a route that Google and Yahoo have experienced successfully so far. Microsoft is developing a generic, centralized search engine that is supposed to have indexed the entire Web for users to search. Yahoo has succeeded with this strategy, and so has Google.

A problem to Microsoft by adopting this "successful experience" is that this strategy makes Microsoft forget its unique and powerful weapon that neither Yahoo nor Google has. The weapon is Windows. By giving up this weapon, Microsoft is just "a three-year-old kid comparing to the 12-year-old big boy Google" on Web search, once said by Mr. Ballmer. However, what might happen if the three-year-old kid decides to pick up a WMD (weapon of mass destruction) on hand? Google will be really afraid when Microsoft starts to assign Windows a new interpretation towards the new Web age.

In addition to the current strategy, Microsoft may think of an alternative strategy of Live Search to defeat Google. In my mind, this new search strategy should have a decentralized paradigm. Based on the network of registered Windows users, Microsoft can develop a novel social-search-style strategy by embedding Live Search into Windows (by contrast to add a Live Search link into Web browsers, I will explain the difference at the end of the post). This new strategy will put Live Search to the center of the new Microsoft Windows.

I will not discuss more details of this new search strategy in this post since it has already gotten be too long. I may start another post on this topic later depending on my time. But I will share the details of my thoughts with Microsoft Live Search scientists and engineers next week at Redmond.

More about the Yahoo Deal

As last, I want to say a little bit more about the Microsoft-Yahoo deal. In an earlier post, I briefly expressed that the deal may hurt Live Search in a long run. But certainly it is not because of Yahoo! Search that Microsoft wants to buy Yahoo.

Microsoft wants to jump into the market of online advertisement and buying Yahoo may be the fastest way to get into it and obtain a decent percentage of market share immediately. Unquestionably there are enough proper reasons for Microsoft to take this action. The problem is, however, that whether the action is really the best option, let it alone the only option, on the table, as some analysts have argued. I think it is not.

Merging with Yahoo could be a huge burden to Microsoft. This action would cost Microsoft both of the time and money to develop new-age Web-resource management strategy that is critical to the next-generation Web search. Anyway, Yahoo is constructed on top of its original successful Web search portal and Google's success on online advertisement is also on top of its successful Web search platform too. Designing a new-age Live Search is much more important for the future of Microsoft than merging Yahoo. Anyway, if Windows can keep on its strength on growing, buying Yahoo is actually much less critical to Yahoo, as the same analysts have implicitly suggested. Let's choose Live Search and new Windows instead of merging Yahoo!

(The binding of Live Search and Windows is not to enforce all the search flow to live.com by Windows. Otherwise Microsoft must get sued immediately by Google and all other search engines. The binding is actually about reinterpreting Microsoft to be a provider of WebROM and Windows is the product. By this reinterpretation, Web search is nothing but another basic function of new Windows such as the other ordinary Windows operations, e.g. creating a new file in PC. By this change, it does not matter who is the default search engine set in a computer. Even if Google is set to be the default search engine in a PC stalled by this new Windows, Microsoft may still gain a big (or probably the greatest) share on online advertisement because Windows is always the default search platform, in contrast to the particular search engine. Windows becomes the manager of the Web.)

Beginning of a Stage Transition, A View of Web Evolution, Series No. 8

(revised May. 25, 2008)

In this installment, we start to examine how we may observe a stage transition of Web evolution in the first place.

Corollary 5: the initiative of a stage transition in web evolution can be recognized by a qualitative upgrade of the Web-resource operating mechanism.

People have produced many valuable Web resources on the Web. The Web-resource operating mechanism (WebROM) is the methodology of declaring, displaying, and transmitting particular collections of Web resources on the Web.

In this corollary, a qualitative upgrade means that newly upgraded WebROM is going to effectively and efficiently support Web resources with new quality. Certain quality Web resources can only be effectively consumed by its respective WebROM (or the WebROMs at the higher evolutionary stages), and they can hardly be effectively consumed by the WebROMs at the lower evolutionary stages. I will adjust this claim later in this installment.

WebROM vs. WebOS

The specification of WebROM is different from the often-heard Web operating system (WebOS). The vision of WebOS is on the basis of that the Web as a platform. If the Web is just a platform, we may construct a uniform operating system that can effectively handles global namespaces, Web-wide resource discovery and management, remote process execution, authentication, and security. The execution of any new Web applications could be guided by such a WebOS. It is similar to that any new PC applications could be effectively managed by PC OS such as Microsoft Windows operating system. Certainly the Web development could become much more efficient if we can make such a WebOS be available.

A problem is, however, that the Web is an open world, which fundamentally contradicts to a personal computer environment that is a closed world. In an open world, we generally have no restrictions on duplicated resources or the potential of mixing Web identifiers with varied references. An open world also means a theoretically infinite large space of search. Hence a global (i.e. Web-wide) resource allocation methodology is too expensive (if possible) to achieve. These problems are main obstacles to produce a WebOS.

The vision of WebROM is on a difference basis of that the Web as a society. In a society, we have loose conventions about Web resource allocation and operation. But the detailed implementations of the conventions are open to individuals. For example, in our society we have federal laws that are general conventions of all the members in our society. A specific execution of these laws, however, is often interpreted slightly differently at varied local places due to their particular local condition. In an open world, nobody can predict all circumstance and hence such a flexibility of interpretation is crucial for the stable of society. Based on this analysis, WebROM is only described at the level of methodology comparing to that WebOS is tending to implement at the level of method.

If the Web is a society, which type of a society it is? In general, there are three basic types we could choose to develop our WebROM methodology. It could be a dictating society, a democratic society, or an anarchistic society. Among the three options, a dictating web is too hard to maintain because basically it mean a WebOS. On the other side, an anarchistic web is too difficult to manage since it has too few rules to follow. Therefore, a democratic web is the one we should and could achieve.

The Beginning of Transition from Web 1.0 to Web 2.0

The intention of WebROM at Web 1.0 is to declare and reference various Web resources. In particular, the main implementation of this methodology happens to be the HTML encoding. HTML allows users uniformly declaring various resources onto the Web. Based on this technology, we have produced Web browsers to uniformly displaying the declared Web resources. In addition to HTML, we have also developed auxiliary WebROM methodologies such as PHP and Javascript at Web-1.0 level for the same goal of 1.0-WebROM. But these additional resolutions are not the fundamental.

In comparison, the intention of WebROM at Web 2.0 becomes to efficiently executing various specified Web resources. The main implementation of this methodology happens to be AJAX. Unlike other 1.0-level WebROM implementations such as PHP and JavaScript, the primary purpose of AJAX is not about initiating Web resources. By contrast, it is about initiating Web resources that can be executed independently. This fundamental change of philosophy causes the effective support of newer quality Web resources---the Web-2.0-quality resources. This is why the invention of AJAX is milestone comparing to other ones such as PHP and JavaScript.

To be clearer, we briefly compare the traditional PHP to AJAX and see their difference on the level of WebROM.

PHP is a well-designed computational language that supports dynamic Web pages. PHP helps dynamically access Web resources when there are too many of them on servers. Hence the purpose of PHP was to improve the quantitative management of Web resource. The invention of PHP significantly accelerated the quantitative accumulation of resource on the Web. Based on this recognition, the invention of PHP aggravated the the primary contradiction of Web evolution by allowing faster quantitative accumulation. This philosophical consequence is different from being a transitional trigger of Web evolution.

The invention of AJAX has a complete different philosophical meaning. In the last installment, we mentioned that the direct technological consequence of the primary Web-evolution contradiction at the level of Web 1.0 was typically represented by the page-refreshing problem. To solve this problem, we expect a mechanism that allows computers to execute and only execute the particularly required pieces of resources in a web space while keeping other resources in the same web space untouched. By following this philosophy, AJAX allows Web resources in any single Web space to be transmitted piece-by-piece on the basis of user request. Hence AJAX solves (by contrast to aggravate) the primary contrition of Web evolution at the 1.0 level. This is why the invention of AJAX became the trigger towards Web 2.0.

Another way to demonstrate the AJAX was the trigger of transition from Web 1.0 to Web 2.0 is that AJAX has encouraged the prevalence of new-quality Web resources. For instance, before AJAX Web-2.0-quality resources, such as Web widgets, were hard to be prevailed. Web widgets are portable chunks of code that can be installed and executed within any separate HTML-based web page by an end user without requiring additional compilation. In theory, the execution of Web widgets does not necessarily depend on AJAX. Supposing there is no AJAX, however, all Web widgets in the same Web space may have to be reloaded synchronously when every time one of them is asked to update. This execution overloading is the major restriction that Web widget could not be popular on Web 1.0 because their installation could hardly be scaled. AJAX solves the problem and makes Web widget be a standard type of 2.0-quality Web resources.

By these discussions, we can learn that the invention of AJAX indeed represents the initiative of the stage transition from Web 1.0 to Web 2.0.

The next: Essence of Web Evolution

• Previous posts in this series
  
• In the Beginning …
  
• Three Evolutionary Elements
  
• Two Postulates
  
• Web Evolution and Human Growth
  
• Evolutionary Stage
  
• Qualities of Evolutionary Stages
  
• Trigger of Transition