Currently, what we perceive as a "web browser" is in fact an easy way to develop and run applications "everywhere" which is much beyond the initial goal of having a visual tool for hyper-pages. In fact, a web browser is a very practical solution of achieving application portability. With the coming of mobiles devices, we are definitely giving more and more power to the web browsers.

However, web browsers are a kind of middleware application and cannot compete with the OS in terms of efficiency, security, and flexibility. Therefore, idealistically, the future will give us anything that allows the easiest development (program language: freedom of choice), with different levels of abstraction, and running in whatever existing platform, with full-support for remote data/services.

In my opinion, the future will bring a OS with different service level agreements (SLAs) and what we know as a "web browser" would be one of these integrated and native OS services. Some devices initially would only support a subset of the existing OS services and they would be limited to the services similar to the ones we have on web browsers.

So, in short, I believe the answer for the question is NO, but the current characteristics of web browsers will be definitely part of "integrated OSs". Going a little further in my expectations, if one application is for SLA-A, it would properly run on any existing OS that supports that SLA-A. Think about: this is what actually occurs when we have a Firefox running, for instance, on Linux and Windows. But we still have incompatibilities and performance issues (I mean, due to the OS, not due to the hardware) because we never formalized a standard SLA for this application.

Tom, look back at the testimony regarding the Consent Decree between the US Government and Microsoft.

One of the fears that Microsoft had was that Netscape's browser would make the Operating System irrelevant. It is why they implemented IE to use a library that was also integrated into the Windows kernel. Any changes to the browser also required changes for the OS.

Yes..It is already replaced. CloudOS is running in browser and capable of operating all the applications. One good example is 'EyeOS' which operates more than 67 applications such as Adobe , Word , Excel , PowerPoint , Paint, and Media-players Etc just on browser itself !!!

It depends on what you means by browser app. If you mean that it uses data or computation from the cloud, then there are situation where this is not possible, e.g., no electricity or no internet connection, or limited connectivity with too much latency. So software for disaster management might not be suitable for browser only applications)

Currently, what we perceive as a "web browser" is in fact an easy way to develop and run applications "everywhere" which is much beyond the initial goal of having a visual tool for hyper-pages. In fact, a web browser is a very practical solution of achieving application portability. With the coming of mobiles devices, we are definitely giving more and more power to the web browsers.

However, web browsers are a kind of middleware application and cannot compete with the OS in terms of efficiency, security, and flexibility. Therefore, idealistically, the future will give us anything that allows the easiest development (program language: freedom of choice), with different levels of abstraction, and running in whatever existing platform, with full-support for remote data/services.

In my opinion, the future will bring a OS with different service level agreements (SLAs) and what we know as a "web browser" would be one of these integrated and native OS services. Some devices initially would only support a subset of the existing OS services and they would be limited to the services similar to the ones we have on web browsers.

So, in short, I believe the answer for the question is NO, but the current characteristics of web browsers will be definitely part of "integrated OSs". Going a little further in my expectations, if one application is for SLA-A, it would properly run on any existing OS that supports that SLA-A. Think about: this is what actually occurs when we have a Firefox running, for instance, on Linux and Windows. But we still have incompatibilities and performance issues (I mean, due to the OS, not due to the hardware) because we never formalized a standard SLA for this application.

One should understand the basic functions of an OS and a web browser. There are two major parts of an OS: kernel, and add-on application programs. The basic functions of the kernel of an OS is allocating and managing the resources that include processor(s), memory(s) (primary or secondary), I/O device(s), etc. A good kernel OS tries to manage the resources as efficiently as possible. The add on programs include user interfaces etc. The browser (be it a web browser or a simple browser to "browse" the directories (or folders) or files within the directories (or folders) is an intermediary between the user and the kernel. The main task of the browser is to make the use of OS "easier" for the user. For an "end user" the browser might also be an important component to "browse" easily (and a web browser to "browse" the web). In other words, a browser or a web browser is an integral part of the OS.

I recall my days of using "lynx" on unix machines for browsing the Web. I remember around fifteen years back or so, many of our students were developing web browsers for windows as part of their mini projects. Now-a-days, we hardly advise them to do so. However, many of them develop some add on applications for Android phones etc.

For Guido Governatori I come from a lengthy background of emergency management and that is what I am studying as a Doctoral student, your thoughts that it may not be suitable for disasters are interesting. Disaster management is one place where cloud OS can be very useful. However we do need the communications and infrastructure backbone to be functioning. Generally one of the first things affected by a natural or anthropogenic disaster is failure advanced communications systems. Cell systems go down and the remaining are overloaded, electricity drops, utility lines are broken etc... However in a large scale disaster, responders from a wide array of services and brought in to to assist (public and private), the responders will come with a wide array of communications systems. For the higher technology we will find computers of all different OS's, Android devices, Windows devices, and many others. Here is why a suitable cloud system is beneficial, we need to be able to interconnect our various systems to communicate efficiently. This communications can extend further with ham radios, they can send packets, albeit at a low speed but they can help with connectivity. In my opinion I think we can. develop a reliable emergency communications system which can be placed in service fairly rapidly within a disaster zone, which then connects us to internet systems. Developing portable power generators for laptops is fairly easy, thus we can address the basic power concern. Portable cell sites are common to use in a disaster, thus we do have a method to address some of the communications concern along with ham radio. I think the bigger concern is methods to get this equipment into a disaster zone rapidly. Here in the U.S. we can do this; however in other more remote parts of the world we may have to deal with other political and logistical issues which slow down a response. What we need to work on is better trust, at least for life saving disaster management between all of our nations.

The Desktop applications will be replaced very soon by Web based applications. We can see the tremendous improvement in Android OS. No one thought it will get this kind of appreciation and promotion. Now everyone working on developing applications that works on Android. In my childhood days i usually play games in my computer by installing in a traditional way, but now all games are available in the Internet itself so need to worry about the RAM capacity or the Disk space availability or whatever else. I heard a news about the Mozilla works on a new kind of browsers and Operating systems using basic HTML, Java Script and Web API's. In a short span of time; the desktop applications will be replaced with the web based applications. No wonder if our OS itself becomes available from the Internet space. :-)

Although online applications have many advantages, they cant replace the desktop applications. The reasons are as follows:

1. you require all time Internet connectivity to access the applications. Consider mobility and bandwidth usage cost factor in mobile networks.

2. online applications may also store your data online (in the cloud). Majority of users have privacy concerns.

3. you have to pay for the online services (applications), where a considerable number of users are using opensource or pirated applications.

4. browsers do not support heavy applications. Consider downloading each stage of game (which can be over 300MB in size) with a slow network connection.

Apart, Google has recently launched cromebook and people are not willing to adopt it, unless an additional OS is installed.

Actually I use a chromebook and it is overall fairly handy. All services I need are available through cloud and websites. However I also partitioned the drive and put GNU/Linux on it also. From my view point cloud services are still growing and maturing and time will only tell how many desktop applications will move to cloud. We can look at computing history and draw some interesting parallels. The early (electircal) computers were giant machines and we communicated with them on punch cards an paper tape, later we included terminals with CRT's and key boards; essentially all computing was done in the server and only communications provided through the terminal. This was even expanded where we could run a serial terminal across a country with a terminal in California and the server in New York (maybe call this a early form of cloud computing). Then as PC's worked into the market we reduced load on our servers by placing applications on the desktop. Now we seem to be moving towards placing applications back on the server so we can access them from anywhere. I think the need for mobility is driving the cloud market; however this is just my observation. Where it will go in the future is going to be an exciting development to watch.

Just like Tom, I am also using the Chromebook. For me it's more or less a additional computer, light and small enough to carry it with me to presentations, meeting etc. And just like Tom I added an Ubuntu installation since I wasn't able to do all things I wanted to do with ChromeOS. After using it for several months my opinion is:

- purely browser-based application will not be a real alternative to desktop computers in near future

- but cloud-based applications will rise as soon as there are three requirements satisfied: a higher level of security for the data, better network coverage and a real OpenSource alternative.

When the US Government brought anti-trust charges against Microsoft (circa 2001), the goal by Microsoft was to include specific APIs into the kernel so that they could claim that they were required by the OS - even though there was no architectural reason to do so. They saw the browser (Netscape Navigator back then) as a direct threat to their PC OS monopoly.

Now, exactly which applications can/are/could be replaced by making them run inside a browser? The obvious ones are word processing, spreadsheets, presentations - the basis for Microsoft's cash cow, Office. Microsoft saw this as a threat, and had that exposed during the anti-trust trial.

It is my opinion that video/audio applications, and others that need to make use of the ancillary hardware on a PC (think GPU) won't work very well from inside a javascript written (or java vm) that is spawned by a browser.