But now that the news is out, has Chrome OS lost its shine? People had high expectations for Google's new operating system but the end result doesn't look like the revolutionary, "change the world" product many had hoped for.

Don't get us wrong - Google's OS is different than whatever Windows, Mac, or Linux build you have running on your computer today. The new OS does away with desktop applications entirely - everything you use on Google Chrome OS runs on the web. Of course, the company hopes you'll use a lot of Google products like Gmail and YouTube, but it doesn't limit you to just Google-branded services. In the built-in applications area, there are also links to other web apps like the online TV streaming service Hulu.com and music sites Lala and Pandora. To be fair, Chrome OS even links to Yahoo and Microsoft's webmail offerings right out of the box.

Google's major goal with Chrome OS is to moving computing off our personal hard drives and into the cloud...the Google cloud. To accomplish such a feat, they've made the web browser the OS. Everything you need (in theory) is accessible through the included Google Chrome browser, the same browser the company currently offers to Windows users with Mac and Linux versions expected by the end of this year.

As exciting as that vision is, we have to wonder if people - especially the mainstream netbook users the OS is aimed at - are ready for this big of a switch. And more importantly, is the technology itself ready to make the change a comfortable and seamless experience?

...but is it Better?

After digesting yesterday's news, some lingering questions remain. Was this the OS everyone was hoping for or has Google let us down?

You Can't Just Install Chrome OS - You Have to Buy a New Netbook

To begin with, one of the more surprising reveals that came out of yesterday's news is that the OS cannot be installed on your own computer. Oh sure, there are downloads available that use Google's open-sourced code to create bootable builds tech-savvy users and developers can play with, but the official word from the search giant is that anyone wanting to use the "real" Google Chrome OS will have to purchase a new netbook to do so. You cannot simply download it from the web and install it on any machine.

Part of the reason for this restriction is driver support. Google is working with carefully selected manufacturers to offer a handful of netbooks running the OS in the coming year. By going this route, they don't have to provide an entire ecosystem of drivers for every piece of hardware out there - they can pick and choose which ones to support. They'll likely limit the number of peripherals supported, too. According to what was said yesterday, the company will support "mass storage devices" (think USB flash drives and digital cameras) but were cagey on how they plan on offering printing support. All they would say is that they're planning on an "innovative approach" when it comes to printing, whatever that means. Hopefully, they're planning to do something more than just integrating with Kinko's and FedEx's online document services, for example. Printing, (sorry Google) is not a web app just yet.

No Other Web Browsers Supported

Another big disappointment is the company's decision to limit all web surfing to the one included browser, Google Chrome. Firefox and Safari users are out of luck - no other browsers will be supported. But before you cry out "antitrust!," be warned - Google has this covered. The code base used to build the OS is open-source - that means anyone take the code and create their own version of Chrome OS. As was carefully - and haltingly - explained by Google's VP of Product Management, Sundar Pichai, other browser makers can take the code and build their own OS if they want to. But let's get real - Firefox Chrome OS? We don't think so. The reality is that fans of other browsers are simply out of luck if they want to use this operating system.

Offline Access is Limited. Your New Netbook is Now a Brick.

One of the questions that got glossed over during the Q&A session at the end of the event is how Google's OS plans to deal with offline access. The world is not blanketed in Wi-Fi yet, so what can this web-based OS do without the web? Surprisingly, the answer given didn't refer to any subsidized deals with cellular providers regarding deals to offer built-in 3G connectivity for the new netbooks. Instead, Pichai explained that the OS was built for use with Wi-Fi.

Of course, a handful of Google products use Google Gears, a technology that makes websites available offline. For example, Gmail uses Gears to create an offline version of your webmail inbox which you can use to read and respond to email until internet connectivity becomes available again. At that point, all the changes are synced back to Google's servers. Although Google didn't specifically refer to Gears when answering the question, there's no reason to doubt that it will work in Chrome OS's web browser the same as it does now in the standard Chrome browser.

However, Pichai did make note of Chrome OS's support for HTML5, an upcoming revision to the core markup language used to build the web. In the new specification, a key feature is offline support for web apps. However, web application developers will have to rebuild their apps in order to use HTML5, so users will be dependent on each individual company to make this change. While it's believed that one day this spec could make the whole web an offline app, the reality is that most developers have yet to implement this technology in their services yet. Even by Chrome OS's launch next year, there's no reason to believe the landscape will have changed significantly by then.

Do You Really Need an OS or Just the Chrome Web Browser?

Finally, the big question regarding Chrome OS is why? What can the OS do that any operating system running the Chrome browser cannot? Based on what was shown yesterday, the answer is very little. Chrome OS's brand-new features consist of two things: application tabs and panels. The panels are persistent windows that pop-up in front of your web browser's main window. For example, Google Chat, the company's IM service, can live in a panel that stays on top no matter what window you're viewing.

Application tabs, meanwhile, are special tabs that give you easy access to your most frequently used web apps from the browser. Any page tab can be made into an application tab with one click and the resulting "tab" is represented with the colorful icon for that site or service. While that's certainly a cool feature, it alone isn't a major selling point for the OS. That would be like saying you have to buy Mac OS X because of the dock or Windows because of the taskbar. You need a million of these little features combined to add up to a compelling reason to buy an OS.

That's not to say that Chrome OS itself doesn't have worthwhile features of its own - like its built-in security mechanisms or its auto-update system, it's just that these aren't the kinds of things that sell it to an end user. The questions consumers want answers to are what does it do that's special? What does it look like? And for now, the answer is "it's basically just a web browser."

Revolution? Maybe Not Just Yet.

At the end of the day, Chrome OS is an exciting, but not fully realized, vision. Although it has potential, the world may not be ready for a web-based netbook right now. Also, the technology needed to make the Wi-Fi only netbook useful without an internet connection isn't up to full speed either. At the end of the day, the netbook will be marginally more useful than an iPod Touch - when connected, it's amazing. Offline, not so much.

While you might not rush right out to buy a Chrome OS netbook when they first launch, there could come a time - sooner than you think - when it becomes a reasonable choice. When the majority of apps work offline and you've fully transitioned away from desktop apps, a web-connected netbook, especially one that's affordable, could easily become your everyday computer. That day hasn't arrived yet. For now, Chrome OS is an exciting glimpse at the future of computing, but not a practical device for the majority of users.

Disclosure: Sarah Perez freelances for Microsoft's Channel 10 blog, but is not a Microsoft employee. Her primary web browser is, in fact, Google Chrome which she uses exclusively.

While the in-home kick-off concept is an interesting launch tactic, we have difficulty believing that the people in the campaign's imagery are actually huddled around the computer enthralled by the Windows 7 experience. Are you sure this isn't a YouTube party? Balloons, shrimp cocktail AND models? If it's going to be this kind of a party, let's just hope no one does any keg stands over your keyboard. To register as a party host visit the house party page.

The WebOS space is made up of a number of small startups, many of them in Europe, and it seems to be prone to flare ups and disagreements between developers. In May 2006 I blogged that eyeOS claimed Orca was vaporware - and a heated discussion took place. Part of the issue there was confusion about whether Orca is open source, which is the bone of contention with MyBooo too.

Although the eyeOS 0.9x code has since been superseded by the 1.0 version, eyeOS wants MyBooo to "recognize that they broke the GPL license". In the comments to wwwhatsnew.com's post, a representative from MyBooo (Jeremy, no last name given) admits they used some eyeOS code, but otherwise he vigorously defends his company:

"Today few developers of EyeOs cummunity who spread a rumor in the Blogs and have shamefully attacked MyBooo. They mentioned and I quote "stolen their code and broken the GPL(General Public Licence)". We DO NOT appreciate this kind of slander and insult to our work via Web.

We are totally transparent with our sources and we have never intended nor thought of stealing the codes of EyeOS.

We confirm on the other hand that we have used a small part of EyeOS for developing MyBooo. We have found a small part, again a small part only, of EyeOS code useful to use hence we had integrated it to our system rather than re-developing the same. We were also encouraged that our project is compatible to them in case of any future partnership (which has been largely breached because of false accusations of our open source vision)."

It's unclear to me whether or not MyBooo has broken the GPL license terms, but they have certainly used some eyeOS code. It seems the stand-off can easily be rectified by MyBooo giving eyeOS the credit they deserve, in the MyBooo licensing.

The WebOS market is a fiercely competitive one, so these tiffs will continue I suspect until a clear market leader emerges. Right now, eyeOS is a leading contender, along with others such as YouOS and DesktopOnDemand.

eyeOS hails from Spain and was one of the early entries into a market that, as yet, has little evidence of profitability. However eyeOS now boasts 120,518 users (compared to 53,500 in August 2006), so it is ramping up well.

Changes in eyeOS 1.0

Back in August, Eduardo Perez Orue of eyeOS told me that their product was used by basically three types of people - for reasons ranging from word processing between home and office, to file sharing, to game-playing. The 3 types of users are:

• People who download the code and use it in their own server.
• People who use the miniserver (a small windows-only program that installs apache, php and eyeOS in your PC, so you can run it locally).
• The users at the Public Server at http://eyeOS.info. These are people who (like me) simply want to sign up for an eyeOS account and start using the product.

Since August eyeOS has been completely rewritten and redesigned; however eyeOS is still an Open Source project, under the GPL license. Eduardo told Read/WriteWeb today that "the importance of eyeOS 1.0 isn't in its look or in its apps, but in its core." He says they "designed it as a normal operating system, running over a microkernel, which loads services (such as virtual filesystem, global security, graphics/screen shows...) and libraries (such as support for XML, ZIP, TAR, widgets for making apps...)." The statement that eyeOS is like "a normal operating system" is sure to get under the noses of the OS purists out there, but I find it admirable that eyeOS is aiming to be a real OS and hence development platform - and not just a virtual desktop for apps such as word processing (although that it is part of it).

Eduardo told R/WW that "we've created a full toolkit to develop apps on", calling it "a new way of creating internet applications" - i.e. rich web apps fully based on ajax, and with "great control over their security and management with the eyeOS core." Indeed everything, says Eduardo, is made through the eyeOS Toolkit. Developers only need to join app "widgets" together and create funcitons for them - "eyeOS will do the rest."

eyeOS in action

Here is the screen you're confronted with when you first enter eyeOS:

The top menu looks like this (below image shows state when the "Office" icon selected):

Unfortunately my test drive tonight didn't get much further, as I ran into a problem with the test environment I was using (note this was prior to go-live). I'm sure this has been resolved in the live version, which went up just minutes ago.

Conclusion

eyeOS is making great progress and it seems to offer a compelling development platform for apps builders. It may even make the WebOS doubting Thomases think again! Check out eyeOS 1.0 and let us know what you think in the comments below.

Desktop On Demand (DOD) is the latest WebOS to come onto the market. It is a free "online desktop service" that lets you access your desktop remotely. DOD comes with 1GB of free disk space, with extra space available for a fee. DOD is aiming to go beyond simply being a virtual desktop - it wants to be "a wide ranging and powerful remote computing platform". We'll explore what that means in a minute, but first some context...

The WebOS, in its literal sense as a web-based virtual desktop, has been a pet topic of the Read/WriteWeb team for some time. For context, check out my overview of the WebOS market in April 2006 along with a case study of EyeOS in August. Ebrahim Ezzy covered the territory for Read/WriteWeb in September last year, then Emre Sokullu wrote about the mythical GoogleOS in November. A WebOS is also known as a Webtop, and Ebrahim's definition probably explains it best:

"A webtop (derived from 'desktop') pushes that replication to its limit. Also known as a WebOS, it is basically a virtual desktop on the web. It is a simple, less bloated, less featured and remotely accessible operating environment that runs in a browser. It delivers a rich desktop-like experience, coupled with various built-in applications."

The point of a WebOS/webtop is to give you remote access to all your usual desktop applications. It's an ambitious undertaking and there is no shortage of (usually small) vendors pursuing the dream - such as YouOS, Goowy, DesktopTwo, Xin and eyeOS. Also last year Firefox co-founder Blake Ross announced his effort called Parakey, and Laszlo is also building one. So it's a popular, if still very niche, product.

So back now to Desktop on Demand. DOD comes with a customizable interface, applications such as online word processing and "full featured POP/IMAP email client", drag-and-drop file transfer capability, file sharing and anonymous Internet access with identity protection. Perhaps the most intriguing feature is the ability to "browse the web anonymously and download files direct to your DOD hard drive, without tying up your own Internet connection."

DOD is putting a lot of emphasis on privacy and is offering users "end to end encryption" when accessing their remote desktop and data - also when transferring data between a local computer and their remote desktop or file space. This is going to be key for DOD - and other webtops - because there is still a significant amount of user hesitation about putting their personal files in an online service.

DOD is attempting to differentiate itself from the other webtops by not just mimicking a desktop. The product is being built by a company called Defuturo, which says that the core of DOD is the Linux Gnome desktop environment "and the growing myriad of applications constantly being developed for it." So it is using an open source platform and building a virtual desktop service on top of that. DOD aims to extend rather than replace a local computer, but I think they need to work on their messaging on how exactly they'll do this. As a user of the service, I'd want to know just how DOD complements my personal computers. The private file sharing is a good start though.

All in all, DOD looks to be another interesting entrant into the WebOS market. I'm still testing it out, but let us know your thoughts.

Over the past few weeks, we have examined possible GoogleOS scenarios and covered top WebOS offerings. GravityZoo is a very early stage WebOS, with a whole different - and noteworthy - approach.

What the heck is a WebOS again?

Firstly, let's revisit what a WebOS is. A WebOS is a web-based app (generally powered by either AJAX or Flash technologies) that emulates the application capabilities of an operating system. It's basically a virtual desktop that gives you communication tools (like email and instant messaging), productivity tools like word processing, and ability to play games and any other application that you'd find on a typical OS like Microsoft Windows. 

But that's not all - a WebOS should be an expandable platform (just like Windows) that gives developers the right tools (IDE, APIs) to develop new applications and add new utilities. And that's the advantage of a Web-based "OS" - it gives you the new ability to use it remotely and not be restricted to one PC. This is a huge need in today's world, because increasingly we use more than one computer from various different locations. So each time we sit down to a computer, a different scene awaits us - different applications and different documents. WebOS can solve this inefficiency, by taking your productivity with you everywhere you go.

GravityZoo is more the second type. You need to install an app to start using it; just browsing to http://gravityzoo.com is not enough. Although this makes it more of a pain to get started, it offers several advantages in functionality - namely the potential to be much faster and more secure.

The GravityZoo Experience

Brace yourself, GravityZoo is a daunting program and the terminology alone took Richard and I some time to figure out! 

When you launch GravityZoo, by default it literally covers the whole screen. You virtually lose all your connection with the operating system and enter the GravityZoo space. However after launch, you have the option to reduce the GravityZoo UI to simply another window inside your Windows screen. Here's a screenshot illustrating the full screen launch:

One immediate problem though is that GravityZoo is extremely slow! This is because GravityZoo is a platform on top of a platform: Microsoft .NET 2.0. 

When you start the program, as shown above, you see two options Web and Net:

The Web option is a custom browser - there is a version powered by the Gecko Engine of Firefox, plus a version based on the IE engine (kind of like the Maxthon browser). Gecko is a good choice for future cross-compatibility. 

The Net option is GravityZoo's own desktop client, very confusingly called "NetBrowser". They seem to have a widget server that communicates with the client app. Is this a better approach? In one sense it is, because it is probably more efficient and secure than standard HTTP protocols. But on the other hand, it is not a standard web app platform and GravityZoo has created its own APIs, with the hope that it will become a de-facto standard. This is similar to the approach of Laszlo, which recently entered into the WebOS space. But Laszlo did so by open sourcing and releasing under a liberal license their platform, called OpenLaszlo. 

Net Browser - currently there are a very limited number of applications

Sample Application - Four In a Row

Web Browser - exactly the same as Internet Explorer or Firefox

Criticism

Apart from the terminology confusing the heck out of Read/WriteWeb's editor (!), as of now GravityZoo only supports Windows. This may be considered a big disadvantage, because if the purpose is to usurp the applications that run on an OS, then why does it only run on a single operating system - particularly a non-free one? However, as stated above, GravityZoo is still in the prototype stage - so Windows is an acceptable platform as the most popular operating system.

One concern is GravityZoo's .NET 2.0 framework foundation. This is absolutely not a dependency that the average user can satisfy without guidance. However this won't be a problem for Vista users, as .NET 2.0 is pre-installed on that.

It should be noted that we encountered frequent errors, no doubt because the product is still early in its development.

Conclusion

Although it's confusing now for the average user, even tech-savvy ones, GravityZoo is definitely a unique concept that is worth keeping an eye on. But my biggest concerns are:

• the hot competition in the WebOS space;
• it's not entirely clear if this is actually a WebOS, because the "Net" app does not use web standards;
• and as Paul Graham stated after the failure of Kiko, this is a project that Google internal staff are probably already working on!

WebOS services are going to be utilized by thousands of companies - and will power the next generation of web applications. Amazon is at this point leading the charge of the big Internet companies to capture this potentially huge market.

There is a very long, but interesting, cover story in today's BusinessWeek entitled Jeff Bezos' Risky Bet. The article focuses on the transformation of the e-commerce giant into a software company. The growing stack of Amazon Web Services clearly points to a sea change in the Seattle e-commerce giant. Indeed Amazon is beginning to look more like an alternative Microsoft for the web computing era! 

In short, Jeff Bezos' big bet is a bet on the software infrastructure of the Web. We here at Read/WriteWeb think this is a visionary strategy by Amazon - and it is likely to pay off... 

Amazon completes its Web Services stack

In August, I wrote a series of articles about Amazon's Web Services strategy for a Web 2.0 magazine. The article that summed up what Amazon is up to was called: Amazon - the Real Web Sevices Company. Based on the piece in Business Week, it is clear that during the Web 2.0 conference next week, Amazon's Web Services strategy will become official. As a software engineer, I can't hide my joy. This is indeed a triumph of software engineering - a large company has managed to productize the pieces of its own infrastructure.

Not only that, but Amazon is very serious about making money on this endeavor. The web giant is carefully and methodically rolling out the building blocks of its next generation Web Platform. It started with the Amazon eCommerce API and Alexa services. But not until the Simple Storage Service rolled out, did it became clear that Amazon is building a full web services stack. Here is our diagram showing what it looks like:

Amazon's Web Services stack is evidence of a new computing paradigm, where web services in aggregate give rise to a new web-based operating system. Like a classical operating system, this new one has the key ingredients - infinitely scalable storage, dynamic indexing service, adaptive grid, etc. These pieces, put together, provide a compelling new way to think about application development. Amazon is actively working to both define and implement the ingredients of this new Web Platform.

Why this makes sense

Building large-scale web software is a big challenge. Amazon solves this problem by offering the infrastructure that has powered one of the biggest online stores for the past decade. Amazon hides complexity behind simple, minimalist APIs and offers their services for a very reasonable cost. The Amazon team takes the concepts of search, storage, lookup and management of data - and turns them into pay-per-fetch and pay-per-space web services. 

To begin with, it'll be small and medium businesses that take up Amazon's services. As Business Week points out, Wall Street is not going to jump on this. But the SmugMug photo service did and other startups and small businesses will follow suit. So even if large corporations will not come, there is plenty of money to be made. The Long Tail anyone?

What can we expect?

In the near term, we will probably see more services from Amazon which focus on completing their Web Services stack. For example, S3 does not have querying capabilities - which is a fairly big limitation. The elastic cloud is very powerful, but at the same time complex - so we can expect additional offerings that simplify deployment and management of the grid. 

We are also likely to see other players entering the WebOS market. Google has already made moves with its Google Base API and is rumored to be working on the GDrive. Microsoft also has Live Drive in the works. Both Google and Microsoft are no doubt working on other web services initiatives. Also watch out for smaller but more innovative players, like 3Tera - which we profiled in September.

Regardless of the provider, WebOS services are going to be utilized by thousands of companies - and will power the next generation of web applications. Amazon is at this point leading the charge of the big Internet companies to capture this potentially huge market.

In upcoming posts, we will highlight the use cases for Amazon and other web services. In the mean time, let us know if you're currently using Amazon Web Services - and what you think of the experience so far.

See Also: Web Platform Primer - what's available via API?; GData API for Google Base released; Amazon Launches Elastic Compute Cloud

"As he describes it, from a user’s point of view, Parakey is “a Web operating system that can do everything an OS can do.” Translation: it makes it really easy to store your stuff and share it with the world. Most or all of Parakey will be open source, under a license similar to Firefox’s."

Readers of my ZDNet blog will be familiar with my obsession earlier this year with the WebOS. There are a lot of small, mostly European, startups building a WebOS system - a virtual desktop, if you will. Check out my WebOS market review from April, to see some of the startups in this space. None of them have made any headway into the mainstream market yet - in fact most are still building out beta versions of their products. Let's not forget also that Google (in particular) and Microsoft are very well positioned to come stampeding into this market, like a herd of.... er, elephants.

Here's another key quote from the IEEE Spectrum article:

"Today, something like e-mail can involve two completely different experiences, depending on whether or not you’re using the Web—Outlook versus Hotmail, for example. A Parakey e-mail program, on the other hand, provides a single access point for your mail, “unifying the desktop and the Web,” in Ross’s words. Parakey is intended to be a platform for tools that can manipulate just about anything on your hard drive—e-mail, photos, videos, recipes, calendars."

Also Om Malik has some details about the funding and Matt Mullenweg has a post too.

To use Parakey, you'll need to download a small app onto your computer - which essentially turns your computer into a local server, thus enabling offline access. Kind of similar to how Dave Winer's Radio Userland blogging tool works. This description could equally apply to Radio Userland:

"...everything is ultimately stored locally, your computer being synchronized with remote servers whenever you are online. “You never have to care about the uploading process,” says Ross. “That just happens transparently.”

I'm looking forward to testing Parakey out. It sounds like it has the potential to help bridge the current gap between Web and desktop applications, in an open source way. I'm also interested in the reaction of the other WebOS vendors.