No matter how pretty, nobody really wants to live in a walled garden. And yet, on the Web that’s what we all do when we visit sites like Yahoo, Facebook, AOL and all the rest. What people are yearning for is the ability to move seamlessly between all these sites because each of them offers some unique service. As a result, the ultimate Web application of the future is going to be a mashup, also known as a composite application, that allows people to seamlessly interoperate with data from all of these sources.
Unfortunately, the owners of all these sites have always thought in terms of owning the customer versus making it easier for more people to use the services on their sites. As a result, there is no standard today for importing and exporting data across these Web sites.
Microsoft, in its desperate efforts to battle Google, has now made a $44.6 billion bid to buy Yahoo. Integrating Yahoo with its existing MSN services on the Web would be a Herculean task so the best immediate course of action would be for Microsoft to create a standard way to integrate services across Yahoo and MSN. More importantly, it would give Microsoft a unique set of capabilities across its Web sites that would ultimately attract more users because the sites themselves would fundamentally be more useful.
There’s a lot of talk today about the need for data portability and OpenID 2.0 across all these Web sites, but the problem with all these initiatives is they are all mired in the industry politics of what it means to actually be open.
If the backers of these efforts really want them to succeed, they need to first create some sort of compelling application that shows users the real benefits of an open Web architecture. Once people really comprehend that their personal productivity is being limited by all these walls surrounding all our Web gardens, they will on their own accord starting taking sledge hammers to those walls as part of a mass rising of users that no short-sighted business plan is going to be able to stop.
In Microsoft’s case, this is an imperative because it will show skeptical users a clear differentiation compared to competing Web services. In the case of the rest of the Web, it’s only a matter of time before it happens so better to embrace it now than be seen resisting it later.
For more on this topic, check out http://googlewatch.eweek.com/content/data_portability/feeling_social_yahoo_avails_openid_to_its_masses_1.html
For more on the rising importance of cloud computing, check out http://www.baselinemag.com/c/a/Projects-Management/App-Development-Head-in-the-Cloud/
|
|
The next big database war is going to seem very familiar to a lot of people who are familiar with Microsoft Access and all the derivatives of dBase.
For years now the average end user has been grappling with the challenges associated with building a database for either individual use or for a small workgroup. Truth be told, the vast majority of what should be database applications are instead deployed as spreadsheet applications because that’s the tool that most people know how to use. This isn’t because other alternatives are not available. Quite the contrary, a number of desktop database products have been on the market for years. The problem is that for the average end user, these products are difficult to master and neither end users or the IT department really wants to support a bunch of arcane database applications running on workstations or servers that typically sit under somebody’s desk.
Unfortunately, the only alternative to creating a spreadsheet is to put in a request to IT to develop an application, which inevitably falls to the bottom of the application development priority list because more often than not it is not broadly applicable to the entire business. The end result of this is that IT sees its application backlog get longer while more users become increasingly frustrated waiting for applications that even though they will boost productivity will never actually arrive.
Against this backdrop a number new software-as-a-service offerings have been launched to give users the equivalent of access to an easy to use desktop database system that is only a click away. This idea itself isn’t all that new as QuickBase, for one, has been offering a similar service for about seven years. But thanks to advent of Salesforce.com and the increasing interest of Google and Microsoft, a raft of venture capital dollars is pouring into this space to fund startup companies such as Trackvia, Coghead, WuFoo, ZoHo, Iceberg OnDemand and LongJump are all jumping into this space. You should also expect to see Microsoft in this space with some version of Access made available as a service through Windows Live. Logically Google will either build out its own offering or acquire any one of the startup companies already in this space.
Within IT there might be some resistance to these applications because there is a real security issue in terms of having so much corporate data residing outside the firewall. And the folks managing compliance and risk might also look askance at such practices. But there isn’t all that much that IT can do to stop people from trying to become more productive so rather than resisting these services, the better course of action is probably to figure out which ones offer the most security and compliance within the established guideline of the corporation. For example, given the history of QuickBase, the odds are good that this service will pass security and compliance muster.
Whatever ultimately shakes out in this space, end users are going to see these services as a significant advance in the battle to make their work days easier. And anything that ultimately serves to make IT more valuable to end users can not be a bad thing at the end of the day.
|
|
One of the tenets of Microsoft’s future plans for the continuing relevance of Windows is that the advent of multicore computing platforms that are going to make it easier to further integrate new features into the core Windows platform.
On the face of it this is basically the same as it ever was strategy for Windows that Microsoft ever had. That strategy essentially calls for any piece of software functionality that more than 40 percent of users are likely to want should be included in the core operating system. And as much competitors like to take issue with that strategy, the history of computing prior to the invention of the PC tends to support the argument that commonly used piece of software functionality should be included in the core operating system.
So with the advent of multicore computing, it could conceivably become easier for Microsoft to integrate more functionality into Windows by leveraging different processor cores to run different pieces of code in parallel. But as much as multicore platforms creates an opportunity for Microsoft to continue its dominance, it also creates one of the most series threats to Microsoft’s dominance to come along in a very long time.
That threat stems from the fact that Windows owes its dominance to all the developers that built Windows applications using tools largely brought to market by Microsoft. But none of those existing tools really takes advantage of the parallel processing capabilities of a multicore processor. And we have already seen a number of startup companies such as PeakStream coming to market with new development environments that promise to make parallel processing easy to master for the average developer. In fact, what PeakStream is up to was so interesting, Google moved to acquire them.
So it’s distinctly possible that a new development environment running on a new operating system platform could usurp Windows in the next few years, which is a threat that Microsoft takes seriously enough to fund the development of a new F# development environment designed for parallel processing environments that are inherent capabilities of multicore processors.
There’s a tendency in this industry to think that thinks will always remain the way they are despite the fact that history tells us that it usually takes us a half dozen years to realize we reached some inflection point that changed everything that came after it. The rise of multicore processors is just such an inflection point that will probably take another three to five years to really appreciate how they changed everything we think about computing today.
|
|
We all know that XML and Web services are driving a revolution in enterprise computing but it seems that the actual timeline of this fundamental shift in computing models is on an evolutionary rather than revolutionary timeline.
One more sign that Web services has gone mainstream will come today when Red Hat announces a strategic alliance with SOA Software to bring the latter company’s Web services management tools to the Red Hat Linux platform. This is both a sign that Web services has reached enough critical mass on a mainstream Linux platform to require a management system and an acknowledgement that Linux is now a mainstream platform for developing applications built on a service oriented architecture powered by XML-based Web services.
But the question that this and other recent development s raise is what took so long? The whole concept of Web services first came to light at the beginning of this decade. We’re now well into 2007 and the whole concept of service oriented architecture driven by Web services is just now starting to go mainstream. And if you look hard at the types of SOA applications that have been deployed thus far, a lot of them feel like they are in a perpetual state of pilot testing.
A lot of the issues surrounding the evolutionary timeline of Web services and SOA have to do with the need to train a whole new generation of developers on the concept. But perhaps even more significant was a two year slowdown in the economy that forced a general retreat from anything new and risky and a late realization that most of the existing network infrastructure wasn’t up to the task of supporting latency sensitive SOA applications.
Whatever the cause, the movement towards SOA applications continues at a pace that augers well in terms of some general sense of forward movement, but still leaves a general sense of frustration when it comes to really unleashing the full potential of the technology before the end of this decade.
|
|
