On Demand meets the Enterprise Portal

Mike Minelli of Compuware Covisint. Rooted in the auotmotive industry Covisint provides hosted portals, messaging and trusted identity frameworks to clients in an on-demand model of business service delivery. At the Gartner Conference they talked about "Leveraging an OnDemand Model to Consolidate Shared Enterprise Portals"

I wonder if this will just be a 30 minute infomercial or will there be something of value?

The evolution of the portal - the phases of growth:

• Information Explosion
• Information Consolidation
• Information Convergence
• Communications Convergence

Gartner's evolution of the portal places the portal in the business integration phase but not yet contributing to Business Transformation. This would seem to indicate that the lessons of Web 2.0 and Enterprise Mashups provides the potential to enable the agile enterprise - built on the infrastructure strengths of the portal.

 

A well designed and implemented portal enables new information feeds and composite applications to be delivered quickly without having to specifically develop functionality for security, personalization and other basic features of a good portal.

Okay - so yes - mainly an infomercial. Still it gave me a chance to catch up on twitter. On that front I am getting to like Quotably a lot. 

Portals and Mashups - Putting the Powr of SOA in our hands

Ray Valdes presented a session on Portals and Mashups at the Gartner conference in Baltimore.

Portals bring the personalization in the presentation of information. Mashups enable information feeds to be filtered and adapted to meet emerging needs.

Heavyweight Portal Deployments bring complexity and this leads to an "Agility Gap." If we are to create agile organizations we need the ability to adapt to business needs quickly.

Portal technology is a massively layered solution requiring extensive skills to support and leverage. However, portals are still an essential tool to enable personalization, manage access and control the delivery of information.

Service Oriented Architecture (SOA) enables composite applications but these are typically heavyweight designs using complex, over-engineered protocols. 

 

Web 2.0 Mashups are lightweight and often use simpler technologies, such as REST. The emerging Enterprise Mashup market, led by tools such as JackBe's Presto offer an opportunity to combine the best of SOA and Web 2.0 and deliver the results via the portal infrastructure.

The challenge is to enable mashups without unleashing anarchy. IT organizations need to apply some governance around the use of SOA to ensure consistent use of feeds that implement codified business processes.

Interestingly in this large audience over 50% are using portals and SOA. More interestingly only about 10% of those feel their SOA initiative has been a success. The usual situation is a long deployment cycle, complex environment and little benefit to be demonstrated until well in to the deployment phase - typically 2 years or more after initiating the project.

Services are autonomous, connected, contracted (typically via WSDL's), loosely coupled, independent, discoverable (via a registry such as Systinet) and built on standards.

Portals bring application modules (portlets), user and system administration, personalization, search, access control and identity management all built on an Application Integration Interface. Think of it this way. Information is delivered through a Portal, Not by a Portal.

The next evolution of portals will be to expose the portal sub-systems as services themselves. In our quest for reusability this is a good thing.

The objective of a portal should be to deliver strategic services at tactical speeds. The basic infrastructure such as access management is delivered by the portal framework.

The next evolution in the use of portals on the Internet needs to address Distributed Identity and Shared Metadata.

Web 2.0 is driving the agenda in SOA with moves in to the Open cloud, lightweight, AJAX enabled, loosely coupled and even embracing community development. 

Enterprises have had Mashups in the Enterprise for many years. It is not a new requirement. It is just that the capability was delivered using Microsoft Excel and manual cut-and-paste. Mashups provide a dramatically enhanced capability with automatic updating and the ability to merge and filter information much more easily and then deliver it wherever the user needs it.

Examples of Web 2.0 mashups included Housing Maps and Tripit. Facebook, while a social network has delivered an application platform and now has the ability to deliver applications anywhere.

A couple of great quotes:

"Organization with warped lines of communication can build systems that are fragile and don't interoperate."

"dysfunctional organizations can get stuck in constant battles with design defects, bugs and performance issues."

As we have learned with the wisdom of crowds the social architecture of the system can be as important as the technical architecture.

What did I take away from Ray's session? Portals are a delivery service. Enterprise Mashups that leverage Web 2.0 approaches and enable self-service development are a critical component in the solutions armoury if we want to build and enable an agile enterprise.

Portals going beyond Web 2.0 - Really?

David Gootzit talking about the Portal's future and beyond Web 2.0.

What is a portal? Web software that provides relevant access to web assets in a highly personalized fashion to target users and groups.

The portal is dead, long live the portal. It is just the tool to provide personalized views to information. Enterprise Portals are serving as an entry point to mashups such as JackBe's Presto platform.

The "MyPortal" will enter the Enterprise and will be a portal service aggregator.

Portals are being decomposed to extend the life of legacy portal technology.

Users and workers already engage with multiple portals. We adapt to the portal. The portals are meta-stovepipes. They only aggregate data within their own environment.

A portal fabric or portal mesh will emerge through inter-operability efforts. What Gartner is really talking about but not verbalizing is an evolution of OpenSocial. I want to take my profile and background information and contacts between social networks and portals.

Directory, security and Identity management will be critical services provided by and for Portals.

I am thinking that the evolution of widget-capable Wikis represent a potential MyPortal platform. 

Approximately 30% of the audience have MyYahoo and iGoogle pages.

The five touchpoints for portals:

• Portlets
• User profiles
• Directory
• Security
• Metadata

Federated Identity Management is a critical capability and OpenID is an increasingly serious player in this area. I believe this because one lesson from the Internet is "Simple is best."

60% of enterprise portals have been implemented as replacements for Intranets.

Corporate blogs have not been implemented often on internal enterprise sites. More often used in Business to Consumer roles. On the other hand Wikis supplement portals and are not competing against portals. 

 Wiki and blogs benefit the enterprise more than email because the content can be shared and made searchable.

Enterprise Mashups will be integrated by 50% of enterprises by 2011. They will typically be integrated into portals. Enterprise Mashups are open standard, user defined platforms. Typically consumer and enterprise data is merged. For example, a customer list overlaid on a Google map. 

Adobe Flex and Air, Mozilla Prism and Microsoft Silverlight will confuse the landscape by driving the use of rich clients. Up until now the browser-based thin web client has been the predominant delivery method.

At the same time expect portals to de-compose and effectively deliver portal functionality as services.

The portal of the future will:

- Be aggregation friendly

- support end-user mashups

- integrate social networking

- part of the overall enterprise user experience

- extend the life of legacy web applications as portal services extend to non-portal applications.

The last item is probably where the best Return on Investment will be found.

Finally, as you evolve a portal strategy you need to understand the needs of the digital native.

Portals or Social Network - Who wins?

The Portal v. Social Network debate continues to move along. Stowe Boyd and Steve Rubel are fueling that debate. I respect the knowledge and views of both Stowe and Steve. On this subject I have to side more with Steve. However, there is a caveat. The portals will only win out if they embrace social networks and not try to become one. This is why OpenSocial is such a significant development. The portals will win out if they allow us to bring the essence of our preferred social network or networks and encapsulate it inside my home page. The portals that allow us, the indigenous content creators, to create our own window on to the world we have created will win. The portals that try to replicate social networks and strong arm us in to using their flavor will fail. The successful portal will provide the open framework and underlying communications tools that allow us to be lazy. In this instance I believe laziness is the key. My attention is be scattered across a multitude of sites. The platform provider that allows me to consolidate those sites - on my own terms - will be successful. Don't you want to be able to fire up your browser with it showing your favorite search engine, a snapshot of your email and IM's and a window in to your social networks all there on one page? Or am I the only lazy one around?