 |
|
|
Corporate
Portals An NSS Group White Paper Table of Contents INTRODUCTION
At the beginning of the application software era a quarter of a century ago, neither users nor administrators needed to worry about where data was stored, which applications were being run, and which desktop operating systems and clients were installed. Centralised computing was the order of the day. Applications were designed and created to be run on a single, central host – the mainframe – and these applications were designed to support specific business functions, such as accounting or payroll. Desktop access was provided by dumb terminal, each with a direct connection to the central host which would actually run the applications for the user. All processing was done, and all data stored, on the mainframe, and the whole infrastructure was easily managed and controlled by the IT department. The introduction of the PC into the work place triggered a steady erosion of this centralised control. Users demanded more freedom and flexibility, and the provision of stand alone word processing, spreadsheet and database applications provided this. Users were finally in control of their own computing environment without having to rely totally on the IT department. The growth of networking allowed workgroup and departmental systems to evolve slowly from these early “personal” applications. The marketing department created marketing-related databases and apps, whilst the sales department customised things for their own use. These applications developed with little regard to the needs of other users in the company, and so little “knowledge islands” were created. Over time, these islands grew bigger and bigger, creating a huge number of highly-compartmentalised, segregated information stores. Of course, as the importance of the corporate network became apparent, there was an increased need to make this data an enterprise-wide resource. In order to achieve this, however, users required multiple applications on their desktop, each operating in a completely different way and each with its own unique user interface. Sometimes they even needed multiple access devices, as the new data was accessed via a standard PC whilst the legacy data remained available via a dumb, green screen terminal only. One of the biggest problems was that those “islands” of information were all stored in different formats and in different places. It was extremely difficult and time consuming for a single user to access data from multiple departments and pull it all together into a single, coherent report. One analogy might be the situation where you were forced to use a different type of telephone depending on who you wanted to call. Less experienced users were bewildered by the plethora of user interfaces and data stores and found themselves completely unable to bring up related data sets onto a single screen. Key users were being starved of that most vital corporate asset – informationThe advent of the Internet and the widespread deployment of Internet standard technology is now changing the information environment in which this fragmentation has been the unavoidable ruling condition. With every desktop including a standard Web browser, and with Internet access almost mandatory these days, the basic infrastructure is in place to provide the unified desktop. More and more applications are being Web-enabled out of the box, and the new breed of corporate portal software provides the means to “Webify” those remote islands of data. The intranet applied Internet technology to our internal networks to allow Web publishing of all our data – now the portal extends that to allow different types of data in multiple locations to be unified in a single view. By converting existing data stores into HTML content we can bring them all together in a single user interface via the Web browser. Occasional and experienced users alike need learn only one desktop client interface, and access to all data – no matter where it resides and in what format – is achieved in the same way. With Web-to-host publishing tools, even the legacy data buried on those hard-to-kill mainframe systems can be accessed and presented in a modern, graphical format alongside all the more recent information repositories. Anyone who has browsed the Internet will be familiar with the portal sites there – sites like Yahoo!, Lycos, Infoseek and Excite – that provide a means of pulling together a wide range of information in a single place. The advantage they offer the casual browser is not only their ability to provide a single point of entry to the Internet, but also the ability to personalise that entry point, so that at each visit the user is presented with the latest news and information that is directly relevant to him or her. As it stands today, the corporate portal is straddling the fence between the data processing era of computing and the knew “knowledge era” heralded by the emergence of the Internet. On one side of the fence is the old application-centric view of the business, where every user or workgroup is defined by the applications and data they use, and it is not possible to tell from looking at a user’s desktop what his or her function is within the company. On the other side is the new task-centric view, where the user’s desktop is defined by, and reflects, the job they are doing. In its current incarnation as information kiosk, the Portal is clearly an example of a large publishing database. However, as interactive applications - from collaboration platforms to vertically-focussed professional desktops - begin to emerge, the Portal will become the platform for tomorrow’s function-centred computing. The aim of the corporate portal is very similar to that of its Internet-based predecessor – to provide added value for its customers. It has to make life simpler than the current model with is array of incompatible applications and data stores in order to improve user efficiency. Internet portals started off as little more than search engines, attempting to provide a means to filter and search the vast repository that is the Internet. Navigation sites moved on from simple full text indexing to categorisation of documents, allowing users to narrow down their search to a particular group of documents that may be of interest. Sites were thus filtered into pre-configured groups according to their content – sport, finance, news, and so on. The final stage was the full-blown portal site, which provides search functionality and categorisation as before, but adds personalisation of content by the user, real time chat, access to communities of interest, and links to specialised functions such as on-line trading of stocks and shares, shopping networks, and so on. Through these new sites, users have a single point of access from which they can make connections to all their Web-based information feeds, whether they are going shopping, reading the news, checking their stock portfolios, or just browsing. The needs of the corporate user are not that different. They require the same single point of access to external sites, but they want this to be combined with equal ease of access to all their internal information feeds too. The birth of the Internet raised expectations of being able to search for all types of information outside corporate information networks. Intranets promised the same type of access to internal data too, but have moved quickly from being the answer for information sharing, to creating an array of unorganised data across the enterprise. More often than not, the most valuable and useful company information is locked away in a myriad of applications that are either off-limits for most users or don't provide any means of getting an overall picture. Hard copy reports still circulate. As business moves faster, companies need to be more dynamic and flexible. Getting the right information to the right people in the organisation is critical to making timely decisions. The biggest challenge faced by the corporate portal today is that very fragmentation of data, poorly organised and stored in many different formats across multiple platforms. Traditionally employees spend too much time searching for information in order to do their jobs. The successful portal will provide the means to re-integrate and publish this data in a function-centric manner, thus helping users to do their individual tasks that much better. Corporate portals unlock hard-to-reach data and present it to users in a personalised Web page, offering a more efficient way to communicate corporate information. IDC's report, Sourcebook for Knowledge Superconductivity, identifies portals as the ideal medium for knowledge management. It also recognises the evolution of the corporate portal market place and has identified four main categories of corporate information portals: Enterprise Information Portals (EIP) - which provide personalised information to users on a subscription and query basis Enterprise Collaborative Portals (ECP) - which provide virtual places for people to work together Enterprise Expertise Portals (EEP) - which provide connections between people based on their abilities Enterprise Knowledge Portals (EKP) - which provide all of the above and proactively deliver links to content and people that are directly relevant to user’s tasks in real time. It takes little imagination to determine just how a corporate portal could benefit the average company. Even a reasonably well-organised network directory structure can be a nightmare to navigate when you are trying to find a particular document or spreadsheet. New sub-directories appear every day as users struggle to categorise their output in the only way they know how, and this presents serious navigation problems to those users outside a specific workgroup when they need access to that information. The portal would allow corporate data in all its forms to be fully indexed and categorised in a number of ways. This would allow clients to execute comprehensive search and find operations across a range of document formats and workgroup boundaries, as well as navigate through a logical tree structure of relevant categories to zero in on the data they require. Nor would the user need to worry about the native format of the data they were accessing. The portal would simply present them with the data requested by converting to HTML on the fly, allowing data from any location and any application to be presented in a standard Web browser without any input from either administrator or user. Even legacy data, previously only available via application-specific green screen terminals, can be pulled from the mainframe files and formatted in HTML for inclusion in standard reports and enquiry screens. The administrator and/or department heads can classify the data and determine who should have write access, who should have read access, and who should have no access at all. The portal would then apply these fine-grained access controls at the time of the request depending on how the client has authenticated him or herself to the system. In short, the portal provides a mechanism to support and simplify the otherwise complex process of publishing material to the Web. Even those organisations with well-developed intranet services will find that the portal will save many hours of converting source data into a format suitable for publishing on a Web site. And when that source data changes, the Webmaster usually has to start all over again with the conversion and publishing process. This is not necessary with the portal, which can provide an interface both to relatively static data (such as Word documents) and dynamic data (such as stock levels in modern or legacy database files) for immediate conversion and publication on the fly. Whilst some of the benefits mentioned will accrue to only the larger organisations – those with legacy mainframe applications and data stores, for example – many will apply to small to medium enterprises too. Even the smallest company can benefit from the ability to access all their corporate literature, documents, spreadsheets, and product databases – not to mention the integration of their favourite applications and Internet information feeds – from a single unified desktop: the Web browser. Arising seemingly out of nowhere in 1998 the corporate portal market is extending at a fast and furious pace. Given its relatively short period of existence, there is currently no single dominant player in this market. The extent of portal adoption today has lead researchers to the conclusion that corporations show a strong interest in experimenting with portal sites. They have also noted as a general observation, the adoption pattern for portals closely resembles the beginning stages of the intranet implementation in 1996. A recent survey conducted by the Delphi Group suggestions that 9% of companies surveyed have recently implemented a pilot version of a corporate portal, while 45% of the interviewed companies have no portal plans. Additionally the research shows that by January 2001 83% of companies are expected to implemented a corporate portal solution. The corporate portal market is still in its infancy, and the adoption cycle is still in the early stages. Unlike the abortive client-server initiatives of a few years ago, however, the corporate portal is in for an easier time. This is because so many companies have already adopted the intranet model, and are frantically hunting for tools that will help them streamline it. This is exactly where the corporate portal fits in. It is not so much a revolution, as an evolution of the intranet, bringing all the associated benefits to the experienced and casual user alike, through a single common interface. This doesn’t mean that implementation is easy, since most intranets suffer from the common problem of fragmented islands of information. It is all available to everybody, just as it should be, but no one currently has a clue how to categorise it and publish it in the most effective manner. The corporate portal will not automate the categorisation process, though some products will offer tools that can help. In most cases, however, someone still has to wade through the mire that is the corporate information repository and create something wonderful and logical out of it. Not everything should be transferred to the Web, and knowledge managers need to make the distinction between data that will create a value proposition for portal users and data that will not. Once that has been done, however, the portal takes over and brings it all to the Web and keeps it up to date, no matter how often it is changed. Once information has been categorised in this manner, users will have a clearer idea in the future where, and how, to store their everyday data so that it can be made available quickly and easily via the portal. It is vital that organisations do not skimp on this first process, since rushing to implement the portal tools will only serve to bring a poorly organised and essentially useless glut of information to the portal user. Once it has been completed, only then should the next step be taken, which is to select the portal tools themselves. Once again, their selection will depend on what type of data and environment they are expected to support. Is it all Notes databases you need to publish, or SQL Server, or Access, or a mixture? And does it include legacy data? Does the portal provide tools to extract data from different formats, or does it expect to access everything via standard SQL queries? If the latter, you will need to source third party tools to convert your legacy data to a SQL feed. How easy is it to publish your standard Microsoft files such as Word documents, Excel spreadsheets or PowerPoint slides? Can you easily add links to applications from your portal (i.e. to allow users to fire up Word directly form the portal screen) and is it easy to insert links to external Web resources? Perhaps most important of all, how easy is it for your users to personalise the portal environment for themselves? Finally, what about security? Can the portal provide a secure single sign-on environment and fine grained access control? Although you need to be able to offer seamless access to a broad range of data across your organisation (without forcing users to authenticate themselves each time they need to access a new data store), you also need to enforce corporate security policy to ensure that users still cannot access data they are not meant to. The third stage is implementation, and this also needs to be considered carefully. Try to do too much in the first round and you will make life difficult. The first step for most organisations is to bring the information to the portal in its native presentation format (though converted to HTML, of course), often in separate windows. This will provide a single access point for information throughout the company, but will still result in numerous windows, each with its own information feed. This stage is often very easy to accomplish with the portal’s built-in tools. The next step is to customise this by linking those separate windows and information feeds together, creating truly unified data views using information sources from anywhere in the organisation. As business becomes increasingly based on electronic systems, both knowledge management and the corporate portal will depend heavily on the organisation’s information infrastructure. It is at the level of the corporate portal that issues of information design and presentation strategies become a primary concern. It is vitally important that organisations do not simply plough ahead and implement a replica of today’s “messy desktop” with its multiple disconnected windows, but in a Web environment instead. To deliver on the dream of the truly task-centric desktop, companies need to consider carefully what information is truly vital to the day-to-day running of the business, as well as how best to present that in different ways to different users. Once that mapping has been established, the corporate portal is the device that will bring it to the end user. |
Send mail to webmaster
with questions or
|