Letting users loose on system integration?
Is that really such a good idea?
Reg Reader Studies There has been a lot of talk in recent times about portal frameworks and other technologies that allow end users to "compose" their own applications and essentially perform their own development and system integration, in the front end at least.
Advocates of Web 2.0 often take this idea to the extreme, and sometimes appear to argue that developers and integration specialists will somehow become redundant as users solve their own application and information access problems.
Cries of "the death of applications" soon follow, motivated by a desire to disrupt the status quo in the software application vendor community by people who object to others making more from the intellectual property they have developed.
Wherever you stand on such ideas, ideals, and arguments, there are a couple of practicalities that need to be considered. The first, which many overlook, is that the majority of users in a business environment simply couldn't be bothered with all this. These are the guys who are not "information workers", to steal a Microsoft label, but people who just turn up in the morning (or at the beginning of their shift) and simply want to get on with their job with the minimum of complication then go home again at the end. With all due respect to the average store man, neither they nor the company are likely to benefit much if the chap starts mashing up the organisation's inventory data with supplier catalogues over the internet.
But even when some kind of development and integration "self service" is appropriate, this is only manageable (from a cost and overhead point view) and safe (from a security and integrity perspective) if the starting point for the process is a set of robust pre-built components or services that have been designed, developed, and tested by IT professionals, whether internal or external.
This whole discussion is particularly relevant to users trying to get information out of systems in order to do their jobs, whether that be transactional level data to investigate a customer service query or performance statistics to manage some aspect of the business overall.
Recent research into information related risk suggests that despite lots of investment in various categories of information management solutions, fragmented and inconsistent data is still a major problem in many organisations.
The user self-service approach may obviously have a place here, but what are IT departments doing to enable this and/or other implement mechanisms to deal with the problem of information access, particularly across multiple sources and systems?
Good question, and in fact one we would like your input on. We are trying to gather as many views and ideas as we can then we'll summarise insights for The Register readership in general – perhaps a more community form of self help.
So, if you have experience in information management and delivery, or are a non-specialist who can help us understand the nature and extent of some of the challenges out there, we would really appreciate your help with the latest Reg reader survey.
Sponsored: Benefits from the lessons learned in HPC