Student information systems continue to evolve, with many moving to a web-based interface for managing student, school, and district data. While this has obvious benefits, many vendors are choosing to maintain client-server or, more and more frequently, Windows Terminal Services/Citrix Presentation Server implementations.
Many early systems relied on proprietary client software to access centralized databases. These databases were often building-based, requiring significant manual data processing to update district- and state-level data stores. Chancery Software (recently purchased by Pearson), among others, attempted to unify data for entire districts with additional software modules, but eventually abandoned this approach for a centralized, web/SQL Server model. All schools could access data via a Windows web server and database/application server farm. Great in theory, but many districts are finding that this sort of system requires an incredible amount of maintenance on the back end. Quite literally hundreds, if not thousands of tables are required to manage these sorts of data in a proper relational model. This is hardly the kind of system that most school secretaries can get their heads around, let alone full-time IT staff.
The user experience inherent in a browser-based student information system can rarely compare to the richness of a client program that can support multiple windows, speedy data entry, and easy (at least relatively speaking) report creation. In fact, while modern web-enabled systems have a lot to offer administrative staff, the end users at the schools are the first to feel the limitations of a browser interface. If one more secretary complains about having to use a back button I'm going to lose it. I get it. Windows: good; back, click, hyperlink, back, forward, submit, OK, cancel: bad.
Of course, the teachers, who primarily enter attendance and grades actually like a browser interface, not because it's in a web browser, but because the system can generally be accessed remotely. Who wouldn't rather enter grades in their boxers? Often, client-server applications are designed to be accessed from within a local area network, putting up a few barriers to remote access. Similarly, while many browser-based systems tend to be platform-independent (at least in theory), client-server systems tend to be platform specific. Older operating systems (or non-Wintel OS's) can present additional challenges.
Enter application servers. Citrix, Windows Terminal Services, and even Mac OS X Server provide mechanisms for accessing applications through a variety of web browsers or cross-platform clients (please, no talkbacks about my omission of LTSP here - I'm afraid there just aren't any student information systems running on Linux). Ubiquitous bandwidth and dropping prices on high-powered 64-bit server hardware mean that districts no longer need to choose between remote access/platform independence and the robust nature of many client-server systems. Just run the client-server systems remotely!
I'm not suggesting that the web-based systems lack value. However, if we allow some of the big players to go head to head, what would end users (read secretaries, guidance counselors, and administrators) rather deal with? Complicated SQL-based tables, intensive querying and reporting tools like Crystal Reports, and single-window browsing capabilities limited by sessions and cookies or a fast, multi-window, tightly integrated system with built-in report-writing tools? Most modern client-server apps tend to fall into the latter category. Web-based systems, no matter how well-built, continue to suffer from the limitations of web technology. As Web 2.0 applications emerge, some of the progress in refining web-based user interfaces will trickle down to student management systems. For now, however, the industry push to view these applications in a browser window, just for the sake of being web-based, seems misguided.
Not all browser-based systems are bad, nor are all client-server systems good. However, as always, a careful examination of user requirements and thoughtful testing with the primary users of the system needs to drive your choice in an SIS. I have trouble imagining a secretary who would prefer one of the current crop of web-based systems over modern client-server student management applications.