At Blue Cross Blue Shield where I work, we have 54 different systems. They are divided into projects. We provide queries against either Db2 or Microsoft extracted data. You can write an SQL code against a COBOL file. You can type in there, you can read and write any type of data sets, big or little, and you can merge things.
We're supporting millions of cases. People can go to our website and find every one of their queries. Every doctor visit, what happened at the visit, and that type of stuff. We also provide information for providers. We've also created things for groups and companies that are nationwide and have their own plan, with their logo and reports, so their HR people can access the information.
Previously, I set up a series of WebFOCUS HTML pages that would run a job and query when you clicked on them. I worked at a university, and we could get any historical data about how many people were in a major for the last 10 years by gender, race, sex, or whatever. We could generate close to 800 different reports just with one prompted page.
The solution has what is called a mainframe product. It runs on any operating system: Windows, Linux, whatever. Their history goes back to the 60s. While they were on the mainframe they basically invented a core language, or their own scripting language. That language has been used since the 60s and is in all of their products, so if you get to a command line prompt, you can do pretty much anything. Tools like Tableau, Crystal, SAP, and BusinessObjects don't have this scripting language, so they expect you to do everything through SQL. A lot of the tools out there are GUI or ETL-based, but you can do all of that within this core language.
The solution can read and write 500 different types of outputs. You can read and output an XML file. You can write output takes in Excel or CSV files and you can write out to databases. All of that is built in.
Another thing they do that's a little bit unique is that you can read a file and then put out a hold file, which is a temporary file in its own binary format. When you're doing very complex inner and outer joins on SQL, you don't know if the data is right or if your filtering is correct. In WebFOCUS, you can go off to table A, do something, put it out to a temporary or permanent file, then go to table B, do a bunch of stuff, put out a file, and then join the two hold files. You can join up to 32 different files simultaneously in one run, which is a unique feature that I haven't seen in any other tool.
They have a scheduling product called ReportCaster that is rock solid. It works, and it can do load balancing, so if you have tons of schedules on the first week of the month, it can kick off multiple clients to run simultaneously and off they go. If you're into a mainframe product, they're unique. You may not think they can run TCP/IP on a mainframe, but they do.
It is a versatile tool. Now, since they got bought out by TIBCO, there's more of a push on the GUI front end and analytics.
