I’ve only had a little opportunity to work with Crystal Reports, but, as you can tell, I was very underwhelmed. It’s not even in the same league with BO, especially for those of us who ocassionaly like to write freehand SQL. It’s just aweful for that.
Our company is trying to make us (the finance dept) migrate from BusObj to Crystal, even though a)the licenses are more expensive, and b)we have hundreds and hundreds of BusObj reports that we’ve developed over the past five years. Crystal has a lot of nice features, esp. in the area of formulas, but it’s much less user-friendly, doesn’t allow the flexibility of variables and user-defined objects, and because it lacks the “semantic layer” of the universe, it’s more difficult to maintain data consistency between users. I will resist until they pry my mouse from my cold, dead hand . . .
This may be a bit off topic but I sat in a meeting this morning with our programmers and listened to them pick on BO for an hour because one of our users wants to write her own BO reports against operational data. The DBA cautioned about data contention issues if this report was refreshed while hundreds of students were trying to register for classes thru the web apps. That seemed to be a bit of a stretch as a read-only query is going to zip thru those rows pretty darn quick. More likely, it seems her query would be waiting for db locks, not the other way.
Then the prgrammers started talking about doing these reports in java and the assistant director, a pretty smart guy, said that if they gave him the reports she wanted, he could have his java programmers knock it out - ahem - within 3 months.
I think a company has to carefully weigh the advantages/disadvantages of using a packaged solution such as BusObjects versus a in-house developed solution.
At an upcoming project my client is concerned about the cost of licensing BusObjects to over 2500 users (plus the infrastructure). Compare this to the cost of writing your own reporting interface via java etc. (with better use of resources compared to a deployment of BusObjects on a Windows platform). I am right now thinking on how many of their reporting needs we maybe could satisfy with WEbi reports instead (which will allow the servers to handle more concurrent users).
Regarding BusObjects v6/Code name Tosca:
I heard it will have a BO light process for the Windows platforms.
Can anyone chime in on it? Maybe, give some estimate numbers how many concurrent BO light serv processes a CPU under Windows will be able to handle?
The biggest problem I’ve run into is that we have so much work that has been done in java and that’s the only way to retrieve those classes…that it’s quite possible my department will be doing away with Business Objects completely. Even to the point of ignoring the opportunity for using it for the things is does so well!